From 1c8922586a0919c407692d524792f3ce29e99704 Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 10:29:38 +0000 Subject: [PATCH 01/10] feat: Add B2B CDP entity design documentation Co-authored-by: 13585811473 <13585811473@163.com> --- docs/b2b-cdp-entity-design.md | 1353 +++++++++++++++++++++++++++++++++ 1 file changed, 1353 insertions(+) create mode 100644 docs/b2b-cdp-entity-design.md diff --git a/docs/b2b-cdp-entity-design.md b/docs/b2b-cdp-entity-design.md new file mode 100644 index 000000000..e00f4399b --- /dev/null +++ b/docs/b2b-cdp-entity-design.md @@ -0,0 +1,1353 @@ +# B2B CDP 实体设计详细方案 + +## 📋 目录 +- [整体架构](#整体架构) +- [核心实体设计](#核心实体设计) +- [实体关系图](#实体关系图) +- [业务流程](#业务流程) +- [数据模型详细设计](#数据模型详细设计) + +--- + +## 整体架构 + +### 实体分层架构图 + +```mermaid +graph TB + subgraph "数据源层" + DS1[官网] + DS2[微信生态] + DS3[抖音] + DS4[线下活动] + DS5[电话/邮件] + DS6[CRM系统] + DS7[第三方平台] + end + + subgraph "渠道层" + Channel[Channel 渠道] + end + + subgraph "交互层" + Touchpoint[Touchpoint 触点] + Event[Event 事件] + Campaign[Campaign 营销活动] + end + + subgraph "客户主体层" + Contact[Contact 联系人] + Lead[Lead 线索] + Account[Account 企业账户] + end + + subgraph "业务层" + Opportunity[Opportunity 商机] + Product[Product 产品] + Order[Order 订单] + end + + subgraph "洞察层" + Segment[Segment 客户分群] + Tag[Tag 标签] + Score[Score 评分] + Journey[Journey 客户旅程] + Attribution[Attribution 归因] + end + + DS1 & DS2 & DS3 & DS4 & DS5 & DS6 & DS7 --> Channel + Channel --> Touchpoint + Channel --> Event + Campaign --> Touchpoint + Touchpoint --> Contact + Touchpoint --> Lead + Event --> Contact + Contact --> Account + Lead --> Contact + Lead --> Opportunity + Account --> Opportunity + Opportunity --> Product + Opportunity --> Order + Contact --> Segment + Account --> Segment + Lead --> Tag + Contact --> Tag + Account --> Tag + Contact --> Score + Account --> Score + Lead --> Score + Contact --> Journey + Campaign --> Attribution + Opportunity --> Attribution +``` + +--- + +## 核心实体设计 + +### 1. Account（企业账户）- 核心实体 + +```mermaid +classDiagram + class Account { + +String account_id PK + +String account_name + +String unified_social_credit_code + +String account_type + +String industry_id FK + +String account_status + +String account_level + +Decimal annual_revenue + +Integer employee_count + +String company_website + +String company_address + +String province + +String city + +String district + +String account_source + +String primary_channel_id FK + +String owner_user_id + +DateTime created_at + +DateTime updated_at + +JSON custom_fields + +String[] tags + +Integer health_score + +String lifecycle_stage + } + + class AccountChannelIdentity { + +String identity_id PK + +String account_id FK + +String channel_id FK + +String channel_account_id + +String identity_type + +Boolean is_verified + +DateTime first_seen_at + +DateTime last_seen_at + +JSON additional_info + } + + class AccountRelation { + +String relation_id PK + +String parent_account_id FK + +String child_account_id FK + +String relation_type + +DateTime created_at + } + + Account "1" --> "*" AccountChannelIdentity + Account "1" --> "*" AccountRelation : parent + Account "1" --> "*" AccountRelation : child +``` + +**字段说明：** +- `account_id`: 全局唯一账户ID +- `unified_social_credit_code`: 统一社会信用代码（企业唯一标识） +- `account_type`: 客户类型（潜在客户/现有客户/合作伙伴/竞争对手） +- `account_status`: 账户状态（活跃/休眠/流失/黑名单） +- `account_level`: 客户等级（战略级/重要级/普通级） +- `lifecycle_stage`: 生命周期阶段（认知期/考虑期/决策期/留存期/扩展期） + +--- + +### 2. Contact（联系人）- 核心实体 + +```mermaid +classDiagram + class Contact { + +String contact_id PK + +String contact_name + +String mobile_phone + +String email + +String wechat_id + +String job_title + +String department + +String contact_status + +String primary_account_id FK + +String contact_source + +String primary_channel_id FK + +String owner_user_id + +DateTime created_at + +DateTime updated_at + +JSON custom_fields + +String[] tags + +Integer engagement_score + +String lifecycle_stage + +Boolean is_decision_maker + +Boolean is_verified + } + + class ContactChannelIdentity { + +String identity_id PK + +String contact_id FK + +String channel_id FK + +String channel_user_id + +String identity_type + +Boolean is_primary + +Boolean is_verified + +DateTime first_seen_at + +DateTime last_seen_at + +JSON additional_info + } + + class AccountContactRelation { + +String relation_id PK + +String account_id FK + +String contact_id FK + +String role_in_account + +String decision_level + +Boolean is_primary_contact + +String relationship_status + +DateTime relation_start_date + +DateTime relation_end_date + +DateTime created_at + +DateTime updated_at + } + + Contact "1" --> "*" ContactChannelIdentity + Contact "1" --> "*" AccountContactRelation +``` + +**字段说明：** +- `decision_level`: 决策层级（决策者/影响者/使用者/把关者） +- `engagement_score`: 参与度评分（基于互动频率和深度） +- `role_in_account`: 在企业中的角色（CEO/CTO/采购经理等） + +--- + +### 3. Lead（线索）- 核心实体 + +```mermaid +classDiagram + class Lead { + +String lead_id PK + +String lead_name + +String company_name + +String mobile_phone + +String email + +String wechat_id + +String job_title + +String lead_source + +String channel_id FK + +String campaign_id FK + +String lead_status + +Integer lead_score + +String lead_grade + +String industry_id FK + +String province + +String city + +String owner_user_id + +DateTime created_at + +DateTime updated_at + +DateTime last_contacted_at + +DateTime converted_at + +String converted_contact_id FK + +String converted_account_id FK + +String converted_opportunity_id FK + +JSON custom_fields + +String[] tags + +Boolean is_qualified + } + + class LeadChannelIdentity { + +String identity_id PK + +String lead_id FK + +String channel_id FK + +String channel_user_id + +DateTime captured_at + +JSON utm_params + +JSON additional_info + } + + Lead "1" --> "*" LeadChannelIdentity +``` + +**字段说明：** +- `lead_status`: 线索状态（新建/联系中/已限定/已转化/无效） +- `lead_score`: 线索评分（基于行为和画像的综合评分） +- `lead_grade`: 线索等级（A/B/C/D） +- `is_qualified`: 是否为合格线索（MQL/SQL） + +--- + +### 4. Opportunity（商机）- 核心实体 + +```mermaid +classDiagram + class Opportunity { + +String opportunity_id PK + +String opportunity_name + +String account_id FK + +String primary_contact_id FK + +String lead_id FK + +String opportunity_type + +String opportunity_source + +Decimal amount + +String currency + +String stage + +Integer probability + +Date expected_close_date + +Date actual_close_date + +String close_reason + +String owner_user_id + +String[] product_ids FK + +String campaign_id FK + +DateTime created_at + +DateTime updated_at + +JSON custom_fields + +String[] tags + +Integer days_in_stage + +Boolean is_won + +Boolean is_lost + } + + class OpportunityStageHistory { + +String history_id PK + +String opportunity_id FK + +String from_stage + +String to_stage + +DateTime changed_at + +String changed_by_user_id + +String change_reason + +Integer duration_days + } + + class OpportunityProduct { + +String opp_product_id PK + +String opportunity_id FK + +String product_id FK + +Integer quantity + +Decimal unit_price + +Decimal total_price + +Decimal discount + +String product_description + } + + Opportunity "1" --> "*" OpportunityStageHistory + Opportunity "1" --> "*" OpportunityProduct +``` + +**字段说明：** +- `stage`: 阶段（线索/需求确认/方案设计/商务谈判/合同签订/已赢单/已输单） +- `probability`: 赢单概率（0-100） +- `opportunity_type`: 商机类型（新客户/追加销售/续约/交叉销售） + +--- + +### 5. Channel（渠道）- 核心实体 + +```mermaid +classDiagram + class Channel { + +String channel_id PK + +String channel_name + +String channel_type + +String channel_category + +String parent_channel_id FK + +String channel_status + +JSON channel_config + +Decimal cost + +DateTime created_at + +DateTime updated_at + +JSON custom_fields + } + + class ChannelPerformance { + +String performance_id PK + +String channel_id FK + +Date stat_date + +Integer lead_count + +Integer contact_count + +Integer opportunity_count + +Decimal revenue + +Decimal cost + +Decimal roi + +Integer conversion_count + +Decimal conversion_rate + } + + Channel "1" --> "*" ChannelPerformance +``` + +**渠道类型枚举：** +- 线上渠道：官网、SEO、SEM、社交媒体、内容营销 +- 社交渠道：微信、企业微信、抖音、快手、小红书 +- 线下渠道：展会、研讨会、地推活动 +- 合作渠道：合作伙伴、代理商、分销商 +- 直销渠道：电话、邮件、销售拜访 + +--- + +### 6. Campaign（营销活动）- 核心实体 + +```mermaid +classDiagram + class Campaign { + +String campaign_id PK + +String campaign_name + +String campaign_type + +String campaign_status + +String[] channel_ids FK + +Date start_date + +Date end_date + +Decimal budget + +Decimal actual_cost + +String target_audience + +String owner_user_id + +DateTime created_at + +DateTime updated_at + +JSON custom_fields + +String[] tags + } + + class CampaignPerformance { + +String performance_id PK + +String campaign_id FK + +Date stat_date + +Integer impressions + +Integer clicks + +Integer leads_generated + +Integer opportunities_generated + +Decimal revenue + +Decimal roi + +Decimal cpl + +Decimal cpa + } + + class CampaignMember { + +String member_id PK + +String campaign_id FK + +String member_type + +String member_ref_id + +String member_status + +DateTime joined_at + +DateTime responded_at + +String response_status + +JSON response_data + } + + Campaign "1" --> "*" CampaignPerformance + Campaign "1" --> "*" CampaignMember +``` + +**活动类型：** +- 网络研讨会（Webinar） +- 线下会议/展会 +- 邮件营销 +- 内容营销（白皮书、案例分享） +- 产品试用活动 +- 行业峰会 + +--- + +### 7. Touchpoint（触点/互动记录）- 核心实体 + +```mermaid +classDiagram + class Touchpoint { + +String touchpoint_id PK + +String touchpoint_type + +String channel_id FK + +String campaign_id FK + +String contact_id FK + +String lead_id FK + +String account_id FK + +DateTime touchpoint_time + +String touchpoint_direction + +String touchpoint_status + +String content_type + +String content_id FK + +String subject + +Text description + +Integer duration_seconds + +String owner_user_id + +JSON metadata + +JSON utm_params + +DateTime created_at + } + + class TouchpointAttachment { + +String attachment_id PK + +String touchpoint_id FK + +String file_name + +String file_url + +String file_type + +Integer file_size + +DateTime uploaded_at + } + + Touchpoint "1" --> "*" TouchpointAttachment +``` + +**触点类型：** +- 网站浏览 +- 表单提交 +- 内容下载 +- 邮件互动（打开/点击） +- 电话沟通 +- 会议/拜访 +- 在线聊天 +- 社交媒体互动 + +--- + +### 8. Event（行为事件）- 核心实体 + +```mermaid +classDiagram + class Event { + +String event_id PK + +String event_name + +String event_type + +String channel_id FK + +String contact_id FK + +String lead_id FK + +String account_id FK + +DateTime event_time + +String session_id + +String device_type + +String browser + +String os + +String ip_address + +String page_url + +String referrer_url + +JSON event_properties + +JSON utm_params + +DateTime created_at + } +``` + +**事件类型：** +- 页面浏览（page_view） +- 按钮点击（button_click） +- 表单开始（form_start） +- 表单提交（form_submit） +- 文件下载（file_download） +- 视频播放（video_play） +- 产品试用（product_trial） +- 搜索（search） + +--- + +### 9. Product（产品/解决方案）- 核心实体 + +```mermaid +classDiagram + class Product { + +String product_id PK + +String product_name + +String product_code + +String product_category + +String product_type + +String product_status + +Text description + +Decimal list_price + +String currency + +String pricing_model + +String[] feature_list + +DateTime created_at + +DateTime updated_at + +JSON custom_fields + } + + class ProductCategory { + +String category_id PK + +String category_name + +String parent_category_id FK + +Integer sort_order + } + + Product "*" --> "1" ProductCategory +``` + +--- + +### 10. Tag（标签）- 核心实体 + +```mermaid +classDiagram + class Tag { + +String tag_id PK + +String tag_name + +String tag_category + +String tag_type + +String description + +String color + +DateTime created_at + +String created_by_user_id + } + + class TagRelation { + +String relation_id PK + +String tag_id FK + +String entity_type + +String entity_id + +DateTime tagged_at + +String tagged_by_user_id + +Boolean is_auto_tagged + +String tag_source + } + + Tag "1" --> "*" TagRelation +``` + +**标签类型：** +- 行为标签（高活跃度、近期浏览过产品A） +- 画像标签（互联网行业、大型企业、决策者） +- 业务标签（重点客户、流失风险、高价值） +- 兴趣标签（关注AI、关注云计算） + +--- + +### 11. Segment（客户分群）- 核心实体 + +```mermaid +classDiagram + class Segment { + +String segment_id PK + +String segment_name + +String segment_type + +Text description + +JSON segment_rules + +String target_entity_type + +Integer member_count + +Boolean is_dynamic + +DateTime last_calculated_at + +String created_by_user_id + +DateTime created_at + +DateTime updated_at + } + + class SegmentMember { + +String member_id PK + +String segment_id FK + +String entity_type + +String entity_id + +DateTime joined_at + +DateTime left_at + +Boolean is_active + } + + Segment "1" --> "*" SegmentMember +``` + +--- + +### 12. Score（评分模型）- 核心实体 + +```mermaid +classDiagram + class ScoreModel { + +String model_id PK + +String model_name + +String model_type + +String target_entity_type + +JSON scoring_rules + +Integer max_score + +String status + +DateTime created_at + +DateTime updated_at + } + + class ScoreRecord { + +String record_id PK + +String model_id FK + +String entity_type + +String entity_id + +Integer score + +String grade + +JSON score_details + +DateTime calculated_at + } + + class ScoreHistory { + +String history_id PK + +String entity_type + +String entity_id + +String model_id FK + +Integer score + +DateTime recorded_at + } + + ScoreModel "1" --> "*" ScoreRecord +``` + +**评分类型：** +- Lead评分：基于行为和画像的线索评分 +- Account评分：企业健康度评分 +- Contact评分：联系人参与度评分 + +--- + +### 13. Industry（行业）- 核心实体 + +```mermaid +classDiagram + class Industry { + +String industry_id PK + +String industry_name + +String industry_code + +String parent_industry_id FK + +Integer level + +Integer sort_order + +DateTime created_at + } +``` + +--- + +### 14. Attribution（归因）- 核心实体 + +```mermaid +classDiagram + class Attribution { + +String attribution_id PK + +String entity_type + +String entity_id + +String attribution_model + +JSON touchpoint_sequence + +JSON attribution_weights + +DateTime created_at + +DateTime updated_at + } + + class TouchpointAttribution { + +String ta_id PK + +String attribution_id FK + +String touchpoint_id FK + +String campaign_id FK + +String channel_id FK + +Decimal attribution_weight + +Integer position_in_journey + +DateTime touchpoint_time + } + + Attribution "1" --> "*" TouchpointAttribution +``` + +**归因模型：** +- 首次触点归因 +- 末次触点归因 +- 线性归因 +- 时间衰减归因 +- U型归因 +- W型归因 + +--- + +### 15. CustomerJourney（客户旅程）- 核心实体 + +```mermaid +classDiagram + class CustomerJourney { + +String journey_id PK + +String journey_name + +String entity_type + +String entity_id + +String journey_stage + +DateTime journey_start_at + +DateTime journey_end_at + +Integer total_touchpoints + +Integer journey_duration_days + +JSON stage_history + +DateTime created_at + +DateTime updated_at + } + + class JourneyStage { + +String stage_id PK + +String stage_name + +Integer stage_order + +String stage_category + +JSON milestone_criteria + } +``` + +--- + +## 实体关系图 + +### 核心实体关系总览 + +```mermaid +erDiagram + Account ||--o{ AccountContactRelation : has + Contact ||--o{ AccountContactRelation : belongs + Account ||--o{ Opportunity : has + Contact ||--o{ Opportunity : has + Lead ||--o| Contact : converts_to + Lead ||--o| Account : converts_to + Lead ||--o| Opportunity : converts_to + + Channel ||--o{ Touchpoint : generates + Campaign ||--o{ Touchpoint : generates + Contact ||--o{ Touchpoint : receives + Lead ||--o{ Touchpoint : receives + Account ||--o{ Touchpoint : receives + + Contact ||--o{ Event : generates + Lead ||--o{ Event : generates + Channel ||--o{ Event : tracks + + Opportunity ||--o{ OpportunityProduct : contains + Product ||--o{ OpportunityProduct : included_in + + Campaign ||--o{ Lead : generates + Campaign ||--o{ CampaignMember : has + + Account ||--o{ AccountChannelIdentity : has + Contact ||--o{ ContactChannelIdentity : has + Lead ||--o{ LeadChannelIdentity : has + Channel ||--o{ AccountChannelIdentity : identifies + Channel ||--o{ ContactChannelIdentity : identifies + Channel ||--o{ LeadChannelIdentity : identifies + + Tag ||--o{ TagRelation : applies_to + Segment ||--o{ SegmentMember : contains + + ScoreModel ||--o{ ScoreRecord : calculates + + Industry ||--o{ Account : categorizes + Industry ||--o{ Lead : categorizes + + Attribution ||--o{ TouchpointAttribution : analyzes + Touchpoint ||--o{ TouchpointAttribution : contributes_to + Campaign ||--o{ TouchpointAttribution : contributes_to + + CustomerJourney ||--o{ Touchpoint : tracks +``` + +--- + +### 全渠道身份关联图 + +```mermaid +graph TB + subgraph "渠道身份体系" + WX[微信渠道身份] + WEB[官网渠道身份] + DY[抖音渠道身份] + EMAIL[邮件渠道身份] + PHONE[电话渠道身份] + OFFLINE[线下活动身份] + end + + subgraph "统一Contact" + Contact[Contact 联系人
统一ID: C001] + end + + subgraph "统一Account" + Account[Account 企业
统一ID: A001] + end + + WX --> |Identity Mapping| Contact + WEB --> |Identity Mapping| Contact + DY --> |Identity Mapping| Contact + EMAIL --> |Identity Mapping| Contact + PHONE --> |Identity Mapping| Contact + OFFLINE --> |Identity Mapping| Contact + + Contact --> |Belongs To| Account + + style Contact fill:#ff9999 + style Account fill:#99ccff +``` + +--- + +## 业务流程 + +### 线索到商机转化流程 + +```mermaid +stateDiagram-v2 + [*] --> NewLead: 捕获线索 + NewLead --> Contacted: 首次联系 + Contacted --> Qualified: 资格验证 + Qualified --> Converted: 转化 + + Converted --> Contact: 创建联系人 + Converted --> Account: 创建/关联企业 + Converted --> Opportunity: 创建商机 + + Contact --> [*] + Account --> [*] + Opportunity --> [*] + + NewLead --> Invalid: 标记无效 + Contacted --> Invalid: 标记无效 + Invalid --> [*] +``` + +--- + +### 商机阶段流转流程 + +```mermaid +stateDiagram-v2 + [*] --> Lead: 线索阶段 + Lead --> Qualification: 需求确认 + Qualification --> SolutionDesign: 方案设计 + SolutionDesign --> Negotiation: 商务谈判 + Negotiation --> Contract: 合同签订 + Contract --> Won: 赢单 + Contract --> Lost: 输单 + + Lead --> Lost: 丢失 + Qualification --> Lost: 丢失 + SolutionDesign --> Lost: 丢失 + Negotiation --> Lost: 丢失 + + Won --> [*] + Lost --> [*] +``` + +--- + +### 客户生命周期流程 + +```mermaid +stateDiagram-v2 + [*] --> Awareness: 认知阶段 + Awareness --> Consideration: 考虑阶段 + Consideration --> Decision: 决策阶段 + Decision --> Retention: 留存阶段 + Retention --> Expansion: 扩展阶段 + + Retention --> Churn: 流失 + Expansion --> Churn: 流失 + + Churn --> Winback: 召回 + Winback --> Retention: 成功召回 + Winback --> [*]: 永久流失 + + Expansion --> [*]: 持续合作 +``` + +--- + +### 全渠道数据流转流程 + +```mermaid +sequenceDiagram + participant User as 用户 + participant Channel as 渠道 + participant Event as 事件系统 + participant Identity as 身份识别 + participant CDP as CDP核心 + participant Lead as Lead管理 + participant Contact as Contact管理 + participant Account as Account管理 + + User->>Channel: 1. 访问/互动 + Channel->>Event: 2. 记录事件 + Event->>Identity: 3. 身份识别 + + alt 新用户 + Identity->>Lead: 4a. 创建Lead + Lead->>CDP: 5a. 保存Lead数据 + else 已识别用户 + Identity->>Contact: 4b. 关联Contact + Contact->>Account: 5b. 关联Account + end + + CDP->>Event: 6. 触发规则引擎 + Event->>Channel: 7. 个性化响应 + Channel->>User: 8. 返回内容 +``` + +--- + +## 数据模型详细设计 + +### Account 详细字段设计 + +| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | +|--------|------|------|------|------|------| +| account_id | VARCHAR | 64 | ✓ | 账户唯一ID（PK） | ACC_20231105001 | +| account_name | VARCHAR | 200 | ✓ | 企业名称 | 阿里巴巴网络技术有限公司 | +| unified_social_credit_code | VARCHAR | 18 | | 统一社会信用代码 | 91330000MA27XYZ123 | +| account_type | VARCHAR | 50 | ✓ | 客户类型 | CUSTOMER（客户）/PARTNER（合作伙伴）/COMPETITOR（竞争对手）/PROSPECT（潜在客户） | +| industry_id | VARCHAR | 64 | | 行业ID（FK） | IND_001 | +| account_status | VARCHAR | 50 | ✓ | 账户状态 | ACTIVE（活跃）/DORMANT（休眠）/CHURNED（流失）/BLACKLIST（黑名单） | +| account_level | VARCHAR | 50 | | 客户等级 | STRATEGIC（战略级）/IMPORTANT（重要级）/NORMAL（普通级） | +| annual_revenue | DECIMAL | (18,2) | | 年营收（万元） | 50000.00 | +| employee_count | INT | | | 员工人数 | 5000 | +| company_website | VARCHAR | 500 | | 公司网站 | https://www.alibaba.com | +| company_address | VARCHAR | 500 | | 公司地址 | 浙江省杭州市余杭区文一西路969号 | +| province | VARCHAR | 50 | | 省份 | 浙江省 | +| city | VARCHAR | 50 | | 城市 | 杭州市 | +| district | VARCHAR | 50 | | 区县 | 余杭区 | +| account_source | VARCHAR | 100 | | 来源 | WEBSITE/EXHIBITION/PARTNER/COLD_CALL | +| primary_channel_id | VARCHAR | 64 | | 主渠道ID（FK） | CH_001 | +| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | +| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | +| custom_fields | JSON | | | 自定义字段 | {"crm_id": "CRM001"} | +| tags | JSON | | | 标签数组 | ["高价值客户","AI行业"] | +| health_score | INT | | | 健康度评分（0-100） | 85 | +| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | AWARENESS/CONSIDERATION/DECISION/RETENTION/EXPANSION | + +--- + +### Contact 详细字段设计 + +| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | +|--------|------|------|------|------|------| +| contact_id | VARCHAR | 64 | ✓ | 联系人唯一ID（PK） | CNT_20231105001 | +| contact_name | VARCHAR | 100 | ✓ | 联系人姓名 | 张伟 | +| mobile_phone | VARCHAR | 20 | | 手机号 | 13800138000 | +| email | VARCHAR | 200 | | 邮箱 | zhangwei@company.com | +| wechat_id | VARCHAR | 100 | | 微信ID | wx_zhangwei | +| job_title | VARCHAR | 100 | | 职位 | CTO | +| department | VARCHAR | 100 | | 部门 | 技术部 | +| contact_status | VARCHAR | 50 | ✓ | 联系人状态 | ACTIVE（活跃）/INACTIVE（不活跃）/BOUNCED（退订）/UNSUBSCRIBED（取消订阅） | +| primary_account_id | VARCHAR | 64 | | 主要关联企业ID（FK） | ACC_20231105001 | +| contact_source | VARCHAR | 100 | | 来源 | WEBSITE/FORM/IMPORT/API | +| primary_channel_id | VARCHAR | 64 | | 主渠道ID（FK） | CH_001 | +| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | +| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | +| custom_fields | JSON | | | 自定义字段 | {"birthday": "1985-01-01"} | +| tags | JSON | | | 标签数组 | ["决策者","技术背景"] | +| engagement_score | INT | | | 参与度评分（0-100） | 75 | +| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | SUBSCRIBER/LEAD/MQL/SQL/OPPORTUNITY/CUSTOMER | +| is_decision_maker | BOOLEAN | | | 是否决策者 | true | +| is_verified | BOOLEAN | | | 是否已验证 | true | + +--- + +### Lead 详细字段设计 + +| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | +|--------|------|------|------|------|------| +| lead_id | VARCHAR | 64 | ✓ | 线索唯一ID（PK） | LEAD_20231105001 | +| lead_name | VARCHAR | 100 | ✓ | 线索姓名 | 李明 | +| company_name | VARCHAR | 200 | | 公司名称 | XX科技有限公司 | +| mobile_phone | VARCHAR | 20 | | 手机号 | 13900139000 | +| email | VARCHAR | 200 | | 邮箱 | liming@company.com | +| wechat_id | VARCHAR | 100 | | 微信ID | wx_liming | +| job_title | VARCHAR | 100 | | 职位 | 产品经理 | +| lead_source | VARCHAR | 100 | ✓ | 线索来源 | WEBSITE/FORM/CAMPAIGN/COLD_CALL/REFERRAL | +| channel_id | VARCHAR | 64 | | 渠道ID（FK） | CH_001 | +| campaign_id | VARCHAR | 64 | | 营销活动ID（FK） | CMP_001 | +| lead_status | VARCHAR | 50 | ✓ | 线索状态 | NEW（新建）/CONTACTED（已联系）/QUALIFIED（已限定）/CONVERTED（已转化）/DISQUALIFIED（无效） | +| lead_score | INT | | | 线索评分（0-100） | 80 | +| lead_grade | VARCHAR | 10 | | 线索等级 | A/B/C/D | +| industry_id | VARCHAR | 64 | | 行业ID（FK） | IND_001 | +| province | VARCHAR | 50 | | 省份 | 广东省 | +| city | VARCHAR | 50 | | 城市 | 深圳市 | +| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | +| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | +| last_contacted_at | DATETIME | | | 最后联系时间 | 2023-11-05 14:00:00 | +| converted_at | DATETIME | | | 转化时间 | 2023-11-10 09:00:00 | +| converted_contact_id | VARCHAR | 64 | | 转化后联系人ID（FK） | CNT_20231110001 | +| converted_account_id | VARCHAR | 64 | | 转化后企业ID（FK） | ACC_20231110001 | +| converted_opportunity_id | VARCHAR | 64 | | 转化后商机ID（FK） | OPP_20231110001 | +| custom_fields | JSON | | | 自定义字段 | {"product_interest": "AI"} | +| tags | JSON | | | 标签数组 | ["高意向","下载过白皮书"] | +| is_qualified | BOOLEAN | | | 是否为合格线索 | true | + +--- + +### Opportunity 详细字段设计 + +| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | +|--------|------|------|------|------|------| +| opportunity_id | VARCHAR | 64 | ✓ | 商机唯一ID（PK） | OPP_20231105001 | +| opportunity_name | VARCHAR | 200 | ✓ | 商机名称 | XX公司-AI平台采购项目 | +| account_id | VARCHAR | 64 | ✓ | 关联企业ID（FK） | ACC_20231105001 | +| primary_contact_id | VARCHAR | 64 | | 主要联系人ID（FK） | CNT_20231105001 | +| lead_id | VARCHAR | 64 | | 来源线索ID（FK） | LEAD_20231105001 | +| opportunity_type | VARCHAR | 50 | | 商机类型 | NEW_BUSINESS（新客户）/UPSELL（追加销售）/RENEWAL（续约）/CROSS_SELL（交叉销售） | +| opportunity_source | VARCHAR | 100 | | 商机来源 | LEAD_CONVERSION/DIRECT_SALES/PARTNER | +| amount | DECIMAL | (18,2) | | 金额 | 1000000.00 | +| currency | VARCHAR | 10 | | 货币 | CNY | +| stage | VARCHAR | 50 | ✓ | 阶段 | QUALIFICATION/NEEDS_ANALYSIS/PROPOSAL/NEGOTIATION/CLOSED_WON/CLOSED_LOST | +| probability | INT | | | 赢单概率（0-100） | 60 | +| expected_close_date | DATE | | | 预计成交日期 | 2023-12-31 | +| actual_close_date | DATE | | | 实际成交日期 | 2023-12-25 | +| close_reason | VARCHAR | 200 | | 关闭原因 | 价格因素/竞争对手/预算不足/成功签约 | +| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | +| product_ids | JSON | | | 产品ID数组 | ["PRD_001", "PRD_002"] | +| campaign_id | VARCHAR | 64 | | 来源活动ID（FK） | CMP_001 | +| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | +| custom_fields | JSON | | | 自定义字段 | {"contract_type": "annual"} | +| tags | JSON | | | 标签数组 | ["重点项目","Q4目标"] | +| days_in_stage | INT | | | 当前阶段停留天数 | 15 | +| is_won | BOOLEAN | | | 是否赢单 | false | +| is_lost | BOOLEAN | | | 是否输单 | false | + +--- + +### Channel 详细字段设计 + +| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | +|--------|------|------|------|------|------| +| channel_id | VARCHAR | 64 | ✓ | 渠道唯一ID（PK） | CH_001 | +| channel_name | VARCHAR | 100 | ✓ | 渠道名称 | 官网-产品页 | +| channel_type | VARCHAR | 50 | ✓ | 渠道类型 | WEBSITE/WECHAT/DOUYIN/EMAIL/PHONE/OFFLINE/PARTNER | +| channel_category | VARCHAR | 50 | | 渠道分类 | ONLINE（线上）/OFFLINE（线下）/SOCIAL（社交）/DIRECT（直销） | +| parent_channel_id | VARCHAR | 64 | | 父渠道ID（FK） | CH_PARENT_001 | +| channel_status | VARCHAR | 50 | ✓ | 渠道状态 | ACTIVE（活跃）/INACTIVE（停用）/TESTING（测试中） | +| channel_config | JSON | | | 渠道配置 | {"api_key": "xxx", "webhook_url": "xxx"} | +| cost | DECIMAL | (18,2) | | 成本 | 50000.00 | +| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | +| custom_fields | JSON | | | 自定义字段 | {"partner_name": "XX合作伙伴"} | + +--- + +### Touchpoint 详细字段设计 + +| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | +|--------|------|------|------|------|------| +| touchpoint_id | VARCHAR | 64 | ✓ | 触点唯一ID（PK） | TP_20231105001 | +| touchpoint_type | VARCHAR | 50 | ✓ | 触点类型 | PAGE_VIEW/FORM_SUBMIT/DOWNLOAD/EMAIL/CALL/MEETING/CHAT/SOCIAL | +| channel_id | VARCHAR | 64 | | 渠道ID（FK） | CH_001 | +| campaign_id | VARCHAR | 64 | | 活动ID（FK） | CMP_001 | +| contact_id | VARCHAR | 64 | | 联系人ID（FK） | CNT_20231105001 | +| lead_id | VARCHAR | 64 | | 线索ID（FK） | LEAD_20231105001 | +| account_id | VARCHAR | 64 | | 企业ID（FK） | ACC_20231105001 | +| touchpoint_time | DATETIME | ✓ | ✓ | 触点时间 | 2023-11-05 14:30:00 | +| touchpoint_direction | VARCHAR | 20 | | 触点方向 | INBOUND（入站）/OUTBOUND（出站） | +| touchpoint_status | VARCHAR | 50 | | 触点状态 | COMPLETED（完成）/SCHEDULED（已安排）/CANCELLED（取消） | +| content_type | VARCHAR | 50 | | 内容类型 | WHITEPAPER/CASE_STUDY/WEBINAR/DEMO/PROPOSAL | +| content_id | VARCHAR | 64 | | 内容ID（FK） | CONTENT_001 | +| subject | VARCHAR | 200 | | 主题 | 产品演示会议 | +| description | TEXT | | | 描述 | 讨论了AI平台的技术架构... | +| duration_seconds | INT | | | 持续时长（秒） | 3600 | +| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | +| metadata | JSON | | | 元数据 | {"device": "mobile", "location": "Beijing"} | +| utm_params | JSON | | | UTM参数 | {"utm_source": "baidu", "utm_medium": "cpc"} | +| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 14:30:00 | + +--- + +### Event 详细字段设计 + +| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | +|--------|------|------|------|------|------| +| event_id | VARCHAR | 64 | ✓ | 事件唯一ID（PK） | EVT_20231105001 | +| event_name | VARCHAR | 100 | ✓ | 事件名称 | page_view | +| event_type | VARCHAR | 50 | ✓ | 事件类型 | PAGE_VIEW/CLICK/FORM_START/FORM_SUBMIT/DOWNLOAD/VIDEO_PLAY/SEARCH | +| channel_id | VARCHAR | 64 | | 渠道ID（FK） | CH_001 | +| contact_id | VARCHAR | 64 | | 联系人ID（FK） | CNT_20231105001 | +| lead_id | VARCHAR | 64 | | 线索ID（FK） | LEAD_20231105001 | +| account_id | VARCHAR | 64 | | 企业ID（FK） | ACC_20231105001 | +| event_time | DATETIME | | ✓ | 事件时间 | 2023-11-05 14:35:20 | +| session_id | VARCHAR | 64 | | 会话ID | SESSION_20231105001 | +| device_type | VARCHAR | 50 | | 设备类型 | DESKTOP/MOBILE/TABLET | +| browser | VARCHAR | 50 | | 浏览器 | Chrome | +| os | VARCHAR | 50 | | 操作系统 | Windows 10 | +| ip_address | VARCHAR | 50 | | IP地址 | 192.168.1.1 | +| page_url | VARCHAR | 1000 | | 页面URL | https://www.example.com/product/ai-platform | +| referrer_url | VARCHAR | 1000 | | 来源URL | https://www.baidu.com/s?wd=AI平台 | +| event_properties | JSON | | | 事件属性 | {"button_text": "申请试用", "form_id": "trial_form"} | +| utm_params | JSON | | | UTM参数 | {"utm_source": "baidu", "utm_campaign": "Q4_campaign"} | +| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 14:35:20 | + +--- + +## 全渠道身份映射方案 + +### 身份识别优先级 + +```mermaid +graph LR + A[多渠道数据] --> B{身份识别} + B --> C[1. 手机号匹配] + B --> D[2. 邮箱匹配] + B --> E[3. 企业微信ID] + B --> F[4. 统一社会信用代码] + B --> G[5. 自定义ID] + + C --> H[合并至统一Contact] + D --> H + E --> H + F --> I[合并至统一Account] + G --> H + + H --> J[建立ChannelIdentity关联] + I --> K[建立AccountChannelIdentity关联] +``` + +### 身份合并规则 + +```mermaid +flowchart TD + Start[接收新数据] --> Check{是否存在标识符} + Check -->|有手机号| Phone[手机号匹配] + Check -->|有邮箱| Email[邮箱匹配] + Check -->|有微信ID| WeChat[微信ID匹配] + + Phone --> Match{找到匹配?} + Email --> Match + WeChat --> Match + + Match -->|是| Merge[合并到现有Contact] + Match -->|否| Create[创建新Contact] + + Merge --> AddIdentity[添加渠道身份] + Create --> AddIdentity + + AddIdentity --> UpdateScore[更新评分] + UpdateScore --> End[完成] +``` + +--- + +## 数据字典总结 + +### 核心实体数量统计 + +| 实体类型 | 实体数量 | 说明 | +|---------|---------|------| +| 客户主体实体 | 3 | Account, Contact, Lead | +| 业务实体 | 2 | Opportunity, Product | +| 营销实体 | 2 | Campaign, Channel | +| 交互实体 | 2 | Touchpoint, Event | +| 关系实体 | 6 | AccountContactRelation, AccountRelation, OpportunityProduct, CampaignMember, TagRelation, SegmentMember | +| 身份实体 | 3 | AccountChannelIdentity, ContactChannelIdentity, LeadChannelIdentity | +| 分析实体 | 5 | Segment, Tag, Score, Attribution, CustomerJourney | +| 支撑实体 | 2 | Industry, ProductCategory | +| **合计** | **25** | 覆盖B2B CDP核心业务场景 | + +--- + +## 扩展建议 + +### 可选扩展实体（根据业务需要） + +1. **Content（内容资产）** + - 营销内容管理 + - 内容效果追踪 + +2. **Order（订单）** + - 如需管理订单详情 + - 支持电商场景 + +3. **Contract（合同）** + - 合同管理 + - 续约提醒 + +4. **Partner（合作伙伴）** + - 渠道伙伴管理 + - 分销体系 + +5. **Competitor（竞争对手）** + - 竞争对手分析 + - 竞品情报 + +6. **Task（任务）** + - 销售任务管理 + - 跟进提醒 + +7. **Note（备注）** + - 客户备注 + - 沟通记录 + +--- + +## 技术实现建议 + +### 数据库选型建议 + +```mermaid +graph TB + subgraph "主数据存储" + PG[PostgreSQL
关系型数据
Account/Contact/Lead/Opportunity] + end + + subgraph "行为数据存储" + CH[ClickHouse
海量事件数据
Event/Touchpoint] + end + + subgraph "搜索引擎" + ES[Elasticsearch
全文搜索
客户搜索/标签搜索] + end + + subgraph "缓存层" + Redis[Redis
热数据缓存
评分/标签] + end + + subgraph "数据仓库" + DW[数据仓库
分析报表
BI分析] + end + + PG --> ES + PG --> Redis + CH --> DW + PG --> DW +``` + +### 关键索引建议 + +**Account表索引：** +```sql +-- 主键索引 +PRIMARY KEY (account_id) + +-- 业务索引 +CREATE INDEX idx_account_name ON Account(account_name); +CREATE INDEX idx_unified_code ON Account(unified_social_credit_code); +CREATE INDEX idx_account_status ON Account(account_status); +CREATE INDEX idx_account_owner ON Account(owner_user_id); +CREATE INDEX idx_account_created ON Account(created_at); + +-- 组合索引 +CREATE INDEX idx_account_type_status ON Account(account_type, account_status); +CREATE INDEX idx_account_city ON Account(province, city); +``` + +**Contact表索引：** +```sql +-- 主键索引 +PRIMARY KEY (contact_id) + +-- 业务索引 +CREATE INDEX idx_contact_phone ON Contact(mobile_phone); +CREATE INDEX idx_contact_email ON Contact(email); +CREATE INDEX idx_contact_wechat ON Contact(wechat_id); +CREATE INDEX idx_contact_account ON Contact(primary_account_id); +CREATE INDEX idx_contact_status ON Contact(contact_status); + +-- 组合索引 +CREATE INDEX idx_contact_phone_email ON Contact(mobile_phone, email); +``` + +**Event表索引（ClickHouse）：** +```sql +-- 主排序键 +ORDER BY (event_time, contact_id, event_type) + +-- 分区键 +PARTITION BY toYYYYMM(event_time) +``` + +--- + +## 总结 + +本实体设计方案包含： + +✅ **25个核心实体**，覆盖B2B CDP全业务场景 +✅ **全渠道身份映射**方案，支持跨渠道客户识别 +✅ **完整的客户生命周期**管理 +✅ **从线索到商机**的完整转化流程 +✅ **灵活的标签和分群**能力 +✅ **多维度的归因分析**能力 +✅ **详细的字段设计**和数据字典 + +该方案可以支撑： +- 全渠道客户数据整合 +- 客户360度画像 +- 精准营销和客户分群 +- 销售线索管理和转化 +- 客户旅程分析 +- 营销归因分析 +- 客户价值评估 + +根据实际业务需要，可以选择性实现部分实体，并在后续迭代中逐步完善。 From cd569c2bb03f2dd8e4deac72180306dec6552651 Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 10:43:53 +0000 Subject: [PATCH 02/10] Refactor: Improve B2B CDP entity design documentation Co-authored-by: 13585811473 <13585811473@163.com> --- docs/b2b-cdp-entity-design.md | 2324 +++++++++++++++++++-------------- 1 file changed, 1361 insertions(+), 963 deletions(-) diff --git a/docs/b2b-cdp-entity-design.md b/docs/b2b-cdp-entity-design.md index e00f4399b..db74c2aa4 100644 --- a/docs/b2b-cdp-entity-design.md +++ b/docs/b2b-cdp-entity-design.md @@ -1,6 +1,6 @@ # B2B CDP 实体设计详细方案 -## 📋 目录 +## 目录 - [整体架构](#整体架构) - [核心实体设计](#核心实体设计) - [实体关系图](#实体关系图) @@ -14,752 +14,858 @@ ### 实体分层架构图 ```mermaid -graph TB - subgraph "数据源层" - DS1[官网] - DS2[微信生态] - DS3[抖音] - DS4[线下活动] - DS5[电话/邮件] - DS6[CRM系统] - DS7[第三方平台] +flowchart TD + subgraph DataSource["数据源层 Data Source Layer"] + direction LR + DS1["官方网站
Official Website"] + DS2["微信生态
WeChat Ecosystem"] + DS3["抖音平台
Douyin Platform"] + DS4["线下活动
Offline Events"] + DS5["电话邮件
Phone & Email"] + DS6["CRM系统
CRM System"] + DS7["第三方平台
3rd Party Platform"] end - subgraph "渠道层" - Channel[Channel 渠道] + subgraph ChannelLayer["渠道层 Channel Layer"] + CH["Channel
渠道管理"] end - subgraph "交互层" - Touchpoint[Touchpoint 触点] - Event[Event 事件] - Campaign[Campaign 营销活动] + subgraph InteractionLayer["交互层 Interaction Layer"] + TP["Touchpoint
客户触点"] + EV["Event
行为事件"] + CM["Campaign
营销活动"] end - subgraph "客户主体层" - Contact[Contact 联系人] - Lead[Lead 线索] - Account[Account 企业账户] + subgraph EntityLayer["客户实体层 Customer Entity Layer"] + CT["Contact
联系人"] + LD["Lead
线索"] + AC["Account
企业账户"] + AS["AccountSummary
账户汇总数据"] end - subgraph "业务层" - Opportunity[Opportunity 商机] - Product[Product 产品] - Order[Order 订单] + subgraph BusinessLayer["业务层 Business Layer"] + OP["Opportunity
商机"] + PR["Product
产品"] + OR["Order
订单"] end - subgraph "洞察层" - Segment[Segment 客户分群] - Tag[Tag 标签] - Score[Score 评分] - Journey[Journey 客户旅程] - Attribution[Attribution 归因] + subgraph InsightLayer["洞察层 Insight Layer"] + SG["Segment
客户分群"] + TG["Tag
标签体系"] + SC["Score
评分模型"] + JN["Journey
客户旅程"] + AT["Attribution
营销归因"] end - DS1 & DS2 & DS3 & DS4 & DS5 & DS6 & DS7 --> Channel - Channel --> Touchpoint - Channel --> Event - Campaign --> Touchpoint - Touchpoint --> Contact - Touchpoint --> Lead - Event --> Contact - Contact --> Account - Lead --> Contact - Lead --> Opportunity - Account --> Opportunity - Opportunity --> Product - Opportunity --> Order - Contact --> Segment - Account --> Segment - Lead --> Tag - Contact --> Tag - Account --> Tag - Contact --> Score - Account --> Score - Lead --> Score - Contact --> Journey - Campaign --> Attribution - Opportunity --> Attribution + DataSource --> CH + CH --> TP + CH --> EV + CM --> TP + TP --> CT + TP --> LD + EV --> CT + CT --> AC + AC --> AS + LD --> CT + LD --> OP + AC --> OP + OP --> PR + OP --> OR + CT --> SG + AC --> SG + LD --> TG + CT --> TG + AC --> TG + CT --> SC + AC --> SC + LD --> SC + CT --> JN + CM --> AT + OP --> AT + + classDef sourceStyle fill:#e1f5ff,stroke:#01579b,stroke-width:2px + classDef channelStyle fill:#fff3e0,stroke:#e65100,stroke-width:2px + classDef interactionStyle fill:#f3e5f5,stroke:#4a148c,stroke-width:2px + classDef entityStyle fill:#e8f5e9,stroke:#1b5e20,stroke-width:2px + classDef businessStyle fill:#fff9c4,stroke:#f57f17,stroke-width:2px + classDef insightStyle fill:#fce4ec,stroke:#880e4f,stroke-width:2px + + class DS1,DS2,DS3,DS4,DS5,DS6,DS7 sourceStyle + class CH channelStyle + class TP,EV,CM interactionStyle + class CT,LD,AC,AS entityStyle + class OP,PR,OR businessStyle + class SG,TG,SC,JN,AT insightStyle ``` --- ## 核心实体设计 -### 1. Account（企业账户）- 核心实体 +### 1. Account (企业账户) - 核心实体 ```mermaid -classDiagram - class Account { - +String account_id PK - +String account_name - +String unified_social_credit_code - +String account_type - +String industry_id FK - +String account_status - +String account_level - +Decimal annual_revenue - +Integer employee_count - +String company_website - +String company_address - +String province - +String city - +String district - +String account_source - +String primary_channel_id FK - +String owner_user_id - +DateTime created_at - +DateTime updated_at - +JSON custom_fields - +String[] tags - +Integer health_score - +String lifecycle_stage +erDiagram + Account ||--o{ AccountChannelIdentity : "has" + Account ||--|| AccountSummary : "aggregates to" + Account ||--o{ AccountRelation : "parent of" + Account ||--o{ AccountRelation : "child of" + + Account { + varchar(64) account_id PK + varchar(200) account_name + varchar(18) unified_social_credit_code UK + varchar(50) account_type + varchar(64) industry_id FK + varchar(50) account_status + varchar(50) account_level + decimal(18_2) annual_revenue + int employee_count + varchar(500) company_website + varchar(500) company_address + varchar(50) province + varchar(50) city + varchar(50) district + varchar(100) account_source + varchar(64) primary_channel_id FK + varchar(64) owner_user_id FK + datetime created_at + datetime updated_at + json custom_fields + varchar(50) lifecycle_stage } - class AccountChannelIdentity { - +String identity_id PK - +String account_id FK - +String channel_id FK - +String channel_account_id - +String identity_type - +Boolean is_verified - +DateTime first_seen_at - +DateTime last_seen_at - +JSON additional_info + AccountChannelIdentity { + varchar(64) identity_id PK + varchar(64) account_id FK + varchar(64) channel_id FK + varchar(200) channel_account_id + varchar(50) identity_type + boolean is_verified + datetime first_seen_at + datetime last_seen_at + json additional_info } - class AccountRelation { - +String relation_id PK - +String parent_account_id FK - +String child_account_id FK - +String relation_type - +DateTime created_at + AccountSummary { + varchar(64) summary_id PK + varchar(64) account_id FK UK + int total_contacts + int total_opportunities + int total_leads + decimal(18_2) total_revenue + decimal(18_2) lifetime_value + int won_opportunities + int lost_opportunities + decimal(5_2) win_rate + int total_touchpoints + int active_campaigns + int health_score + datetime last_activity_at + datetime last_purchase_at + date first_purchase_date + date latest_opportunity_date + int days_since_last_contact + datetime calculated_at + datetime updated_at } - Account "1" --> "*" AccountChannelIdentity - Account "1" --> "*" AccountRelation : parent - Account "1" --> "*" AccountRelation : child + AccountRelation { + varchar(64) relation_id PK + varchar(64) parent_account_id FK + varchar(64) child_account_id FK + varchar(50) relation_type + datetime created_at + } ``` -**字段说明：** +**Account 字段说明:** - `account_id`: 全局唯一账户ID - `unified_social_credit_code`: 统一社会信用代码（企业唯一标识） -- `account_type`: 客户类型（潜在客户/现有客户/合作伙伴/竞争对手） -- `account_status`: 账户状态（活跃/休眠/流失/黑名单） -- `account_level`: 客户等级（战略级/重要级/普通级） -- `lifecycle_stage`: 生命周期阶段（认知期/考虑期/决策期/留存期/扩展期） +- `account_type`: 客户类型（CUSTOMER/PARTNER/COMPETITOR/PROSPECT） +- `account_status`: 账户状态（ACTIVE/DORMANT/CHURNED/BLACKLIST） +- `account_level`: 客户等级（STRATEGIC/IMPORTANT/NORMAL） +- `lifecycle_stage`: 生命周期阶段（AWARENESS/CONSIDERATION/DECISION/RETENTION/EXPANSION） + +**AccountSummary 字段说明:** +- `total_contacts`: 关联的联系人总数 +- `total_opportunities`: 商机总数 +- `total_revenue`: 累计收入 +- `lifetime_value`: 客户生命周期价值 +- `win_rate`: 赢单率 +- `health_score`: 健康度评分（0-100） +- `last_activity_at`: 最后活跃时间 +- `days_since_last_contact`: 距离上次联系天数 --- -### 2. Contact（联系人）- 核心实体 +### 2. Contact (联系人) - 核心实体 ```mermaid -classDiagram - class Contact { - +String contact_id PK - +String contact_name - +String mobile_phone - +String email - +String wechat_id - +String job_title - +String department - +String contact_status - +String primary_account_id FK - +String contact_source - +String primary_channel_id FK - +String owner_user_id - +DateTime created_at - +DateTime updated_at - +JSON custom_fields - +String[] tags - +Integer engagement_score - +String lifecycle_stage - +Boolean is_decision_maker - +Boolean is_verified +erDiagram + Contact ||--o{ ContactChannelIdentity : "has" + Contact ||--o{ AccountContactRelation : "belongs to" + Contact ||--|| ContactSummary : "aggregates to" + + Contact { + varchar(64) contact_id PK + varchar(100) contact_name + varchar(20) mobile_phone UK + varchar(200) email UK + varchar(100) wechat_id + varchar(100) job_title + varchar(100) department + varchar(50) contact_status + varchar(64) primary_account_id FK + varchar(100) contact_source + varchar(64) primary_channel_id FK + varchar(64) owner_user_id FK + datetime created_at + datetime updated_at + json custom_fields + varchar(50) lifecycle_stage + boolean is_decision_maker + boolean is_verified } - class ContactChannelIdentity { - +String identity_id PK - +String contact_id FK - +String channel_id FK - +String channel_user_id - +String identity_type - +Boolean is_primary - +Boolean is_verified - +DateTime first_seen_at - +DateTime last_seen_at - +JSON additional_info + ContactChannelIdentity { + varchar(64) identity_id PK + varchar(64) contact_id FK + varchar(64) channel_id FK + varchar(200) channel_user_id + varchar(50) identity_type + boolean is_primary + boolean is_verified + datetime first_seen_at + datetime last_seen_at + json additional_info } - class AccountContactRelation { - +String relation_id PK - +String account_id FK - +String contact_id FK - +String role_in_account - +String decision_level - +Boolean is_primary_contact - +String relationship_status - +DateTime relation_start_date - +DateTime relation_end_date - +DateTime created_at - +DateTime updated_at + ContactSummary { + varchar(64) summary_id PK + varchar(64) contact_id FK UK + int total_touchpoints + int total_events + int email_opens + int email_clicks + int form_submissions + int content_downloads + int engagement_score + datetime last_activity_at + datetime last_email_sent_at + datetime last_email_opened_at + int days_since_last_activity + int campaign_responses + datetime calculated_at + datetime updated_at } - Contact "1" --> "*" ContactChannelIdentity - Contact "1" --> "*" AccountContactRelation + AccountContactRelation { + varchar(64) relation_id PK + varchar(64) account_id FK + varchar(64) contact_id FK + varchar(100) role_in_account + varchar(50) decision_level + boolean is_primary_contact + varchar(50) relationship_status + date relation_start_date + date relation_end_date + datetime created_at + datetime updated_at + } ``` -**字段说明：** -- `decision_level`: 决策层级（决策者/影响者/使用者/把关者） -- `engagement_score`: 参与度评分（基于互动频率和深度） -- `role_in_account`: 在企业中的角色（CEO/CTO/采购经理等） +**Contact 字段说明:** +- `decision_level`: 决策层级（DECISION_MAKER/INFLUENCER/USER/GATEKEEPER） +- `lifecycle_stage`: 生命周期阶段（SUBSCRIBER/LEAD/MQL/SQL/OPPORTUNITY/CUSTOMER） + +**ContactSummary 字段说明:** +- `total_touchpoints`: 总触点数 +- `engagement_score`: 参与度评分（0-100） +- `email_opens`: 邮件打开次数 +- `email_clicks`: 邮件点击次数 +- `days_since_last_activity`: 距离上次活跃天数 --- -### 3. Lead（线索）- 核心实体 +### 3. Lead (线索) - 核心实体 ```mermaid -classDiagram - class Lead { - +String lead_id PK - +String lead_name - +String company_name - +String mobile_phone - +String email - +String wechat_id - +String job_title - +String lead_source - +String channel_id FK - +String campaign_id FK - +String lead_status - +Integer lead_score - +String lead_grade - +String industry_id FK - +String province - +String city - +String owner_user_id - +DateTime created_at - +DateTime updated_at - +DateTime last_contacted_at - +DateTime converted_at - +String converted_contact_id FK - +String converted_account_id FK - +String converted_opportunity_id FK - +JSON custom_fields - +String[] tags - +Boolean is_qualified +erDiagram + Lead ||--o{ LeadChannelIdentity : "has" + Lead ||--|| LeadSummary : "aggregates to" + + Lead { + varchar(64) lead_id PK + varchar(100) lead_name + varchar(200) company_name + varchar(20) mobile_phone + varchar(200) email + varchar(100) wechat_id + varchar(100) job_title + varchar(100) lead_source + varchar(64) channel_id FK + varchar(64) campaign_id FK + varchar(50) lead_status + int lead_score + varchar(10) lead_grade + varchar(64) industry_id FK + varchar(50) province + varchar(50) city + varchar(64) owner_user_id FK + datetime created_at + datetime updated_at + datetime last_contacted_at + datetime converted_at + varchar(64) converted_contact_id FK + varchar(64) converted_account_id FK + varchar(64) converted_opportunity_id FK + json custom_fields + boolean is_qualified } - class LeadChannelIdentity { - +String identity_id PK - +String lead_id FK - +String channel_id FK - +String channel_user_id - +DateTime captured_at - +JSON utm_params - +JSON additional_info + LeadChannelIdentity { + varchar(64) identity_id PK + varchar(64) lead_id FK + varchar(64) channel_id FK + varchar(200) channel_user_id + datetime captured_at + json utm_params + json additional_info } - Lead "1" --> "*" LeadChannelIdentity + LeadSummary { + varchar(64) summary_id PK + varchar(64) lead_id FK UK + int total_touchpoints + int total_events + int form_submissions + int content_downloads + int page_views + int days_in_pipeline + int contact_attempts + datetime last_activity_at + datetime last_contact_attempt_at + datetime calculated_at + datetime updated_at + } ``` -**字段说明：** -- `lead_status`: 线索状态（新建/联系中/已限定/已转化/无效） -- `lead_score`: 线索评分（基于行为和画像的综合评分） +**Lead 字段说明:** +- `lead_status`: 线索状态（NEW/CONTACTED/QUALIFIED/CONVERTED/DISQUALIFIED） +- `lead_score`: 线索评分（0-100） - `lead_grade`: 线索等级（A/B/C/D） - `is_qualified`: 是否为合格线索（MQL/SQL） --- -### 4. Opportunity（商机）- 核心实体 +### 4. Opportunity (商机) - 核心实体 ```mermaid -classDiagram - class Opportunity { - +String opportunity_id PK - +String opportunity_name - +String account_id FK - +String primary_contact_id FK - +String lead_id FK - +String opportunity_type - +String opportunity_source - +Decimal amount - +String currency - +String stage - +Integer probability - +Date expected_close_date - +Date actual_close_date - +String close_reason - +String owner_user_id - +String[] product_ids FK - +String campaign_id FK - +DateTime created_at - +DateTime updated_at - +JSON custom_fields - +String[] tags - +Integer days_in_stage - +Boolean is_won - +Boolean is_lost +erDiagram + Opportunity ||--o{ OpportunityStageHistory : "has" + Opportunity ||--o{ OpportunityProduct : "contains" + + Opportunity { + varchar(64) opportunity_id PK + varchar(200) opportunity_name + varchar(64) account_id FK + varchar(64) primary_contact_id FK + varchar(64) lead_id FK + varchar(50) opportunity_type + varchar(100) opportunity_source + decimal(18_2) amount + varchar(10) currency + varchar(50) stage + int probability + date expected_close_date + date actual_close_date + varchar(200) close_reason + varchar(64) owner_user_id FK + json product_ids + varchar(64) campaign_id FK + datetime created_at + datetime updated_at + json custom_fields + int days_in_stage + boolean is_won + boolean is_lost } - class OpportunityStageHistory { - +String history_id PK - +String opportunity_id FK - +String from_stage - +String to_stage - +DateTime changed_at - +String changed_by_user_id - +String change_reason - +Integer duration_days + OpportunityStageHistory { + varchar(64) history_id PK + varchar(64) opportunity_id FK + varchar(50) from_stage + varchar(50) to_stage + datetime changed_at + varchar(64) changed_by_user_id FK + varchar(200) change_reason + int duration_days } - class OpportunityProduct { - +String opp_product_id PK - +String opportunity_id FK - +String product_id FK - +Integer quantity - +Decimal unit_price - +Decimal total_price - +Decimal discount - +String product_description + OpportunityProduct { + varchar(64) opp_product_id PK + varchar(64) opportunity_id FK + varchar(64) product_id FK + int quantity + decimal(18_2) unit_price + decimal(18_2) total_price + decimal(18_2) discount + text product_description } - - Opportunity "1" --> "*" OpportunityStageHistory - Opportunity "1" --> "*" OpportunityProduct ``` -**字段说明：** -- `stage`: 阶段（线索/需求确认/方案设计/商务谈判/合同签订/已赢单/已输单） +**Opportunity 字段说明:** +- `stage`: 阶段（LEAD/QUALIFICATION/NEEDS_ANALYSIS/PROPOSAL/NEGOTIATION/CONTRACT/CLOSED_WON/CLOSED_LOST） - `probability`: 赢单概率（0-100） -- `opportunity_type`: 商机类型（新客户/追加销售/续约/交叉销售） +- `opportunity_type`: 商机类型（NEW_BUSINESS/UPSELL/RENEWAL/CROSS_SELL） --- -### 5. Channel（渠道）- 核心实体 +### 5. Channel (渠道) - 核心实体 ```mermaid -classDiagram - class Channel { - +String channel_id PK - +String channel_name - +String channel_type - +String channel_category - +String parent_channel_id FK - +String channel_status - +JSON channel_config - +Decimal cost - +DateTime created_at - +DateTime updated_at - +JSON custom_fields +erDiagram + Channel ||--o{ ChannelPerformance : "tracks" + + Channel { + varchar(64) channel_id PK + varchar(100) channel_name + varchar(50) channel_type + varchar(50) channel_category + varchar(64) parent_channel_id FK + varchar(50) channel_status + json channel_config + decimal(18_2) cost + datetime created_at + datetime updated_at + json custom_fields } - class ChannelPerformance { - +String performance_id PK - +String channel_id FK - +Date stat_date - +Integer lead_count - +Integer contact_count - +Integer opportunity_count - +Decimal revenue - +Decimal cost - +Decimal roi - +Integer conversion_count - +Decimal conversion_rate + ChannelPerformance { + varchar(64) performance_id PK + varchar(64) channel_id FK + date stat_date + int lead_count + int contact_count + int opportunity_count + decimal(18_2) revenue + decimal(18_2) cost + decimal(18_2) roi + int conversion_count + decimal(5_2) conversion_rate + datetime created_at } - - Channel "1" --> "*" ChannelPerformance ``` -**渠道类型枚举：** -- 线上渠道：官网、SEO、SEM、社交媒体、内容营销 -- 社交渠道：微信、企业微信、抖音、快手、小红书 -- 线下渠道：展会、研讨会、地推活动 -- 合作渠道：合作伙伴、代理商、分销商 -- 直销渠道：电话、邮件、销售拜访 +**渠道类型 (channel_type):** +- WEBSITE - 官方网站 +- SEO - 搜索引擎优化 +- SEM - 搜索引擎营销 +- WECHAT - 微信 +- ENTERPRISE_WECHAT - 企业微信 +- DOUYIN - 抖音 +- EMAIL - 邮件营销 +- PHONE - 电话 +- OFFLINE_EVENT - 线下活动 +- EXHIBITION - 展会 +- PARTNER - 合作伙伴 --- -### 6. Campaign（营销活动）- 核心实体 +### 6. Campaign (营销活动) - 核心实体 ```mermaid -classDiagram - class Campaign { - +String campaign_id PK - +String campaign_name - +String campaign_type - +String campaign_status - +String[] channel_ids FK - +Date start_date - +Date end_date - +Decimal budget - +Decimal actual_cost - +String target_audience - +String owner_user_id - +DateTime created_at - +DateTime updated_at - +JSON custom_fields - +String[] tags +erDiagram + Campaign ||--o{ CampaignPerformance : "tracks" + Campaign ||--o{ CampaignMember : "has" + + Campaign { + varchar(64) campaign_id PK + varchar(200) campaign_name + varchar(50) campaign_type + varchar(50) campaign_status + json channel_ids + date start_date + date end_date + decimal(18_2) budget + decimal(18_2) actual_cost + text target_audience + varchar(64) owner_user_id FK + datetime created_at + datetime updated_at + json custom_fields } - class CampaignPerformance { - +String performance_id PK - +String campaign_id FK - +Date stat_date - +Integer impressions - +Integer clicks - +Integer leads_generated - +Integer opportunities_generated - +Decimal revenue - +Decimal roi - +Decimal cpl - +Decimal cpa + CampaignPerformance { + varchar(64) performance_id PK + varchar(64) campaign_id FK + date stat_date + int impressions + int clicks + int leads_generated + int opportunities_generated + decimal(18_2) revenue + decimal(18_2) roi + decimal(18_2) cpl + decimal(18_2) cpa + datetime created_at } - class CampaignMember { - +String member_id PK - +String campaign_id FK - +String member_type - +String member_ref_id - +String member_status - +DateTime joined_at - +DateTime responded_at - +String response_status - +JSON response_data + CampaignMember { + varchar(64) member_id PK + varchar(64) campaign_id FK + varchar(50) member_type + varchar(64) member_ref_id FK + varchar(50) member_status + datetime joined_at + datetime responded_at + varchar(50) response_status + json response_data } - - Campaign "1" --> "*" CampaignPerformance - Campaign "1" --> "*" CampaignMember ``` -**活动类型：** -- 网络研讨会（Webinar） -- 线下会议/展会 -- 邮件营销 -- 内容营销（白皮书、案例分享） -- 产品试用活动 -- 行业峰会 +**活动类型 (campaign_type):** +- WEBINAR - 网络研讨会 +- CONFERENCE - 线下会议 +- EXHIBITION - 展会 +- EMAIL_MARKETING - 邮件营销 +- CONTENT_MARKETING - 内容营销 +- PRODUCT_TRIAL - 产品试用 +- SUMMIT - 行业峰会 --- -### 7. Touchpoint（触点/互动记录）- 核心实体 +### 7. Touchpoint (触点) - 核心实体 ```mermaid -classDiagram - class Touchpoint { - +String touchpoint_id PK - +String touchpoint_type - +String channel_id FK - +String campaign_id FK - +String contact_id FK - +String lead_id FK - +String account_id FK - +DateTime touchpoint_time - +String touchpoint_direction - +String touchpoint_status - +String content_type - +String content_id FK - +String subject - +Text description - +Integer duration_seconds - +String owner_user_id - +JSON metadata - +JSON utm_params - +DateTime created_at +erDiagram + Touchpoint ||--o{ TouchpointAttachment : "has" + + Touchpoint { + varchar(64) touchpoint_id PK + varchar(50) touchpoint_type + varchar(64) channel_id FK + varchar(64) campaign_id FK + varchar(64) contact_id FK + varchar(64) lead_id FK + varchar(64) account_id FK + datetime touchpoint_time + varchar(20) touchpoint_direction + varchar(50) touchpoint_status + varchar(50) content_type + varchar(64) content_id FK + varchar(200) subject + text description + int duration_seconds + varchar(64) owner_user_id FK + json metadata + json utm_params + datetime created_at } - class TouchpointAttachment { - +String attachment_id PK - +String touchpoint_id FK - +String file_name - +String file_url - +String file_type - +Integer file_size - +DateTime uploaded_at + TouchpointAttachment { + varchar(64) attachment_id PK + varchar(64) touchpoint_id FK + varchar(200) file_name + varchar(500) file_url + varchar(50) file_type + int file_size + datetime uploaded_at } - - Touchpoint "1" --> "*" TouchpointAttachment ``` -**触点类型：** -- 网站浏览 -- 表单提交 -- 内容下载 -- 邮件互动（打开/点击） -- 电话沟通 -- 会议/拜访 -- 在线聊天 -- 社交媒体互动 +**触点类型 (touchpoint_type):** +- PAGE_VIEW - 网站浏览 +- FORM_SUBMIT - 表单提交 +- DOWNLOAD - 内容下载 +- EMAIL_OPEN - 邮件打开 +- EMAIL_CLICK - 邮件点击 +- CALL - 电话沟通 +- MEETING - 会议拜访 +- CHAT - 在线聊天 +- SOCIAL_INTERACTION - 社交媒体互动 --- -### 8. Event（行为事件）- 核心实体 +### 8. Event (行为事件) - 核心实体 ```mermaid -classDiagram - class Event { - +String event_id PK - +String event_name - +String event_type - +String channel_id FK - +String contact_id FK - +String lead_id FK - +String account_id FK - +DateTime event_time - +String session_id - +String device_type - +String browser - +String os - +String ip_address - +String page_url - +String referrer_url - +JSON event_properties - +JSON utm_params - +DateTime created_at +erDiagram + Event { + varchar(64) event_id PK + varchar(100) event_name + varchar(50) event_type + varchar(64) channel_id FK + varchar(64) contact_id FK + varchar(64) lead_id FK + varchar(64) account_id FK + datetime event_time + varchar(64) session_id + varchar(50) device_type + varchar(50) browser + varchar(50) os + varchar(50) ip_address + varchar(1000) page_url + varchar(1000) referrer_url + json event_properties + json utm_params + datetime created_at } ``` -**事件类型：** -- 页面浏览（page_view） -- 按钮点击（button_click） -- 表单开始（form_start） -- 表单提交（form_submit） -- 文件下载（file_download） -- 视频播放（video_play） -- 产品试用（product_trial） -- 搜索（search） +**事件类型 (event_type):** +- PAGE_VIEW - 页面浏览 +- BUTTON_CLICK - 按钮点击 +- FORM_START - 表单开始 +- FORM_SUBMIT - 表单提交 +- FILE_DOWNLOAD - 文件下载 +- VIDEO_PLAY - 视频播放 +- PRODUCT_TRIAL - 产品试用 +- SEARCH - 搜索行为 --- -### 9. Product（产品/解决方案）- 核心实体 +### 9. Product (产品) - 核心实体 ```mermaid -classDiagram - class Product { - +String product_id PK - +String product_name - +String product_code - +String product_category - +String product_type - +String product_status - +Text description - +Decimal list_price - +String currency - +String pricing_model - +String[] feature_list - +DateTime created_at - +DateTime updated_at - +JSON custom_fields +erDiagram + Product }o--|| ProductCategory : "belongs to" + + Product { + varchar(64) product_id PK + varchar(200) product_name + varchar(100) product_code UK + varchar(64) product_category_id FK + varchar(50) product_type + varchar(50) product_status + text description + decimal(18_2) list_price + varchar(10) currency + varchar(50) pricing_model + json feature_list + datetime created_at + datetime updated_at + json custom_fields } - class ProductCategory { - +String category_id PK - +String category_name - +String parent_category_id FK - +Integer sort_order + ProductCategory { + varchar(64) category_id PK + varchar(100) category_name + varchar(64) parent_category_id FK + int level + int sort_order + datetime created_at } - - Product "*" --> "1" ProductCategory ``` --- -### 10. Tag（标签）- 核心实体 +### 10. Tag (标签) - 核心实体 ```mermaid -classDiagram - class Tag { - +String tag_id PK - +String tag_name - +String tag_category - +String tag_type - +String description - +String color - +DateTime created_at - +String created_by_user_id +erDiagram + Tag ||--o{ TagRelation : "applies to" + + Tag { + varchar(64) tag_id PK + varchar(100) tag_name UK + varchar(50) tag_category + varchar(50) tag_type + text description + varchar(20) color + datetime created_at + varchar(64) created_by_user_id FK } - class TagRelation { - +String relation_id PK - +String tag_id FK - +String entity_type - +String entity_id - +DateTime tagged_at - +String tagged_by_user_id - +Boolean is_auto_tagged - +String tag_source + TagRelation { + varchar(64) relation_id PK + varchar(64) tag_id FK + varchar(50) entity_type + varchar(64) entity_id FK + datetime tagged_at + varchar(64) tagged_by_user_id FK + boolean is_auto_tagged + varchar(100) tag_source } - - Tag "1" --> "*" TagRelation ``` -**标签类型：** -- 行为标签（高活跃度、近期浏览过产品A） -- 画像标签（互联网行业、大型企业、决策者） -- 业务标签（重点客户、流失风险、高价值） -- 兴趣标签（关注AI、关注云计算） +**标签类型 (tag_type):** +- BEHAVIOR - 行为标签 +- PROFILE - 画像标签 +- BUSINESS - 业务标签 +- INTEREST - 兴趣标签 --- -### 11. Segment（客户分群）- 核心实体 +### 11. Segment (客户分群) - 核心实体 ```mermaid -classDiagram - class Segment { - +String segment_id PK - +String segment_name - +String segment_type - +Text description - +JSON segment_rules - +String target_entity_type - +Integer member_count - +Boolean is_dynamic - +DateTime last_calculated_at - +String created_by_user_id - +DateTime created_at - +DateTime updated_at +erDiagram + Segment ||--o{ SegmentMember : "contains" + + Segment { + varchar(64) segment_id PK + varchar(200) segment_name + varchar(50) segment_type + text description + json segment_rules + varchar(50) target_entity_type + int member_count + boolean is_dynamic + datetime last_calculated_at + varchar(64) created_by_user_id FK + datetime created_at + datetime updated_at } - class SegmentMember { - +String member_id PK - +String segment_id FK - +String entity_type - +String entity_id - +DateTime joined_at - +DateTime left_at - +Boolean is_active + SegmentMember { + varchar(64) member_id PK + varchar(64) segment_id FK + varchar(50) entity_type + varchar(64) entity_id FK + datetime joined_at + datetime left_at + boolean is_active } - - Segment "1" --> "*" SegmentMember ``` --- -### 12. Score（评分模型）- 核心实体 +### 12. Score (评分模型) - 核心实体 ```mermaid -classDiagram - class ScoreModel { - +String model_id PK - +String model_name - +String model_type - +String target_entity_type - +JSON scoring_rules - +Integer max_score - +String status - +DateTime created_at - +DateTime updated_at +erDiagram + ScoreModel ||--o{ ScoreRecord : "generates" + ScoreRecord ||--o{ ScoreHistory : "tracks" + + ScoreModel { + varchar(64) model_id PK + varchar(100) model_name + varchar(50) model_type + varchar(50) target_entity_type + json scoring_rules + int max_score + varchar(50) status + datetime created_at + datetime updated_at } - class ScoreRecord { - +String record_id PK - +String model_id FK - +String entity_type - +String entity_id - +Integer score - +String grade - +JSON score_details - +DateTime calculated_at + ScoreRecord { + varchar(64) record_id PK + varchar(64) model_id FK + varchar(50) entity_type + varchar(64) entity_id FK + int score + varchar(10) grade + json score_details + datetime calculated_at + datetime updated_at } - class ScoreHistory { - +String history_id PK - +String entity_type - +String entity_id - +String model_id FK - +Integer score - +DateTime recorded_at + ScoreHistory { + varchar(64) history_id PK + varchar(50) entity_type + varchar(64) entity_id FK + varchar(64) model_id FK + int score + datetime recorded_at } - - ScoreModel "1" --> "*" ScoreRecord ``` -**评分类型：** -- Lead评分：基于行为和画像的线索评分 -- Account评分：企业健康度评分 -- Contact评分：联系人参与度评分 +**评分类型 (model_type):** +- LEAD_SCORING - 线索评分 +- ACCOUNT_HEALTH - 企业健康度评分 +- CONTACT_ENGAGEMENT - 联系人参与度评分 --- -### 13. Industry（行业）- 核心实体 +### 13. Industry (行业) - 核心实体 ```mermaid -classDiagram - class Industry { - +String industry_id PK - +String industry_name - +String industry_code - +String parent_industry_id FK - +Integer level - +Integer sort_order - +DateTime created_at +erDiagram + Industry ||--o{ Industry : "parent of" + + Industry { + varchar(64) industry_id PK + varchar(100) industry_name + varchar(50) industry_code UK + varchar(64) parent_industry_id FK + int level + int sort_order + datetime created_at } ``` --- -### 14. Attribution（归因）- 核心实体 +### 14. Attribution (归因) - 核心实体 ```mermaid -classDiagram - class Attribution { - +String attribution_id PK - +String entity_type - +String entity_id - +String attribution_model - +JSON touchpoint_sequence - +JSON attribution_weights - +DateTime created_at - +DateTime updated_at +erDiagram + Attribution ||--o{ TouchpointAttribution : "analyzes" + + Attribution { + varchar(64) attribution_id PK + varchar(50) entity_type + varchar(64) entity_id FK + varchar(50) attribution_model + json touchpoint_sequence + json attribution_weights + datetime created_at + datetime updated_at } - class TouchpointAttribution { - +String ta_id PK - +String attribution_id FK - +String touchpoint_id FK - +String campaign_id FK - +String channel_id FK - +Decimal attribution_weight - +Integer position_in_journey - +DateTime touchpoint_time + TouchpointAttribution { + varchar(64) ta_id PK + varchar(64) attribution_id FK + varchar(64) touchpoint_id FK + varchar(64) campaign_id FK + varchar(64) channel_id FK + decimal(5_4) attribution_weight + int position_in_journey + datetime touchpoint_time } - - Attribution "1" --> "*" TouchpointAttribution ``` -**归因模型：** -- 首次触点归因 -- 末次触点归因 -- 线性归因 -- 时间衰减归因 -- U型归因 -- W型归因 +**归因模型 (attribution_model):** +- FIRST_TOUCH - 首次触点归因 +- LAST_TOUCH - 末次触点归因 +- LINEAR - 线性归因 +- TIME_DECAY - 时间衰减归因 +- U_SHAPED - U型归因 +- W_SHAPED - W型归因 --- -### 15. CustomerJourney（客户旅程）- 核心实体 +### 15. CustomerJourney (客户旅程) - 核心实体 ```mermaid -classDiagram - class CustomerJourney { - +String journey_id PK - +String journey_name - +String entity_type - +String entity_id - +String journey_stage - +DateTime journey_start_at - +DateTime journey_end_at - +Integer total_touchpoints - +Integer journey_duration_days - +JSON stage_history - +DateTime created_at - +DateTime updated_at +erDiagram + CustomerJourney ||--o{ JourneyStage : "follows" + + CustomerJourney { + varchar(64) journey_id PK + varchar(200) journey_name + varchar(50) entity_type + varchar(64) entity_id FK + varchar(50) journey_stage + datetime journey_start_at + datetime journey_end_at + int total_touchpoints + int journey_duration_days + json stage_history + datetime created_at + datetime updated_at } - class JourneyStage { - +String stage_id PK - +String stage_name - +Integer stage_order - +String stage_category - +JSON milestone_criteria + JourneyStage { + varchar(64) stage_id PK + varchar(100) stage_name + int stage_order + varchar(50) stage_category + json milestone_criteria + datetime created_at } ``` @@ -771,86 +877,118 @@ classDiagram ```mermaid erDiagram - Account ||--o{ AccountContactRelation : has - Contact ||--o{ AccountContactRelation : belongs - Account ||--o{ Opportunity : has - Contact ||--o{ Opportunity : has - Lead ||--o| Contact : converts_to - Lead ||--o| Account : converts_to - Lead ||--o| Opportunity : converts_to - - Channel ||--o{ Touchpoint : generates - Campaign ||--o{ Touchpoint : generates - Contact ||--o{ Touchpoint : receives - Lead ||--o{ Touchpoint : receives - Account ||--o{ Touchpoint : receives - - Contact ||--o{ Event : generates - Lead ||--o{ Event : generates - Channel ||--o{ Event : tracks - - Opportunity ||--o{ OpportunityProduct : contains - Product ||--o{ OpportunityProduct : included_in - - Campaign ||--o{ Lead : generates - Campaign ||--o{ CampaignMember : has - - Account ||--o{ AccountChannelIdentity : has - Contact ||--o{ ContactChannelIdentity : has - Lead ||--o{ LeadChannelIdentity : has - Channel ||--o{ AccountChannelIdentity : identifies - Channel ||--o{ ContactChannelIdentity : identifies - Channel ||--o{ LeadChannelIdentity : identifies - - Tag ||--o{ TagRelation : applies_to - Segment ||--o{ SegmentMember : contains - - ScoreModel ||--o{ ScoreRecord : calculates - - Industry ||--o{ Account : categorizes - Industry ||--o{ Lead : categorizes - - Attribution ||--o{ TouchpointAttribution : analyzes - Touchpoint ||--o{ TouchpointAttribution : contributes_to - Campaign ||--o{ TouchpointAttribution : contributes_to - - CustomerJourney ||--o{ Touchpoint : tracks + Account ||--o{ AccountContactRelation : "has" + Account ||--|| AccountSummary : "aggregates" + Account ||--o{ Opportunity : "owns" + Account ||--o{ Touchpoint : "receives" + Account ||--o{ AccountChannelIdentity : "identifies in" + + Contact ||--o{ AccountContactRelation : "belongs to" + Contact ||--|| ContactSummary : "aggregates" + Contact ||--o{ Opportunity : "participates" + Contact ||--o{ Touchpoint : "receives" + Contact ||--o{ Event : "generates" + Contact ||--o{ ContactChannelIdentity : "identifies in" + + Lead ||--o| Contact : "converts to" + Lead ||--o| Account : "converts to" + Lead ||--o| Opportunity : "converts to" + Lead ||--|| LeadSummary : "aggregates" + Lead ||--o{ Touchpoint : "receives" + Lead ||--o{ Event : "generates" + Lead ||--o{ LeadChannelIdentity : "identifies in" + + Opportunity ||--o{ OpportunityStageHistory : "tracks" + Opportunity ||--o{ OpportunityProduct : "contains" + + Product ||--o{ OpportunityProduct : "included in" + Product }o--|| ProductCategory : "categorized by" + + Channel ||--o{ Touchpoint : "generates" + Channel ||--o{ Event : "tracks" + Channel ||--o{ ChannelPerformance : "measures" + Channel ||--o{ AccountChannelIdentity : "identifies" + Channel ||--o{ ContactChannelIdentity : "identifies" + Channel ||--o{ LeadChannelIdentity : "identifies" + + Campaign ||--o{ Touchpoint : "drives" + Campaign ||--o{ Lead : "generates" + Campaign ||--o{ CampaignMember : "includes" + Campaign ||--o{ CampaignPerformance : "measures" + + Tag ||--o{ TagRelation : "tags" + Segment ||--o{ SegmentMember : "groups" + + ScoreModel ||--o{ ScoreRecord : "scores" + ScoreRecord ||--o{ ScoreHistory : "history" + + Industry ||--o{ Account : "classifies" + Industry ||--o{ Lead : "classifies" + Industry ||--o{ Industry : "parent of" + + Attribution ||--o{ TouchpointAttribution : "attributes" + Touchpoint ||--o{ TouchpointAttribution : "contributes" + + CustomerJourney ||--o{ JourneyStage : "progresses" ``` --- -### 全渠道身份关联图 +### 全渠道身份统一架构 ```mermaid graph TB - subgraph "渠道身份体系" - WX[微信渠道身份] - WEB[官网渠道身份] - DY[抖音渠道身份] - EMAIL[邮件渠道身份] - PHONE[电话渠道身份] - OFFLINE[线下活动身份] + subgraph Sources["多渠道数据源"] + direction TB + S1["微信公众号
WeChat Official"] + S2["企业微信
Enterprise WeChat"] + S3["官方网站
Website"] + S4["抖音账号
Douyin"] + S5["邮件系统
Email"] + S6["电话系统
Phone"] + S7["线下活动
Offline Event"] end - subgraph "统一Contact" - Contact[Contact 联系人
统一ID: C001] + subgraph Identity["身份识别中心 Identity Resolution"] + direction TB + IR["Identity Resolution Engine
身份识别引擎"] + IM["Identity Matching
身份匹配"] + IG["Identity Grouping
身份聚合"] end - subgraph "统一Account" - Account[Account 企业
统一ID: A001] + subgraph Entities["统一客户实体 Unified Entities"] + direction TB + UC["Contact
统一联系人
ID: CNT_001"] + UA["Account
统一企业
ID: ACC_001"] end - WX --> |Identity Mapping| Contact - WEB --> |Identity Mapping| Contact - DY --> |Identity Mapping| Contact - EMAIL --> |Identity Mapping| Contact - PHONE --> |Identity Mapping| Contact - OFFLINE --> |Identity Mapping| Contact - - Contact --> |Belongs To| Account + subgraph Mapping["渠道身份映射 Channel Identity Mapping"] + direction TB + CI1["ContactChannelIdentity
微信OpenID"] + CI2["ContactChannelIdentity
企业微信UserID"] + CI3["ContactChannelIdentity
网站CookieID"] + CI4["ContactChannelIdentity
抖音UserID"] + AI1["AccountChannelIdentity
企业微信CorpID"] + AI2["AccountChannelIdentity
官网企业ID"] + end - style Contact fill:#ff9999 - style Account fill:#99ccff + S1 & S2 & S3 & S4 & S5 & S6 & S7 --> IR + IR --> IM + IM --> IG + IG --> UC + IG --> UA + UC --> CI1 & CI2 & CI3 & CI4 + UA --> AI1 & AI2 + + classDef sourceClass fill:#e3f2fd,stroke:#1976d2,stroke-width:2px + classDef identityClass fill:#fff3e0,stroke:#f57c00,stroke-width:2px + classDef entityClass fill:#e8f5e9,stroke:#388e3c,stroke-width:2px + classDef mappingClass fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px + + class S1,S2,S3,S4,S5,S6,S7 sourceClass + class IR,IM,IG identityClass + class UC,UA entityClass + class CI1,CI2,CI3,CI4,AI1,AI2 mappingClass ``` --- @@ -861,22 +999,34 @@ graph TB ```mermaid stateDiagram-v2 - [*] --> NewLead: 捕获线索 - NewLead --> Contacted: 首次联系 - Contacted --> Qualified: 资格验证 - Qualified --> Converted: 转化 - - Converted --> Contact: 创建联系人 - Converted --> Account: 创建/关联企业 - Converted --> Opportunity: 创建商机 - - Contact --> [*] - Account --> [*] - Opportunity --> [*] - - NewLead --> Invalid: 标记无效 - Contacted --> Invalid: 标记无效 - Invalid --> [*] + [*] --> NewLead: 捕获线索
Lead Capture + + state "新线索 New Lead" as NewLead + state "已联系 Contacted" as Contacted + state "已限定 Qualified" as Qualified + state "已转化 Converted" as Converted + state "无效线索 Invalid" as Invalid + + NewLead --> Contacted: 首次联系
First Contact + NewLead --> Invalid: 标记无效
Mark Invalid + + Contacted --> Qualified: 资格验证
Qualification + Contacted --> Invalid: 标记无效
Mark Invalid + + Qualified --> Converted: 转化
Convert + + state Converted { + [*] --> CreateContact: 创建联系人
Create Contact + [*] --> CreateAccount: 创建/关联企业
Create/Link Account + [*] --> CreateOpportunity: 创建商机
Create Opportunity + + CreateContact --> [*] + CreateAccount --> [*] + CreateOpportunity --> [*] + } + + Converted --> [*]: 完成
Complete + Invalid --> [*]: 结束
End ``` --- @@ -885,43 +1035,79 @@ stateDiagram-v2 ```mermaid stateDiagram-v2 - [*] --> Lead: 线索阶段 - Lead --> Qualification: 需求确认 - Qualification --> SolutionDesign: 方案设计 - SolutionDesign --> Negotiation: 商务谈判 - Negotiation --> Contract: 合同签订 - Contract --> Won: 赢单 - Contract --> Lost: 输单 - - Lead --> Lost: 丢失 - Qualification --> Lost: 丢失 - SolutionDesign --> Lost: 丢失 - Negotiation --> Lost: 丢失 - - Won --> [*] - Lost --> [*] + [*] --> Lead: 线索阶段
Lead Stage + + state "线索阶段 Lead" as Lead + state "需求确认 Qualification" as Qualification + state "方案设计 Solution Design" as SolutionDesign + state "商务谈判 Negotiation" as Negotiation + state "合同签订 Contract" as Contract + state "赢单 Won" as Won + state "输单 Lost" as Lost + + Lead --> Qualification: 需求分析
Needs Analysis + Lead --> Lost: 丢失
Lost + + Qualification --> SolutionDesign: 方案输出
Solution Output + Qualification --> Lost: 丢失
Lost + + SolutionDesign --> Negotiation: 商务谈判
Negotiation Start + SolutionDesign --> Lost: 丢失
Lost + + Negotiation --> Contract: 签订合同
Sign Contract + Negotiation --> Lost: 丢失
Lost + + Contract --> Won: 成功赢单
Close Won + Contract --> Lost: 失败输单
Close Lost + + Won --> [*]: 完成
Complete + Lost --> [*]: 结束
End + + note right of Won + 更新AccountSummary + Update Account Summary + end note ``` --- -### 客户生命周期流程 +### 客户生命周期管理 ```mermaid stateDiagram-v2 - [*] --> Awareness: 认知阶段 - Awareness --> Consideration: 考虑阶段 - Consideration --> Decision: 决策阶段 - Decision --> Retention: 留存阶段 - Retention --> Expansion: 扩展阶段 - - Retention --> Churn: 流失 - Expansion --> Churn: 流失 - - Churn --> Winback: 召回 - Winback --> Retention: 成功召回 - Winback --> [*]: 永久流失 - - Expansion --> [*]: 持续合作 + [*] --> Awareness: 认知阶段
Awareness + + state "认知阶段 Awareness" as Awareness + state "考虑阶段 Consideration" as Consideration + state "决策阶段 Decision" as Decision + state "留存阶段 Retention" as Retention + state "扩展阶段 Expansion" as Expansion + state "流失 Churn" as Churn + state "召回 Winback" as Winback + + Awareness --> Consideration: 产生兴趣
Show Interest + Consideration --> Decision: 深度评估
Deep Evaluation + Decision --> Retention: 成交购买
Purchase + Retention --> Expansion: 追加购买
Upsell/Cross-sell + + Retention --> Churn: 停止使用
Stop Using + Expansion --> Churn: 停止使用
Stop Using + + Churn --> Winback: 召回营销
Winback Campaign + Winback --> Retention: 成功召回
Successfully Winback + Winback --> [*]: 永久流失
Permanent Loss + + Expansion --> [*]: 持续合作
Ongoing Partnership + + note right of Retention + health_score > 70 + 健康客户 + end note + + note right of Churn + health_score < 30 + 预警流失 + end note ``` --- @@ -930,30 +1116,36 @@ stateDiagram-v2 ```mermaid sequenceDiagram - participant User as 用户 - participant Channel as 渠道 - participant Event as 事件系统 - participant Identity as 身份识别 - participant CDP as CDP核心 - participant Lead as Lead管理 - participant Contact as Contact管理 - participant Account as Account管理 - - User->>Channel: 1. 访问/互动 - Channel->>Event: 2. 记录事件 - Event->>Identity: 3. 身份识别 - - alt 新用户 - Identity->>Lead: 4a. 创建Lead - Lead->>CDP: 5a. 保存Lead数据 - else 已识别用户 - Identity->>Contact: 4b. 关联Contact - Contact->>Account: 5b. 关联Account + actor User as 用户
User + participant CH as 渠道
Channel + participant EV as 事件系统
Event System + participant ID as 身份识别
Identity Resolution + participant CDP as CDP核心
CDP Core + participant LD as Lead管理
Lead Management + participant CT as Contact管理
Contact Management + participant AC as Account管理
Account Management + participant SM as 汇总计算
Summary Calculation + + User->>CH: 1. 访问互动
Visit/Interact + CH->>EV: 2. 记录事件
Log Event + EV->>ID: 3. 身份识别
Identify + + alt 新用户 New User + ID->>LD: 4a. 创建Lead
Create Lead + LD->>CDP: 5a. 保存数据
Save Data + CDP->>SM: 6a. 更新LeadSummary
Update Summary + else 已识别用户 Identified User + ID->>CT: 4b. 关联Contact
Link Contact + CT->>AC: 5b. 关联Account
Link Account + AC->>CDP: 6b. 保存数据
Save Data + CDP->>SM: 7b. 更新Summary
Update Summary end - CDP->>Event: 6. 触发规则引擎 - Event->>Channel: 7. 个性化响应 - Channel->>User: 8. 返回内容 + SM->>CDP: 8. 触发规则引擎
Trigger Rules + CDP->>CH: 9. 个性化响应
Personalized Response + CH->>User: 10. 返回内容
Return Content + + Note over SM: ContactSummary
AccountSummary
LeadSummary ``` --- @@ -962,392 +1154,598 @@ sequenceDiagram ### Account 详细字段设计 -| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | -|--------|------|------|------|------|------| -| account_id | VARCHAR | 64 | ✓ | 账户唯一ID（PK） | ACC_20231105001 | -| account_name | VARCHAR | 200 | ✓ | 企业名称 | 阿里巴巴网络技术有限公司 | -| unified_social_credit_code | VARCHAR | 18 | | 统一社会信用代码 | 91330000MA27XYZ123 | -| account_type | VARCHAR | 50 | ✓ | 客户类型 | CUSTOMER（客户）/PARTNER（合作伙伴）/COMPETITOR（竞争对手）/PROSPECT（潜在客户） | -| industry_id | VARCHAR | 64 | | 行业ID（FK） | IND_001 | -| account_status | VARCHAR | 50 | ✓ | 账户状态 | ACTIVE（活跃）/DORMANT（休眠）/CHURNED（流失）/BLACKLIST（黑名单） | -| account_level | VARCHAR | 50 | | 客户等级 | STRATEGIC（战略级）/IMPORTANT（重要级）/NORMAL（普通级） | -| annual_revenue | DECIMAL | (18,2) | | 年营收（万元） | 50000.00 | +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| account_id | VARCHAR | 64 | PK, NOT NULL | 账户唯一标识 | ACC_20231105001 | +| account_name | VARCHAR | 200 | NOT NULL | 企业名称 | 阿里巴巴网络技术有限公司 | +| unified_social_credit_code | VARCHAR | 18 | UNIQUE | 统一社会信用代码 | 91330000MA27XYZ123 | +| account_type | VARCHAR | 50 | NOT NULL | 客户类型 | CUSTOMER / PARTNER / COMPETITOR / PROSPECT | +| industry_id | VARCHAR | 64 | FK | 行业分类外键 | IND_001 | +| account_status | VARCHAR | 50 | NOT NULL | 账户状态 | ACTIVE / DORMANT / CHURNED / BLACKLIST | +| account_level | VARCHAR | 50 | | 客户等级 | STRATEGIC / IMPORTANT / NORMAL | +| annual_revenue | DECIMAL | 18,2 | | 年营收(万元) | 50000.00 | | employee_count | INT | | | 员工人数 | 5000 | | company_website | VARCHAR | 500 | | 公司网站 | https://www.alibaba.com | | company_address | VARCHAR | 500 | | 公司地址 | 浙江省杭州市余杭区文一西路969号 | | province | VARCHAR | 50 | | 省份 | 浙江省 | | city | VARCHAR | 50 | | 城市 | 杭州市 | | district | VARCHAR | 50 | | 区县 | 余杭区 | -| account_source | VARCHAR | 100 | | 来源 | WEBSITE/EXHIBITION/PARTNER/COLD_CALL | -| primary_channel_id | VARCHAR | 64 | | 主渠道ID（FK） | CH_001 | -| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | -| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | -| custom_fields | JSON | | | 自定义字段 | {"crm_id": "CRM001"} | -| tags | JSON | | | 标签数组 | ["高价值客户","AI行业"] | -| health_score | INT | | | 健康度评分（0-100） | 85 | -| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | AWARENESS/CONSIDERATION/DECISION/RETENTION/EXPANSION | +| account_source | VARCHAR | 100 | | 来源渠道 | WEBSITE / EXHIBITION / PARTNER / COLD_CALL | +| primary_channel_id | VARCHAR | 64 | FK | 主渠道ID | CH_001 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| custom_fields | JSON | | | 自定义扩展字段 | {"crm_id": "CRM001", "region": "east"} | +| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | AWARENESS / CONSIDERATION / DECISION / RETENTION / EXPANSION | + +--- + +### AccountSummary 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | ACCS_20231105001 | +| account_id | VARCHAR | 64 | FK, UNIQUE, NOT NULL | 账户ID | ACC_20231105001 | +| total_contacts | INT | | DEFAULT 0 | 关联联系人总数 | 25 | +| total_opportunities | INT | | DEFAULT 0 | 商机总数 | 8 | +| total_leads | INT | | DEFAULT 0 | 线索总数 | 45 | +| total_revenue | DECIMAL | 18,2 | DEFAULT 0 | 累计收入(元) | 5000000.00 | +| lifetime_value | DECIMAL | 18,2 | DEFAULT 0 | 生命周期价值(元) | 8000000.00 | +| won_opportunities | INT | | DEFAULT 0 | 赢单数量 | 5 | +| lost_opportunities | INT | | DEFAULT 0 | 输单数量 | 2 | +| win_rate | DECIMAL | 5,2 | DEFAULT 0 | 赢单率(%) | 71.43 | +| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 156 | +| active_campaigns | INT | | DEFAULT 0 | 活跃营销活动数 | 3 | +| health_score | INT | | DEFAULT 0 | 健康度评分(0-100) | 85 | +| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | +| last_purchase_at | DATETIME | | | 最后购买时间 | 2023-10-15 09:00:00 | +| first_purchase_date | DATE | | | 首次购买日期 | 2022-03-20 | +| latest_opportunity_date | DATE | | | 最新商机日期 | 2023-11-01 | +| days_since_last_contact | INT | | DEFAULT 0 | 距上次联系天数 | 5 | +| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | + +**AccountSummary 设计说明:** +1. 该表用于存储Account的所有汇总统计数据，避免频繁实时计算 +2. 建议通过定时任务(如每小时)或触发器更新 +3. 用于快速查询和展示客户360度视图 +4. 支持客户健康度监控和预警 --- ### Contact 详细字段设计 -| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | -|--------|------|------|------|------|------| -| contact_id | VARCHAR | 64 | ✓ | 联系人唯一ID（PK） | CNT_20231105001 | -| contact_name | VARCHAR | 100 | ✓ | 联系人姓名 | 张伟 | -| mobile_phone | VARCHAR | 20 | | 手机号 | 13800138000 | -| email | VARCHAR | 200 | | 邮箱 | zhangwei@company.com | +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| contact_id | VARCHAR | 64 | PK, NOT NULL | 联系人唯一标识 | CNT_20231105001 | +| contact_name | VARCHAR | 100 | NOT NULL | 联系人姓名 | 张伟 | +| mobile_phone | VARCHAR | 20 | UNIQUE | 手机号 | 13800138000 | +| email | VARCHAR | 200 | UNIQUE | 邮箱地址 | zhangwei@company.com | | wechat_id | VARCHAR | 100 | | 微信ID | wx_zhangwei | -| job_title | VARCHAR | 100 | | 职位 | CTO | +| job_title | VARCHAR | 100 | | 职位 | 首席技术官 CTO | | department | VARCHAR | 100 | | 部门 | 技术部 | -| contact_status | VARCHAR | 50 | ✓ | 联系人状态 | ACTIVE（活跃）/INACTIVE（不活跃）/BOUNCED（退订）/UNSUBSCRIBED（取消订阅） | -| primary_account_id | VARCHAR | 64 | | 主要关联企业ID（FK） | ACC_20231105001 | -| contact_source | VARCHAR | 100 | | 来源 | WEBSITE/FORM/IMPORT/API | -| primary_channel_id | VARCHAR | 64 | | 主渠道ID（FK） | CH_001 | -| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | -| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | -| custom_fields | JSON | | | 自定义字段 | {"birthday": "1985-01-01"} | -| tags | JSON | | | 标签数组 | ["决策者","技术背景"] | -| engagement_score | INT | | | 参与度评分（0-100） | 75 | -| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | SUBSCRIBER/LEAD/MQL/SQL/OPPORTUNITY/CUSTOMER | -| is_decision_maker | BOOLEAN | | | 是否决策者 | true | -| is_verified | BOOLEAN | | | 是否已验证 | true | +| contact_status | VARCHAR | 50 | NOT NULL | 联系人状态 | ACTIVE / INACTIVE / BOUNCED / UNSUBSCRIBED | +| primary_account_id | VARCHAR | 64 | FK | 主要关联企业ID | ACC_20231105001 | +| contact_source | VARCHAR | 100 | | 来源 | WEBSITE / FORM / IMPORT / API / REFERRAL | +| primary_channel_id | VARCHAR | 64 | FK | 主渠道ID | CH_001 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| custom_fields | JSON | | | 自定义字段 | {"birthday": "1985-01-01", "hobby": "reading"} | +| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | SUBSCRIBER / LEAD / MQL / SQL / OPPORTUNITY / CUSTOMER | +| is_decision_maker | BOOLEAN | | DEFAULT FALSE | 是否决策者 | true | +| is_verified | BOOLEAN | | DEFAULT FALSE | 是否已验证 | true | + +--- + +### ContactSummary 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | CNTS_20231105001 | +| contact_id | VARCHAR | 64 | FK, UNIQUE, NOT NULL | 联系人ID | CNT_20231105001 | +| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 87 | +| total_events | INT | | DEFAULT 0 | 总事件数 | 234 | +| email_opens | INT | | DEFAULT 0 | 邮件打开次数 | 45 | +| email_clicks | INT | | DEFAULT 0 | 邮件点击次数 | 23 | +| form_submissions | INT | | DEFAULT 0 | 表单提交次数 | 12 | +| content_downloads | INT | | DEFAULT 0 | 内容下载次数 | 8 | +| engagement_score | INT | | DEFAULT 0 | 参与度评分(0-100) | 78 | +| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | +| last_email_sent_at | DATETIME | | | 最后邮件发送时间 | 2023-11-04 10:00:00 | +| last_email_opened_at | DATETIME | | | 最后邮件打开时间 | 2023-11-04 15:30:00 | +| days_since_last_activity | INT | | DEFAULT 0 | 距上次活跃天数 | 1 | +| campaign_responses | INT | | DEFAULT 0 | 营销活动响应次数 | 5 | +| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | --- ### Lead 详细字段设计 -| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | -|--------|------|------|------|------|------| -| lead_id | VARCHAR | 64 | ✓ | 线索唯一ID（PK） | LEAD_20231105001 | -| lead_name | VARCHAR | 100 | ✓ | 线索姓名 | 李明 | -| company_name | VARCHAR | 200 | | 公司名称 | XX科技有限公司 | +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| lead_id | VARCHAR | 64 | PK, NOT NULL | 线索唯一标识 | LEAD_20231105001 | +| lead_name | VARCHAR | 100 | NOT NULL | 线索姓名 | 李明 | +| company_name | VARCHAR | 200 | | 公司名称 | 腾讯科技有限公司 | | mobile_phone | VARCHAR | 20 | | 手机号 | 13900139000 | -| email | VARCHAR | 200 | | 邮箱 | liming@company.com | +| email | VARCHAR | 200 | | 邮箱地址 | liming@company.com | | wechat_id | VARCHAR | 100 | | 微信ID | wx_liming | | job_title | VARCHAR | 100 | | 职位 | 产品经理 | -| lead_source | VARCHAR | 100 | ✓ | 线索来源 | WEBSITE/FORM/CAMPAIGN/COLD_CALL/REFERRAL | -| channel_id | VARCHAR | 64 | | 渠道ID（FK） | CH_001 | -| campaign_id | VARCHAR | 64 | | 营销活动ID（FK） | CMP_001 | -| lead_status | VARCHAR | 50 | ✓ | 线索状态 | NEW（新建）/CONTACTED（已联系）/QUALIFIED（已限定）/CONVERTED（已转化）/DISQUALIFIED（无效） | -| lead_score | INT | | | 线索评分（0-100） | 80 | -| lead_grade | VARCHAR | 10 | | 线索等级 | A/B/C/D | -| industry_id | VARCHAR | 64 | | 行业ID（FK） | IND_001 | +| lead_source | VARCHAR | 100 | NOT NULL | 线索来源 | WEBSITE / FORM / CAMPAIGN / COLD_CALL / REFERRAL | +| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | +| campaign_id | VARCHAR | 64 | FK | 营销活动ID | CMP_001 | +| lead_status | VARCHAR | 50 | NOT NULL | 线索状态 | NEW / CONTACTED / QUALIFIED / CONVERTED / DISQUALIFIED | +| lead_score | INT | | DEFAULT 0 | 线索评分(0-100) | 80 | +| lead_grade | VARCHAR | 10 | | 线索等级 | A / B / C / D | +| industry_id | VARCHAR | 64 | FK | 行业ID | IND_001 | | province | VARCHAR | 50 | | 省份 | 广东省 | | city | VARCHAR | 50 | | 城市 | 深圳市 | -| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | -| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | | last_contacted_at | DATETIME | | | 最后联系时间 | 2023-11-05 14:00:00 | | converted_at | DATETIME | | | 转化时间 | 2023-11-10 09:00:00 | -| converted_contact_id | VARCHAR | 64 | | 转化后联系人ID（FK） | CNT_20231110001 | -| converted_account_id | VARCHAR | 64 | | 转化后企业ID（FK） | ACC_20231110001 | -| converted_opportunity_id | VARCHAR | 64 | | 转化后商机ID（FK） | OPP_20231110001 | -| custom_fields | JSON | | | 自定义字段 | {"product_interest": "AI"} | -| tags | JSON | | | 标签数组 | ["高意向","下载过白皮书"] | -| is_qualified | BOOLEAN | | | 是否为合格线索 | true | +| converted_contact_id | VARCHAR | 64 | FK | 转化后联系人ID | CNT_20231110001 | +| converted_account_id | VARCHAR | 64 | FK | 转化后企业ID | ACC_20231110001 | +| converted_opportunity_id | VARCHAR | 64 | FK | 转化后商机ID | OPP_20231110001 | +| custom_fields | JSON | | | 自定义字段 | {"product_interest": "AI", "budget": "100K"} | +| is_qualified | BOOLEAN | | DEFAULT FALSE | 是否为合格线索 | true | + +--- + +### LeadSummary 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | LEADS_20231105001 | +| lead_id | VARCHAR | 64 | FK, UNIQUE, NOT NULL | 线索ID | LEAD_20231105001 | +| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 12 | +| total_events | INT | | DEFAULT 0 | 总事件数 | 45 | +| form_submissions | INT | | DEFAULT 0 | 表单提交次数 | 3 | +| content_downloads | INT | | DEFAULT 0 | 内容下载次数 | 2 | +| page_views | INT | | DEFAULT 0 | 页面浏览次数 | 28 | +| days_in_pipeline | INT | | DEFAULT 0 | 在管道中天数 | 7 | +| contact_attempts | INT | | DEFAULT 0 | 联系尝试次数 | 4 | +| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | +| last_contact_attempt_at | DATETIME | | | 最后联系尝试时间 | 2023-11-05 11:00:00 | +| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | --- ### Opportunity 详细字段设计 -| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | -|--------|------|------|------|------|------| -| opportunity_id | VARCHAR | 64 | ✓ | 商机唯一ID（PK） | OPP_20231105001 | -| opportunity_name | VARCHAR | 200 | ✓ | 商机名称 | XX公司-AI平台采购项目 | -| account_id | VARCHAR | 64 | ✓ | 关联企业ID（FK） | ACC_20231105001 | -| primary_contact_id | VARCHAR | 64 | | 主要联系人ID（FK） | CNT_20231105001 | -| lead_id | VARCHAR | 64 | | 来源线索ID（FK） | LEAD_20231105001 | -| opportunity_type | VARCHAR | 50 | | 商机类型 | NEW_BUSINESS（新客户）/UPSELL（追加销售）/RENEWAL（续约）/CROSS_SELL（交叉销售） | -| opportunity_source | VARCHAR | 100 | | 商机来源 | LEAD_CONVERSION/DIRECT_SALES/PARTNER | -| amount | DECIMAL | (18,2) | | 金额 | 1000000.00 | -| currency | VARCHAR | 10 | | 货币 | CNY | -| stage | VARCHAR | 50 | ✓ | 阶段 | QUALIFICATION/NEEDS_ANALYSIS/PROPOSAL/NEGOTIATION/CLOSED_WON/CLOSED_LOST | -| probability | INT | | | 赢单概率（0-100） | 60 | +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| opportunity_id | VARCHAR | 64 | PK, NOT NULL | 商机唯一标识 | OPP_20231105001 | +| opportunity_name | VARCHAR | 200 | NOT NULL | 商机名称 | 腾讯科技-AI平台采购项目 | +| account_id | VARCHAR | 64 | FK, NOT NULL | 关联企业ID | ACC_20231105001 | +| primary_contact_id | VARCHAR | 64 | FK | 主要联系人ID | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | 来源线索ID | LEAD_20231105001 | +| opportunity_type | VARCHAR | 50 | | 商机类型 | NEW_BUSINESS / UPSELL / RENEWAL / CROSS_SELL | +| opportunity_source | VARCHAR | 100 | | 商机来源 | LEAD_CONVERSION / DIRECT_SALES / PARTNER | +| amount | DECIMAL | 18,2 | | 预计金额(元) | 1000000.00 | +| currency | VARCHAR | 10 | | 货币单位 | CNY / USD / EUR | +| stage | VARCHAR | 50 | NOT NULL | 当前阶段 | QUALIFICATION / NEEDS_ANALYSIS / PROPOSAL / NEGOTIATION / CLOSED_WON / CLOSED_LOST | +| probability | INT | | DEFAULT 0 | 赢单概率(0-100) | 60 | | expected_close_date | DATE | | | 预计成交日期 | 2023-12-31 | | actual_close_date | DATE | | | 实际成交日期 | 2023-12-25 | -| close_reason | VARCHAR | 200 | | 关闭原因 | 价格因素/竞争对手/预算不足/成功签约 | -| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | -| product_ids | JSON | | | 产品ID数组 | ["PRD_001", "PRD_002"] | -| campaign_id | VARCHAR | 64 | | 来源活动ID（FK） | CMP_001 | -| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | -| custom_fields | JSON | | | 自定义字段 | {"contract_type": "annual"} | -| tags | JSON | | | 标签数组 | ["重点项目","Q4目标"] | -| days_in_stage | INT | | | 当前阶段停留天数 | 15 | -| is_won | BOOLEAN | | | 是否赢单 | false | -| is_lost | BOOLEAN | | | 是否输单 | false | +| close_reason | VARCHAR | 200 | | 关闭原因 | 价格因素 / 竞争对手 / 预算不足 / 成功签约 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| product_ids | JSON | | | 关联产品ID数组 | ["PRD_001", "PRD_002"] | +| campaign_id | VARCHAR | 64 | FK | 来源活动ID | CMP_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| custom_fields | JSON | | | 自定义字段 | {"contract_type": "annual", "payment_terms": "quarterly"} | +| days_in_stage | INT | | DEFAULT 0 | 当前阶段停留天数 | 15 | +| is_won | BOOLEAN | | DEFAULT FALSE | 是否赢单 | false | +| is_lost | BOOLEAN | | DEFAULT FALSE | 是否输单 | false | --- ### Channel 详细字段设计 -| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | -|--------|------|------|------|------|------| -| channel_id | VARCHAR | 64 | ✓ | 渠道唯一ID（PK） | CH_001 | -| channel_name | VARCHAR | 100 | ✓ | 渠道名称 | 官网-产品页 | -| channel_type | VARCHAR | 50 | ✓ | 渠道类型 | WEBSITE/WECHAT/DOUYIN/EMAIL/PHONE/OFFLINE/PARTNER | -| channel_category | VARCHAR | 50 | | 渠道分类 | ONLINE（线上）/OFFLINE（线下）/SOCIAL（社交）/DIRECT（直销） | -| parent_channel_id | VARCHAR | 64 | | 父渠道ID（FK） | CH_PARENT_001 | -| channel_status | VARCHAR | 50 | ✓ | 渠道状态 | ACTIVE（活跃）/INACTIVE（停用）/TESTING（测试中） | -| channel_config | JSON | | | 渠道配置 | {"api_key": "xxx", "webhook_url": "xxx"} | -| cost | DECIMAL | (18,2) | | 成本 | 50000.00 | -| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | ✓ | 更新时间 | 2023-11-05 10:30:00 | -| custom_fields | JSON | | | 自定义字段 | {"partner_name": "XX合作伙伴"} | +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| channel_id | VARCHAR | 64 | PK, NOT NULL | 渠道唯一标识 | CH_001 | +| channel_name | VARCHAR | 100 | NOT NULL | 渠道名称 | 官网-产品页 | +| channel_type | VARCHAR | 50 | NOT NULL | 渠道类型 | WEBSITE / WECHAT / DOUYIN / EMAIL / PHONE / OFFLINE / PARTNER | +| channel_category | VARCHAR | 50 | | 渠道分类 | ONLINE / OFFLINE / SOCIAL / DIRECT | +| parent_channel_id | VARCHAR | 64 | FK | 父渠道ID | CH_PARENT_001 | +| channel_status | VARCHAR | 50 | NOT NULL | 渠道状态 | ACTIVE / INACTIVE / TESTING | +| channel_config | JSON | | | 渠道配置信息 | {"api_key": "xxx", "webhook_url": "xxx", "app_id": "xxx"} | +| cost | DECIMAL | 18,2 | | 渠道成本(元/月) | 50000.00 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| custom_fields | JSON | | | 自定义字段 | {"partner_name": "某某合作伙伴", "contract_end": "2024-12-31"} | --- ### Touchpoint 详细字段设计 -| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | -|--------|------|------|------|------|------| -| touchpoint_id | VARCHAR | 64 | ✓ | 触点唯一ID（PK） | TP_20231105001 | -| touchpoint_type | VARCHAR | 50 | ✓ | 触点类型 | PAGE_VIEW/FORM_SUBMIT/DOWNLOAD/EMAIL/CALL/MEETING/CHAT/SOCIAL | -| channel_id | VARCHAR | 64 | | 渠道ID（FK） | CH_001 | -| campaign_id | VARCHAR | 64 | | 活动ID（FK） | CMP_001 | -| contact_id | VARCHAR | 64 | | 联系人ID（FK） | CNT_20231105001 | -| lead_id | VARCHAR | 64 | | 线索ID（FK） | LEAD_20231105001 | -| account_id | VARCHAR | 64 | | 企业ID（FK） | ACC_20231105001 | -| touchpoint_time | DATETIME | ✓ | ✓ | 触点时间 | 2023-11-05 14:30:00 | -| touchpoint_direction | VARCHAR | 20 | | 触点方向 | INBOUND（入站）/OUTBOUND（出站） | -| touchpoint_status | VARCHAR | 50 | | 触点状态 | COMPLETED（完成）/SCHEDULED（已安排）/CANCELLED（取消） | -| content_type | VARCHAR | 50 | | 内容类型 | WHITEPAPER/CASE_STUDY/WEBINAR/DEMO/PROPOSAL | -| content_id | VARCHAR | 64 | | 内容ID（FK） | CONTENT_001 | +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| touchpoint_id | VARCHAR | 64 | PK, NOT NULL | 触点唯一标识 | TP_20231105001 | +| touchpoint_type | VARCHAR | 50 | NOT NULL | 触点类型 | PAGE_VIEW / FORM_SUBMIT / DOWNLOAD / EMAIL / CALL / MEETING / CHAT | +| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | +| campaign_id | VARCHAR | 64 | FK | 营销活动ID | CMP_001 | +| contact_id | VARCHAR | 64 | FK | 联系人ID | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | 线索ID | LEAD_20231105001 | +| account_id | VARCHAR | 64 | FK | 企业ID | ACC_20231105001 | +| touchpoint_time | DATETIME | | NOT NULL | 触点发生时间 | 2023-11-05 14:30:00 | +| touchpoint_direction | VARCHAR | 20 | | 触点方向 | INBOUND / OUTBOUND | +| touchpoint_status | VARCHAR | 50 | | 触点状态 | COMPLETED / SCHEDULED / CANCELLED | +| content_type | VARCHAR | 50 | | 内容类型 | WHITEPAPER / CASE_STUDY / WEBINAR / DEMO / PROPOSAL | +| content_id | VARCHAR | 64 | FK | 内容ID | CONTENT_001 | | subject | VARCHAR | 200 | | 主题 | 产品演示会议 | -| description | TEXT | | | 描述 | 讨论了AI平台的技术架构... | -| duration_seconds | INT | | | 持续时长（秒） | 3600 | -| owner_user_id | VARCHAR | 64 | | 负责人ID | USER_001 | -| metadata | JSON | | | 元数据 | {"device": "mobile", "location": "Beijing"} | -| utm_params | JSON | | | UTM参数 | {"utm_source": "baidu", "utm_medium": "cpc"} | -| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 14:30:00 | +| description | TEXT | | | 描述详情 | 讨论了AI平台的技术架构和实施方案 | +| duration_seconds | INT | | | 持续时长(秒) | 3600 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| metadata | JSON | | | 元数据 | {"device": "mobile", "location": "Beijing", "ip": "192.168.1.1"} | +| utm_params | JSON | | | UTM追踪参数 | {"utm_source": "baidu", "utm_medium": "cpc", "utm_campaign": "Q4"} | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 14:30:00 | --- ### Event 详细字段设计 -| 字段名 | 类型 | 长度 | 必填 | 说明 | 示例 | -|--------|------|------|------|------|------| -| event_id | VARCHAR | 64 | ✓ | 事件唯一ID（PK） | EVT_20231105001 | -| event_name | VARCHAR | 100 | ✓ | 事件名称 | page_view | -| event_type | VARCHAR | 50 | ✓ | 事件类型 | PAGE_VIEW/CLICK/FORM_START/FORM_SUBMIT/DOWNLOAD/VIDEO_PLAY/SEARCH | -| channel_id | VARCHAR | 64 | | 渠道ID（FK） | CH_001 | -| contact_id | VARCHAR | 64 | | 联系人ID（FK） | CNT_20231105001 | -| lead_id | VARCHAR | 64 | | 线索ID（FK） | LEAD_20231105001 | -| account_id | VARCHAR | 64 | | 企业ID（FK） | ACC_20231105001 | -| event_time | DATETIME | | ✓ | 事件时间 | 2023-11-05 14:35:20 | +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| event_id | VARCHAR | 64 | PK, NOT NULL | 事件唯一标识 | EVT_20231105001 | +| event_name | VARCHAR | 100 | NOT NULL | 事件名称 | page_view | +| event_type | VARCHAR | 50 | NOT NULL | 事件类型 | PAGE_VIEW / CLICK / FORM_START / FORM_SUBMIT / DOWNLOAD / VIDEO_PLAY | +| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | +| contact_id | VARCHAR | 64 | FK | 联系人ID | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | 线索ID | LEAD_20231105001 | +| account_id | VARCHAR | 64 | FK | 企业ID | ACC_20231105001 | +| event_time | DATETIME | | NOT NULL | 事件发生时间 | 2023-11-05 14:35:20 | | session_id | VARCHAR | 64 | | 会话ID | SESSION_20231105001 | -| device_type | VARCHAR | 50 | | 设备类型 | DESKTOP/MOBILE/TABLET | -| browser | VARCHAR | 50 | | 浏览器 | Chrome | -| os | VARCHAR | 50 | | 操作系统 | Windows 10 | +| device_type | VARCHAR | 50 | | 设备类型 | DESKTOP / MOBILE / TABLET | +| browser | VARCHAR | 50 | | 浏览器 | Chrome / Safari / Firefox / Edge | +| os | VARCHAR | 50 | | 操作系统 | Windows 10 / iOS 16 / Android 13 | | ip_address | VARCHAR | 50 | | IP地址 | 192.168.1.1 | | page_url | VARCHAR | 1000 | | 页面URL | https://www.example.com/product/ai-platform | | referrer_url | VARCHAR | 1000 | | 来源URL | https://www.baidu.com/s?wd=AI平台 | -| event_properties | JSON | | | 事件属性 | {"button_text": "申请试用", "form_id": "trial_form"} | -| utm_params | JSON | | | UTM参数 | {"utm_source": "baidu", "utm_campaign": "Q4_campaign"} | -| created_at | DATETIME | | ✓ | 创建时间 | 2023-11-05 14:35:20 | +| event_properties | JSON | | | 事件属性 | {"button_text": "申请试用", "form_id": "trial_form", "product": "AI平台"} | +| utm_params | JSON | | | UTM追踪参数 | {"utm_source": "baidu", "utm_campaign": "Q4_campaign", "utm_content": "ad1"} | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 14:35:20 | --- ## 全渠道身份映射方案 -### 身份识别优先级 +### 身份识别匹配规则 ```mermaid -graph LR - A[多渠道数据] --> B{身份识别} - B --> C[1. 手机号匹配] - B --> D[2. 邮箱匹配] - B --> E[3. 企业微信ID] - B --> F[4. 统一社会信用代码] - B --> G[5. 自定义ID] - - C --> H[合并至统一Contact] - D --> H - E --> H - F --> I[合并至统一Account] - G --> H - - H --> J[建立ChannelIdentity关联] - I --> K[建立AccountChannelIdentity关联] +flowchart TD + Start([接收多渠道数据]) --> ExtractID[提取身份标识符] + + ExtractID --> CheckPhone{是否有
手机号?} + ExtractID --> CheckEmail{是否有
邮箱?} + ExtractID --> CheckWechat{是否有
微信ID?} + ExtractID --> CheckCorpID{是否有
企业信用代码?} + + CheckPhone -->|是| MatchPhone[手机号匹配
优先级: 1] + CheckEmail -->|是| MatchEmail[邮箱匹配
优先级: 2] + CheckWechat -->|是| MatchWechat[微信ID匹配
优先级: 3] + CheckCorpID -->|是| MatchCorpID[企业信用代码匹配
优先级: 1] + + MatchPhone --> FoundContact{找到匹配
Contact?} + MatchEmail --> FoundContact + MatchWechat --> FoundContact + + MatchCorpID --> FoundAccount{找到匹配
Account?} + + FoundContact -->|是| MergeContact[合并到现有Contact] + FoundContact -->|否| CreateContact[创建新Contact] + + FoundAccount -->|是| MergeAccount[合并到现有Account] + FoundAccount -->|否| CreateAccount[创建新Account] + + MergeContact --> AddChannelIdentity[添加/更新
ContactChannelIdentity] + CreateContact --> AddChannelIdentity + + MergeAccount --> AddAccountIdentity[添加/更新
AccountChannelIdentity] + CreateAccount --> AddAccountIdentity + + AddChannelIdentity --> UpdateSummary[更新ContactSummary] + AddAccountIdentity --> UpdateAccountSummary[更新AccountSummary] + + UpdateSummary --> End([完成]) + UpdateAccountSummary --> End + + style Start fill:#e1f5ff + style End fill:#e8f5e9 + style MergeContact fill:#fff9c4 + style MergeAccount fill:#fff9c4 + style CreateContact fill:#ffe0b2 + style CreateAccount fill:#ffe0b2 ``` -### 身份合并规则 +### 身份聚合优先级 ```mermaid -flowchart TD - Start[接收新数据] --> Check{是否存在标识符} - Check -->|有手机号| Phone[手机号匹配] - Check -->|有邮箱| Email[邮箱匹配] - Check -->|有微信ID| WeChat[微信ID匹配] - - Phone --> Match{找到匹配?} - Email --> Match - WeChat --> Match +graph LR + subgraph Priority["身份匹配优先级 Identity Matching Priority"] + direction TB + P1["优先级1: 手机号
Priority 1: Mobile Phone
唯一性: 高
准确性: 95%"] + P2["优先级2: 邮箱
Priority 2: Email
唯一性: 高
准确性: 90%"] + P3["优先级3: 企业微信ID
Priority 3: Enterprise WeChat
唯一性: 中
准确性: 85%"] + P4["优先级4: 统一社会信用代码
Priority 4: Unified Credit Code
唯一性: 极高
准确性: 99%"] + P5["优先级5: 自定义ID
Priority 5: Custom ID
唯一性: 中
准确性: 70%"] + end - Match -->|是| Merge[合并到现有Contact] - Match -->|否| Create[创建新Contact] + subgraph Result["匹配结果 Matching Result"] + direction TB + R1["Contact统一身份
Unified Contact"] + R2["Account统一身份
Unified Account"] + end - Merge --> AddIdentity[添加渠道身份] - Create --> AddIdentity + P1 & P2 & P3 & P5 --> R1 + P4 --> R2 - AddIdentity --> UpdateScore[更新评分] - UpdateScore --> End[完成] + style P1 fill:#c8e6c9 + style P2 fill:#c8e6c9 + style P4 fill:#c8e6c9 + style P3 fill:#fff9c4 + style P5 fill:#ffccbc + style R1 fill:#e1f5ff + style R2 fill:#e1f5ff ``` --- ## 数据字典总结 -### 核心实体数量统计 - -| 实体类型 | 实体数量 | 说明 | -|---------|---------|------| -| 客户主体实体 | 3 | Account, Contact, Lead | -| 业务实体 | 2 | Opportunity, Product | -| 营销实体 | 2 | Campaign, Channel | -| 交互实体 | 2 | Touchpoint, Event | -| 关系实体 | 6 | AccountContactRelation, AccountRelation, OpportunityProduct, CampaignMember, TagRelation, SegmentMember | -| 身份实体 | 3 | AccountChannelIdentity, ContactChannelIdentity, LeadChannelIdentity | -| 分析实体 | 5 | Segment, Tag, Score, Attribution, CustomerJourney | -| 支撑实体 | 2 | Industry, ProductCategory | -| **合计** | **25** | 覆盖B2B CDP核心业务场景 | +### 核心实体统计 + +| 实体分类 | 实体名称 | 数量 | 说明 | +|---------|---------|------|------| +| 客户主体实体 | Account, Contact, Lead | 3 | 核心客户数据 | +| 汇总数据实体 | AccountSummary, ContactSummary, LeadSummary | 3 | 统计汇总数据 | +| 业务实体 | Opportunity, Product, ProductCategory | 3 | 业务交易数据 | +| 营销实体 | Campaign, Channel | 2 | 营销活动管理 | +| 交互实体 | Touchpoint, Event | 2 | 客户互动行为 | +| 关系实体 | AccountContactRelation, AccountRelation, OpportunityProduct, CampaignMember, TagRelation, SegmentMember | 6 | 实体关系映射 | +| 身份实体 | AccountChannelIdentity, ContactChannelIdentity, LeadChannelIdentity | 3 | 全渠道身份映射 | +| 分析实体 | Segment, Tag, ScoreModel, ScoreRecord, ScoreHistory, Attribution, TouchpointAttribution, CustomerJourney, JourneyStage | 9 | 数据分析洞察 | +| 性能实体 | ChannelPerformance, CampaignPerformance | 2 | 绩效统计 | +| 历史实体 | OpportunityStageHistory | 1 | 变更历史追踪 | +| 支撑实体 | Industry, TouchpointAttachment | 2 | 基础支撑数据 | +| **合计** | | **36** | 完整覆盖B2B CDP业务场景 | --- -## 扩展建议 +## 汇总数据表设计说明 -### 可选扩展实体（根据业务需要） +### 为什么需要汇总数据表 -1. **Content（内容资产）** - - 营销内容管理 - - 内容效果追踪 +1. **性能优化** + - 避免频繁的多表JOIN和聚合计算 + - 提升客户360度视图的查询速度 + - 降低数据库负载 -2. **Order（订单）** - - 如需管理订单详情 - - 支持电商场景 +2. **业务需求** + - 快速展示客户健康度评分 + - 实时显示客户价值指标 + - 支持客户预警和监控 -3. **Contract（合同）** - - 合同管理 - - 续约提醒 +3. **数据一致性** + - 统一的计算口径 + - 定时批量更新保证数据准确性 + - 避免实时计算的结果不一致 -4. **Partner（合作伙伴）** - - 渠道伙伴管理 - - 分销体系 - -5. **Competitor（竞争对手）** - - 竞争对手分析 - - 竞品情报 - -6. **Task（任务）** - - 销售任务管理 - - 跟进提醒 - -7. **Note（备注）** - - 客户备注 - - 沟通记录 - ---- - -## 技术实现建议 - -### 数据库选型建议 +### 汇总数据更新策略 ```mermaid -graph TB - subgraph "主数据存储" - PG[PostgreSQL
关系型数据
Account/Contact/Lead/Opportunity] +flowchart LR + subgraph Trigger["触发更新 Update Triggers"] + T1[定时任务
Scheduled Job
每小时] + T2[实时触发
Real-time Trigger
关键事件] + T3[手动刷新
Manual Refresh
按需执行] end - subgraph "行为数据存储" - CH[ClickHouse
海量事件数据
Event/Touchpoint] + subgraph Calculation["计算引擎 Calculation Engine"] + C1[聚合计算
Aggregation] + C2[评分计算
Scoring] + C3[规则引擎
Rule Engine] end - subgraph "搜索引擎" - ES[Elasticsearch
全文搜索
客户搜索/标签搜索] + subgraph Summary["汇总表 Summary Tables"] + S1[AccountSummary
企业汇总] + S2[ContactSummary
联系人汇总] + S3[LeadSummary
线索汇总] end - subgraph "缓存层" - Redis[Redis
热数据缓存
评分/标签] + T1 --> C1 + T2 --> C2 + T3 --> C3 + + C1 & C2 & C3 --> S1 & S2 & S3 + + style T1 fill:#e3f2fd + style T2 fill:#fff3e0 + style T3 fill:#f3e5f5 + style S1 fill:#e8f5e9 + style S2 fill:#e8f5e9 + style S3 fill:#e8f5e9 +``` + +--- + +## 技术实现建议 + +### 数据库架构 + +```mermaid +flowchart TB + subgraph AppLayer["应用层 Application Layer"] + API[API服务
API Service] + Worker[后台任务
Background Worker] end - subgraph "数据仓库" - DW[数据仓库
分析报表
BI分析] + subgraph DataLayer["数据层 Data Layer"] + direction TB + + subgraph OLTP["OLTP 在线事务处理"] + PG[(PostgreSQL
主数据存储
Account/Contact/Lead
Opportunity)] + end + + subgraph BigData["大数据存储 Big Data Storage"] + CH[(ClickHouse
行为数据
Event/Touchpoint
10亿+记录)] + end + + subgraph Cache["缓存层 Cache Layer"] + Redis[(Redis
热数据缓存
Summary/Score/Tag)] + end + + subgraph Search["搜索引擎 Search Engine"] + ES[(Elasticsearch
全文检索
客户搜索/内容搜索)] + end + + subgraph Analytics["分析层 Analytics Layer"] + DW[(数据仓库
Data Warehouse
BI报表/数据分析)] + end end - PG --> ES - PG --> Redis - CH --> DW - PG --> DW + API --> PG + API --> Redis + API --> ES + + Worker --> PG + Worker --> CH + Worker --> Redis + + PG -.同步.-> ES + PG -.同步.-> Redis + PG -.ETL.-> DW + CH -.ETL.-> DW + + classDef oltp fill:#e8f5e9,stroke:#388e3c,stroke-width:2px + classDef bigdata fill:#fff3e0,stroke:#f57c00,stroke-width:2px + classDef cache fill:#fce4ec,stroke:#c2185b,stroke-width:2px + classDef search fill:#e1f5ff,stroke:#0277bd,stroke-width:2px + classDef analytics fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px + + class PG oltp + class CH bigdata + class Redis cache + class ES search + class DW analytics ``` -### 关键索引建议 +### 索引设计建议 -**Account表索引：** +**Account表索引:** ```sql -- 主键索引 -PRIMARY KEY (account_id) +PRIMARY KEY (account_id); --- 业务索引 +-- 唯一索引 +CREATE UNIQUE INDEX uk_account_credit_code ON Account(unified_social_credit_code) WHERE unified_social_credit_code IS NOT NULL; + +-- 业务查询索引 CREATE INDEX idx_account_name ON Account(account_name); -CREATE INDEX idx_unified_code ON Account(unified_social_credit_code); CREATE INDEX idx_account_status ON Account(account_status); CREATE INDEX idx_account_owner ON Account(owner_user_id); -CREATE INDEX idx_account_created ON Account(created_at); +CREATE INDEX idx_account_created ON Account(created_at DESC); +CREATE INDEX idx_account_industry ON Account(industry_id); -- 组合索引 CREATE INDEX idx_account_type_status ON Account(account_type, account_status); -CREATE INDEX idx_account_city ON Account(province, city); +CREATE INDEX idx_account_location ON Account(province, city); +CREATE INDEX idx_account_lifecycle ON Account(lifecycle_stage, account_status); +``` + +**AccountSummary表索引:** +```sql +-- 主键和唯一索引 +PRIMARY KEY (summary_id); +CREATE UNIQUE INDEX uk_summary_account ON AccountSummary(account_id); + +-- 查询索引 +CREATE INDEX idx_summary_health_score ON AccountSummary(health_score DESC); +CREATE INDEX idx_summary_last_activity ON AccountSummary(last_activity_at DESC); +CREATE INDEX idx_summary_total_revenue ON AccountSummary(total_revenue DESC); +CREATE INDEX idx_summary_win_rate ON AccountSummary(win_rate DESC); ``` -**Contact表索引：** +**Contact表索引:** ```sql -- 主键索引 -PRIMARY KEY (contact_id) +PRIMARY KEY (contact_id); --- 业务索引 -CREATE INDEX idx_contact_phone ON Contact(mobile_phone); -CREATE INDEX idx_contact_email ON Contact(email); -CREATE INDEX idx_contact_wechat ON Contact(wechat_id); +-- 唯一索引 +CREATE UNIQUE INDEX uk_contact_phone ON Contact(mobile_phone) WHERE mobile_phone IS NOT NULL; +CREATE UNIQUE INDEX uk_contact_email ON Contact(email) WHERE email IS NOT NULL; + +-- 业务查询索引 +CREATE INDEX idx_contact_name ON Contact(contact_name); CREATE INDEX idx_contact_account ON Contact(primary_account_id); CREATE INDEX idx_contact_status ON Contact(contact_status); +CREATE INDEX idx_contact_wechat ON Contact(wechat_id); -- 组合索引 CREATE INDEX idx_contact_phone_email ON Contact(mobile_phone, email); +CREATE INDEX idx_contact_lifecycle ON Contact(lifecycle_stage, contact_status); ``` -**Event表索引（ClickHouse）：** +**Event表索引 (ClickHouse):** ```sql --- 主排序键 -ORDER BY (event_time, contact_id, event_type) +-- 排序键设计 +ORDER BY (channel_id, event_time, contact_id, event_type); --- 分区键 -PARTITION BY toYYYYMM(event_time) +-- 分区键设计 +PARTITION BY toYYYYMM(event_time); + +-- 采样表达式 +SAMPLE BY cityHash64(event_id); +``` + +**Touchpoint表索引:** +```sql +-- 主键索引 +PRIMARY KEY (touchpoint_id); + +-- 查询索引 +CREATE INDEX idx_touchpoint_contact ON Touchpoint(contact_id, touchpoint_time DESC); +CREATE INDEX idx_touchpoint_lead ON Touchpoint(lead_id, touchpoint_time DESC); +CREATE INDEX idx_touchpoint_account ON Touchpoint(account_id, touchpoint_time DESC); +CREATE INDEX idx_touchpoint_channel ON Touchpoint(channel_id, touchpoint_time DESC); +CREATE INDEX idx_touchpoint_campaign ON Touchpoint(campaign_id, touchpoint_time DESC); +CREATE INDEX idx_touchpoint_time ON Touchpoint(touchpoint_time DESC); +CREATE INDEX idx_touchpoint_type ON Touchpoint(touchpoint_type, touchpoint_time DESC); ``` --- ## 总结 -本实体设计方案包含： +本B2B CDP实体设计方案提供: + +**核心能力** +- 36个实体，完整覆盖B2B CDP业务场景 +- 全渠道身份映射方案，支持跨渠道客户识别 +- 汇总数据表设计，优化查询性能 +- 完整的客户生命周期管理 +- 从线索到商机的完整转化流程 -✅ **25个核心实体**，覆盖B2B CDP全业务场景 -✅ **全渠道身份映射**方案，支持跨渠道客户识别 -✅ **完整的客户生命周期**管理 -✅ **从线索到商机**的完整转化流程 -✅ **灵活的标签和分群**能力 -✅ **多维度的归因分析**能力 -✅ **详细的字段设计**和数据字典 +**关键特性** +- 灵活的标签和分群能力 +- 多维度的归因分析能力 +- 详细的字段设计和数据字典 +- 专业的数据库架构建议 +- 完整的索引优化方案 -该方案可以支撑： +**适用场景** - 全渠道客户数据整合 -- 客户360度画像 +- 客户360度画像分析 - 精准营销和客户分群 - 销售线索管理和转化 - 客户旅程分析 - 营销归因分析 - 客户价值评估 +- 客户健康度监控 根据实际业务需要，可以选择性实现部分实体，并在后续迭代中逐步完善。 From 2d19a0d992c4abdaa5895dff3dcd3a5ab5db74c5 Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 10:56:25 +0000 Subject: [PATCH 03/10] Refactor: Update B2B CDP entity design documentation Co-authored-by: 13585811473 <13585811473@163.com> --- docs/b2b-cdp-entity-design.md | 1575 +++++++++++++++------------------ 1 file changed, 698 insertions(+), 877 deletions(-) diff --git a/docs/b2b-cdp-entity-design.md b/docs/b2b-cdp-entity-design.md index db74c2aa4..70a16cb8d 100644 --- a/docs/b2b-cdp-entity-design.md +++ b/docs/b2b-cdp-entity-design.md @@ -14,236 +14,256 @@ ### 实体分层架构图 ```mermaid -flowchart TD - subgraph DataSource["数据源层 Data Source Layer"] - direction LR - DS1["官方网站
Official Website"] - DS2["微信生态
WeChat Ecosystem"] - DS3["抖音平台
Douyin Platform"] - DS4["线下活动
Offline Events"] - DS5["电话邮件
Phone & Email"] - DS6["CRM系统
CRM System"] - DS7["第三方平台
3rd Party Platform"] +graph TB + subgraph Layer1[Data Source Layer] + DS1[Official Website] + DS2[WeChat Ecosystem] + DS3[Social Media] + DS4[Offline Events] + DS5[Phone & Email] + DS6[CRM System] + DS7[3rd Party Platform] end - subgraph ChannelLayer["渠道层 Channel Layer"] - CH["Channel
渠道管理"] + subgraph Layer2[Channel Layer] + Channel[Channel Management] end - subgraph InteractionLayer["交互层 Interaction Layer"] - TP["Touchpoint
客户触点"] - EV["Event
行为事件"] - CM["Campaign
营销活动"] + subgraph Layer3[Interaction Layer] + Touchpoint[Touchpoint] + Event[Event] + Campaign[Campaign] end - subgraph EntityLayer["客户实体层 Customer Entity Layer"] - CT["Contact
联系人"] - LD["Lead
线索"] - AC["Account
企业账户"] - AS["AccountSummary
账户汇总数据"] + subgraph Layer4[Customer Entity Layer] + Contact[Contact] + Lead[Lead] + Account[Account] end - subgraph BusinessLayer["业务层 Business Layer"] - OP["Opportunity
商机"] - PR["Product
产品"] - OR["Order
订单"] + subgraph Layer5[Summary Layer] + ContactSummary[ContactSummary] + LeadSummary[LeadSummary] + AccountSummary[AccountSummary] end - subgraph InsightLayer["洞察层 Insight Layer"] - SG["Segment
客户分群"] - TG["Tag
标签体系"] - SC["Score
评分模型"] - JN["Journey
客户旅程"] - AT["Attribution
营销归因"] + subgraph Layer6[Business Layer] + Opportunity[Opportunity] + Product[Product] end - DataSource --> CH - CH --> TP - CH --> EV - CM --> TP - TP --> CT - TP --> LD - EV --> CT - CT --> AC - AC --> AS - LD --> CT - LD --> OP - AC --> OP - OP --> PR - OP --> OR - CT --> SG - AC --> SG - LD --> TG - CT --> TG - AC --> TG - CT --> SC - AC --> SC - LD --> SC - CT --> JN - CM --> AT - OP --> AT - - classDef sourceStyle fill:#e1f5ff,stroke:#01579b,stroke-width:2px - classDef channelStyle fill:#fff3e0,stroke:#e65100,stroke-width:2px - classDef interactionStyle fill:#f3e5f5,stroke:#4a148c,stroke-width:2px - classDef entityStyle fill:#e8f5e9,stroke:#1b5e20,stroke-width:2px - classDef businessStyle fill:#fff9c4,stroke:#f57f17,stroke-width:2px - classDef insightStyle fill:#fce4ec,stroke:#880e4f,stroke-width:2px - - class DS1,DS2,DS3,DS4,DS5,DS6,DS7 sourceStyle - class CH channelStyle - class TP,EV,CM interactionStyle - class CT,LD,AC,AS entityStyle - class OP,PR,OR businessStyle - class SG,TG,SC,JN,AT insightStyle + subgraph Layer7[Insight Layer] + Segment[Segment] + Tag[Tag] + Score[Score Model] + Journey[Customer Journey] + Attribution[Attribution] + end + + DS1 --> Channel + DS2 --> Channel + DS3 --> Channel + DS4 --> Channel + DS5 --> Channel + DS6 --> Channel + DS7 --> Channel + + Channel --> Touchpoint + Channel --> Event + Campaign --> Touchpoint + + Touchpoint --> Contact + Touchpoint --> Lead + Event --> Contact + Event --> Lead + + Contact --> Account + Lead --> Contact + + Contact --> ContactSummary + Lead --> LeadSummary + Account --> AccountSummary + + Lead --> Opportunity + Account --> Opportunity + Opportunity --> Product + + Contact --> Segment + Account --> Segment + Contact --> Tag + Account --> Tag + Contact --> Score + Account --> Score + Lead --> Score + Contact --> Journey + Campaign --> Attribution + Opportunity --> Attribution +``` + +--- + +### 数据库架构设计 + +```mermaid +graph LR + subgraph ApplicationLayer[Application Layer] + API[API Gateway] + Worker[Background Jobs] + end + + subgraph DatabaseLayer[Database Layer] + PostgreSQL[(PostgreSQL
OLTP Database
Account/Contact/Lead
Opportunity)] + ClickHouse[(ClickHouse
OLAP Database
Event/Touchpoint)] + Redis[(Redis
Cache Layer
Summary/Score)] + end + + subgraph AnalyticsLayer[Analytics Layer] + Elasticsearch[(Elasticsearch
Search Engine)] + DataWarehouse[(Data Warehouse
BI & Analytics)] + end + + API --> PostgreSQL + API --> Redis + API --> Elasticsearch + Worker --> PostgreSQL + Worker --> ClickHouse + Worker --> Redis + PostgreSQL --> Elasticsearch + PostgreSQL --> DataWarehouse + ClickHouse --> DataWarehouse ``` --- ## 核心实体设计 -### 1. Account (企业账户) - 核心实体 +### 1. Account Entity ```mermaid erDiagram - Account ||--o{ AccountChannelIdentity : "has" - Account ||--|| AccountSummary : "aggregates to" - Account ||--o{ AccountRelation : "parent of" - Account ||--o{ AccountRelation : "child of" + Account ||--o{ AccountChannelIdentity : has + Account ||--|| AccountSummary : aggregates + Account ||--o{ AccountRelation : parent + Account ||--o{ AccountRelation : child Account { - varchar(64) account_id PK - varchar(200) account_name - varchar(18) unified_social_credit_code UK - varchar(50) account_type - varchar(64) industry_id FK - varchar(50) account_status - varchar(50) account_level - decimal(18_2) annual_revenue + varchar account_id PK + varchar account_name + varchar unified_social_credit_code UK + varchar account_type + varchar industry_id FK + varchar account_status + varchar account_level + decimal annual_revenue int employee_count - varchar(500) company_website - varchar(500) company_address - varchar(50) province - varchar(50) city - varchar(50) district - varchar(100) account_source - varchar(64) primary_channel_id FK - varchar(64) owner_user_id FK + varchar company_website + varchar province + varchar city + varchar account_source + varchar primary_channel_id FK + varchar owner_user_id FK datetime created_at datetime updated_at - json custom_fields - varchar(50) lifecycle_stage + varchar lifecycle_stage } AccountChannelIdentity { - varchar(64) identity_id PK - varchar(64) account_id FK - varchar(64) channel_id FK - varchar(200) channel_account_id - varchar(50) identity_type + varchar identity_id PK + varchar account_id FK + varchar channel_id FK + varchar channel_account_id + varchar identity_type boolean is_verified datetime first_seen_at datetime last_seen_at - json additional_info } AccountSummary { - varchar(64) summary_id PK - varchar(64) account_id FK UK + varchar summary_id PK + varchar account_id FK UK int total_contacts int total_opportunities int total_leads - decimal(18_2) total_revenue - decimal(18_2) lifetime_value + decimal total_revenue + decimal lifetime_value int won_opportunities int lost_opportunities - decimal(5_2) win_rate + decimal win_rate int total_touchpoints - int active_campaigns int health_score datetime last_activity_at - datetime last_purchase_at - date first_purchase_date - date latest_opportunity_date - int days_since_last_contact datetime calculated_at - datetime updated_at } AccountRelation { - varchar(64) relation_id PK - varchar(64) parent_account_id FK - varchar(64) child_account_id FK - varchar(50) relation_type + varchar relation_id PK + varchar parent_account_id FK + varchar child_account_id FK + varchar relation_type datetime created_at } ``` -**Account 字段说明:** -- `account_id`: 全局唯一账户ID +**Account 核心字段说明:** +- `account_id`: 全局唯一账户标识 - `unified_social_credit_code`: 统一社会信用代码（企业唯一标识） -- `account_type`: 客户类型（CUSTOMER/PARTNER/COMPETITOR/PROSPECT） -- `account_status`: 账户状态（ACTIVE/DORMANT/CHURNED/BLACKLIST） -- `account_level`: 客户等级（STRATEGIC/IMPORTANT/NORMAL） -- `lifecycle_stage`: 生命周期阶段（AWARENESS/CONSIDERATION/DECISION/RETENTION/EXPANSION） - -**AccountSummary 字段说明:** -- `total_contacts`: 关联的联系人总数 -- `total_opportunities`: 商机总数 -- `total_revenue`: 累计收入 +- `account_type`: 客户类型（CUSTOMER, PARTNER, COMPETITOR, PROSPECT） +- `account_status`: 账户状态（ACTIVE, DORMANT, CHURNED, BLACKLIST） +- `account_level`: 客户等级（STRATEGIC, IMPORTANT, NORMAL） +- `lifecycle_stage`: 生命周期阶段（AWARENESS, CONSIDERATION, DECISION, RETENTION, EXPANSION） + +**AccountSummary 核心指标:** +- `total_contacts`: 关联联系人总数 +- `total_revenue`: 累计收入金额 - `lifetime_value`: 客户生命周期价值 -- `win_rate`: 赢单率 +- `win_rate`: 赢单率（百分比） - `health_score`: 健康度评分（0-100） -- `last_activity_at`: 最后活跃时间 -- `days_since_last_contact`: 距离上次联系天数 --- -### 2. Contact (联系人) - 核心实体 +### 2. Contact Entity ```mermaid erDiagram - Contact ||--o{ ContactChannelIdentity : "has" - Contact ||--o{ AccountContactRelation : "belongs to" - Contact ||--|| ContactSummary : "aggregates to" + Contact ||--o{ ContactChannelIdentity : has + Contact ||--o{ AccountContactRelation : belongs_to + Contact ||--|| ContactSummary : aggregates Contact { - varchar(64) contact_id PK - varchar(100) contact_name - varchar(20) mobile_phone UK - varchar(200) email UK - varchar(100) wechat_id - varchar(100) job_title - varchar(100) department - varchar(50) contact_status - varchar(64) primary_account_id FK - varchar(100) contact_source - varchar(64) primary_channel_id FK - varchar(64) owner_user_id FK + varchar contact_id PK + varchar contact_name + varchar mobile_phone UK + varchar email UK + varchar wechat_id + varchar job_title + varchar department + varchar contact_status + varchar primary_account_id FK + varchar contact_source + varchar primary_channel_id FK + varchar owner_user_id FK datetime created_at datetime updated_at - json custom_fields - varchar(50) lifecycle_stage + varchar lifecycle_stage boolean is_decision_maker boolean is_verified } ContactChannelIdentity { - varchar(64) identity_id PK - varchar(64) contact_id FK - varchar(64) channel_id FK - varchar(200) channel_user_id - varchar(50) identity_type + varchar identity_id PK + varchar contact_id FK + varchar channel_id FK + varchar channel_user_id + varchar identity_type boolean is_primary boolean is_verified datetime first_seen_at datetime last_seen_at - json additional_info } ContactSummary { - varchar(64) summary_id PK - varchar(64) contact_id FK UK + varchar summary_id PK + varchar contact_id FK UK int total_touchpoints int total_events int email_opens @@ -252,91 +272,80 @@ erDiagram int content_downloads int engagement_score datetime last_activity_at - datetime last_email_sent_at - datetime last_email_opened_at int days_since_last_activity - int campaign_responses datetime calculated_at - datetime updated_at } AccountContactRelation { - varchar(64) relation_id PK - varchar(64) account_id FK - varchar(64) contact_id FK - varchar(100) role_in_account - varchar(50) decision_level + varchar relation_id PK + varchar account_id FK + varchar contact_id FK + varchar role_in_account + varchar decision_level boolean is_primary_contact - varchar(50) relationship_status + varchar relationship_status date relation_start_date - date relation_end_date datetime created_at - datetime updated_at } ``` -**Contact 字段说明:** -- `decision_level`: 决策层级（DECISION_MAKER/INFLUENCER/USER/GATEKEEPER） -- `lifecycle_stage`: 生命周期阶段（SUBSCRIBER/LEAD/MQL/SQL/OPPORTUNITY/CUSTOMER） +**Contact 核心字段说明:** +- `decision_level`: 决策层级（DECISION_MAKER, INFLUENCER, USER, GATEKEEPER） +- `lifecycle_stage`: 生命周期阶段（SUBSCRIBER, LEAD, MQL, SQL, OPPORTUNITY, CUSTOMER） -**ContactSummary 字段说明:** -- `total_touchpoints`: 总触点数 +**ContactSummary 核心指标:** - `engagement_score`: 参与度评分（0-100） - `email_opens`: 邮件打开次数 -- `email_clicks`: 邮件点击次数 -- `days_since_last_activity`: 距离上次活跃天数 +- `days_since_last_activity`: 距上次活跃天数 --- -### 3. Lead (线索) - 核心实体 +### 3. Lead Entity ```mermaid erDiagram - Lead ||--o{ LeadChannelIdentity : "has" - Lead ||--|| LeadSummary : "aggregates to" + Lead ||--o{ LeadChannelIdentity : has + Lead ||--|| LeadSummary : aggregates Lead { - varchar(64) lead_id PK - varchar(100) lead_name - varchar(200) company_name - varchar(20) mobile_phone - varchar(200) email - varchar(100) wechat_id - varchar(100) job_title - varchar(100) lead_source - varchar(64) channel_id FK - varchar(64) campaign_id FK - varchar(50) lead_status + varchar lead_id PK + varchar lead_name + varchar company_name + varchar mobile_phone + varchar email + varchar wechat_id + varchar job_title + varchar lead_source + varchar channel_id FK + varchar campaign_id FK + varchar lead_status int lead_score - varchar(10) lead_grade - varchar(64) industry_id FK - varchar(50) province - varchar(50) city - varchar(64) owner_user_id FK + varchar lead_grade + varchar industry_id FK + varchar province + varchar city + varchar owner_user_id FK datetime created_at datetime updated_at datetime last_contacted_at datetime converted_at - varchar(64) converted_contact_id FK - varchar(64) converted_account_id FK - varchar(64) converted_opportunity_id FK - json custom_fields + varchar converted_contact_id FK + varchar converted_account_id FK + varchar converted_opportunity_id FK boolean is_qualified } LeadChannelIdentity { - varchar(64) identity_id PK - varchar(64) lead_id FK - varchar(64) channel_id FK - varchar(200) channel_user_id + varchar identity_id PK + varchar lead_id FK + varchar channel_id FK + varchar channel_user_id datetime captured_at - json utm_params - json additional_info } LeadSummary { - varchar(64) summary_id PK - varchar(64) lead_id FK UK + varchar summary_id PK + varchar lead_id FK UK int total_touchpoints int total_events int form_submissions @@ -345,183 +354,166 @@ erDiagram int days_in_pipeline int contact_attempts datetime last_activity_at - datetime last_contact_attempt_at datetime calculated_at - datetime updated_at } ``` -**Lead 字段说明:** -- `lead_status`: 线索状态（NEW/CONTACTED/QUALIFIED/CONVERTED/DISQUALIFIED） +**Lead 核心字段说明:** +- `lead_status`: 线索状态（NEW, CONTACTED, QUALIFIED, CONVERTED, DISQUALIFIED） - `lead_score`: 线索评分（0-100） -- `lead_grade`: 线索等级（A/B/C/D） +- `lead_grade`: 线索等级（A, B, C, D） - `is_qualified`: 是否为合格线索（MQL/SQL） --- -### 4. Opportunity (商机) - 核心实体 +### 4. Opportunity Entity ```mermaid erDiagram - Opportunity ||--o{ OpportunityStageHistory : "has" - Opportunity ||--o{ OpportunityProduct : "contains" + Opportunity ||--o{ OpportunityStageHistory : tracks + Opportunity ||--o{ OpportunityProduct : contains Opportunity { - varchar(64) opportunity_id PK - varchar(200) opportunity_name - varchar(64) account_id FK - varchar(64) primary_contact_id FK - varchar(64) lead_id FK - varchar(50) opportunity_type - varchar(100) opportunity_source - decimal(18_2) amount - varchar(10) currency - varchar(50) stage + varchar opportunity_id PK + varchar opportunity_name + varchar account_id FK + varchar primary_contact_id FK + varchar lead_id FK + varchar opportunity_type + varchar opportunity_source + decimal amount + varchar currency + varchar stage int probability date expected_close_date date actual_close_date - varchar(200) close_reason - varchar(64) owner_user_id FK - json product_ids - varchar(64) campaign_id FK + varchar close_reason + varchar owner_user_id FK + varchar campaign_id FK datetime created_at datetime updated_at - json custom_fields int days_in_stage boolean is_won boolean is_lost } OpportunityStageHistory { - varchar(64) history_id PK - varchar(64) opportunity_id FK - varchar(50) from_stage - varchar(50) to_stage + varchar history_id PK + varchar opportunity_id FK + varchar from_stage + varchar to_stage datetime changed_at - varchar(64) changed_by_user_id FK - varchar(200) change_reason + varchar changed_by_user_id FK + varchar change_reason int duration_days } OpportunityProduct { - varchar(64) opp_product_id PK - varchar(64) opportunity_id FK - varchar(64) product_id FK + varchar opp_product_id PK + varchar opportunity_id FK + varchar product_id FK int quantity - decimal(18_2) unit_price - decimal(18_2) total_price - decimal(18_2) discount - text product_description + decimal unit_price + decimal total_price + decimal discount } ``` -**Opportunity 字段说明:** -- `stage`: 阶段（LEAD/QUALIFICATION/NEEDS_ANALYSIS/PROPOSAL/NEGOTIATION/CONTRACT/CLOSED_WON/CLOSED_LOST） +**Opportunity 核心字段说明:** +- `stage`: 商机阶段（LEAD, QUALIFICATION, NEEDS_ANALYSIS, PROPOSAL, NEGOTIATION, CONTRACT, CLOSED_WON, CLOSED_LOST） - `probability`: 赢单概率（0-100） -- `opportunity_type`: 商机类型（NEW_BUSINESS/UPSELL/RENEWAL/CROSS_SELL） +- `opportunity_type`: 商机类型（NEW_BUSINESS, UPSELL, RENEWAL, CROSS_SELL） --- -### 5. Channel (渠道) - 核心实体 +### 5. Channel Entity ```mermaid erDiagram - Channel ||--o{ ChannelPerformance : "tracks" + Channel ||--o{ ChannelPerformance : tracks Channel { - varchar(64) channel_id PK - varchar(100) channel_name - varchar(50) channel_type - varchar(50) channel_category - varchar(64) parent_channel_id FK - varchar(50) channel_status - json channel_config - decimal(18_2) cost + varchar channel_id PK + varchar channel_name + varchar channel_type + varchar channel_category + varchar parent_channel_id FK + varchar channel_status + decimal cost datetime created_at datetime updated_at - json custom_fields } ChannelPerformance { - varchar(64) performance_id PK - varchar(64) channel_id FK + varchar performance_id PK + varchar channel_id FK date stat_date int lead_count int contact_count int opportunity_count - decimal(18_2) revenue - decimal(18_2) cost - decimal(18_2) roi + decimal revenue + decimal cost + decimal roi int conversion_count - decimal(5_2) conversion_rate + decimal conversion_rate datetime created_at } ``` **渠道类型 (channel_type):** -- WEBSITE - 官方网站 -- SEO - 搜索引擎优化 -- SEM - 搜索引擎营销 -- WECHAT - 微信 -- ENTERPRISE_WECHAT - 企业微信 -- DOUYIN - 抖音 -- EMAIL - 邮件营销 -- PHONE - 电话 -- OFFLINE_EVENT - 线下活动 -- EXHIBITION - 展会 -- PARTNER - 合作伙伴 +- WEBSITE, SEO, SEM +- WECHAT, ENTERPRISE_WECHAT, DOUYIN +- EMAIL, PHONE +- OFFLINE_EVENT, EXHIBITION +- PARTNER --- -### 6. Campaign (营销活动) - 核心实体 +### 6. Campaign Entity ```mermaid erDiagram - Campaign ||--o{ CampaignPerformance : "tracks" - Campaign ||--o{ CampaignMember : "has" + Campaign ||--o{ CampaignPerformance : tracks + Campaign ||--o{ CampaignMember : includes Campaign { - varchar(64) campaign_id PK - varchar(200) campaign_name - varchar(50) campaign_type - varchar(50) campaign_status - json channel_ids + varchar campaign_id PK + varchar campaign_name + varchar campaign_type + varchar campaign_status date start_date date end_date - decimal(18_2) budget - decimal(18_2) actual_cost - text target_audience - varchar(64) owner_user_id FK + decimal budget + decimal actual_cost + varchar owner_user_id FK datetime created_at datetime updated_at - json custom_fields } CampaignPerformance { - varchar(64) performance_id PK - varchar(64) campaign_id FK + varchar performance_id PK + varchar campaign_id FK date stat_date int impressions int clicks int leads_generated int opportunities_generated - decimal(18_2) revenue - decimal(18_2) roi - decimal(18_2) cpl - decimal(18_2) cpa + decimal revenue + decimal roi + decimal cpl + decimal cpa datetime created_at } CampaignMember { - varchar(64) member_id PK - varchar(64) campaign_id FK - varchar(50) member_type - varchar(64) member_ref_id FK - varchar(50) member_status + varchar member_id PK + varchar campaign_id FK + varchar member_type + varchar member_ref_id FK + varchar member_status datetime joined_at datetime responded_at - varchar(50) response_status - json response_data + varchar response_status } ``` @@ -532,127 +524,109 @@ erDiagram - EMAIL_MARKETING - 邮件营销 - CONTENT_MARKETING - 内容营销 - PRODUCT_TRIAL - 产品试用 -- SUMMIT - 行业峰会 --- -### 7. Touchpoint (触点) - 核心实体 +### 7. Touchpoint Entity ```mermaid erDiagram - Touchpoint ||--o{ TouchpointAttachment : "has" + Touchpoint ||--o{ TouchpointAttachment : has Touchpoint { - varchar(64) touchpoint_id PK - varchar(50) touchpoint_type - varchar(64) channel_id FK - varchar(64) campaign_id FK - varchar(64) contact_id FK - varchar(64) lead_id FK - varchar(64) account_id FK + varchar touchpoint_id PK + varchar touchpoint_type + varchar channel_id FK + varchar campaign_id FK + varchar contact_id FK + varchar lead_id FK + varchar account_id FK datetime touchpoint_time - varchar(20) touchpoint_direction - varchar(50) touchpoint_status - varchar(50) content_type - varchar(64) content_id FK - varchar(200) subject - text description + varchar touchpoint_direction + varchar touchpoint_status + varchar content_type + varchar content_id FK + varchar subject int duration_seconds - varchar(64) owner_user_id FK - json metadata - json utm_params + varchar owner_user_id FK datetime created_at } TouchpointAttachment { - varchar(64) attachment_id PK - varchar(64) touchpoint_id FK - varchar(200) file_name - varchar(500) file_url - varchar(50) file_type + varchar attachment_id PK + varchar touchpoint_id FK + varchar file_name + varchar file_url + varchar file_type int file_size datetime uploaded_at } ``` **触点类型 (touchpoint_type):** -- PAGE_VIEW - 网站浏览 -- FORM_SUBMIT - 表单提交 -- DOWNLOAD - 内容下载 -- EMAIL_OPEN - 邮件打开 -- EMAIL_CLICK - 邮件点击 -- CALL - 电话沟通 -- MEETING - 会议拜访 -- CHAT - 在线聊天 -- SOCIAL_INTERACTION - 社交媒体互动 +- PAGE_VIEW, FORM_SUBMIT, DOWNLOAD +- EMAIL_OPEN, EMAIL_CLICK +- CALL, MEETING, CHAT +- SOCIAL_INTERACTION --- -### 8. Event (行为事件) - 核心实体 +### 8. Event Entity ```mermaid erDiagram Event { - varchar(64) event_id PK - varchar(100) event_name - varchar(50) event_type - varchar(64) channel_id FK - varchar(64) contact_id FK - varchar(64) lead_id FK - varchar(64) account_id FK + varchar event_id PK + varchar event_name + varchar event_type + varchar channel_id FK + varchar contact_id FK + varchar lead_id FK + varchar account_id FK datetime event_time - varchar(64) session_id - varchar(50) device_type - varchar(50) browser - varchar(50) os - varchar(50) ip_address - varchar(1000) page_url - varchar(1000) referrer_url - json event_properties - json utm_params + varchar session_id + varchar device_type + varchar browser + varchar os + varchar ip_address + varchar page_url + varchar referrer_url datetime created_at } ``` **事件类型 (event_type):** -- PAGE_VIEW - 页面浏览 -- BUTTON_CLICK - 按钮点击 -- FORM_START - 表单开始 -- FORM_SUBMIT - 表单提交 -- FILE_DOWNLOAD - 文件下载 -- VIDEO_PLAY - 视频播放 -- PRODUCT_TRIAL - 产品试用 -- SEARCH - 搜索行为 +- PAGE_VIEW, BUTTON_CLICK +- FORM_START, FORM_SUBMIT +- FILE_DOWNLOAD, VIDEO_PLAY +- PRODUCT_TRIAL, SEARCH --- -### 9. Product (产品) - 核心实体 +### 9. Product Entity ```mermaid erDiagram - Product }o--|| ProductCategory : "belongs to" + Product }o--|| ProductCategory : belongs_to Product { - varchar(64) product_id PK - varchar(200) product_name - varchar(100) product_code UK - varchar(64) product_category_id FK - varchar(50) product_type - varchar(50) product_status - text description - decimal(18_2) list_price - varchar(10) currency - varchar(50) pricing_model - json feature_list + varchar product_id PK + varchar product_name + varchar product_code UK + varchar product_category_id FK + varchar product_type + varchar product_status + decimal list_price + varchar currency + varchar pricing_model datetime created_at datetime updated_at - json custom_fields } ProductCategory { - varchar(64) category_id PK - varchar(100) category_name - varchar(64) parent_category_id FK + varchar category_id PK + varchar category_name + varchar parent_category_id FK int level int sort_order datetime created_at @@ -661,32 +635,32 @@ erDiagram --- -### 10. Tag (标签) - 核心实体 +### 10. Tag Entity ```mermaid erDiagram - Tag ||--o{ TagRelation : "applies to" + Tag ||--o{ TagRelation : applies_to Tag { - varchar(64) tag_id PK - varchar(100) tag_name UK - varchar(50) tag_category - varchar(50) tag_type - text description - varchar(20) color + varchar tag_id PK + varchar tag_name UK + varchar tag_category + varchar tag_type + varchar description + varchar color datetime created_at - varchar(64) created_by_user_id FK + varchar created_by_user_id FK } TagRelation { - varchar(64) relation_id PK - varchar(64) tag_id FK - varchar(50) entity_type - varchar(64) entity_id FK + varchar relation_id PK + varchar tag_id FK + varchar entity_type + varchar entity_id FK datetime tagged_at - varchar(64) tagged_by_user_id FK + varchar tagged_by_user_id FK boolean is_auto_tagged - varchar(100) tag_source + varchar tag_source } ``` @@ -698,32 +672,30 @@ erDiagram --- -### 11. Segment (客户分群) - 核心实体 +### 11. Segment Entity ```mermaid erDiagram - Segment ||--o{ SegmentMember : "contains" + Segment ||--o{ SegmentMember : contains Segment { - varchar(64) segment_id PK - varchar(200) segment_name - varchar(50) segment_type - text description - json segment_rules - varchar(50) target_entity_type + varchar segment_id PK + varchar segment_name + varchar segment_type + varchar target_entity_type int member_count boolean is_dynamic datetime last_calculated_at - varchar(64) created_by_user_id FK + varchar created_by_user_id FK datetime created_at datetime updated_at } SegmentMember { - varchar(64) member_id PK - varchar(64) segment_id FK - varchar(50) entity_type - varchar(64) entity_id FK + varchar member_id PK + varchar segment_id FK + varchar entity_type + varchar entity_id FK datetime joined_at datetime left_at boolean is_active @@ -732,42 +704,40 @@ erDiagram --- -### 12. Score (评分模型) - 核心实体 +### 12. Score Entity ```mermaid erDiagram - ScoreModel ||--o{ ScoreRecord : "generates" - ScoreRecord ||--o{ ScoreHistory : "tracks" + ScoreModel ||--o{ ScoreRecord : generates + ScoreRecord ||--o{ ScoreHistory : tracks ScoreModel { - varchar(64) model_id PK - varchar(100) model_name - varchar(50) model_type - varchar(50) target_entity_type - json scoring_rules + varchar model_id PK + varchar model_name + varchar model_type + varchar target_entity_type int max_score - varchar(50) status + varchar status datetime created_at datetime updated_at } ScoreRecord { - varchar(64) record_id PK - varchar(64) model_id FK - varchar(50) entity_type - varchar(64) entity_id FK + varchar record_id PK + varchar model_id FK + varchar entity_type + varchar entity_id FK int score - varchar(10) grade - json score_details + varchar grade datetime calculated_at datetime updated_at } ScoreHistory { - varchar(64) history_id PK - varchar(50) entity_type - varchar(64) entity_id FK - varchar(64) model_id FK + varchar history_id PK + varchar entity_type + varchar entity_id FK + varchar model_id FK int score datetime recorded_at } @@ -780,17 +750,17 @@ erDiagram --- -### 13. Industry (行业) - 核心实体 +### 13. Industry Entity ```mermaid erDiagram - Industry ||--o{ Industry : "parent of" + Industry ||--o{ Industry : parent_of Industry { - varchar(64) industry_id PK - varchar(100) industry_name - varchar(50) industry_code UK - varchar(64) parent_industry_id FK + varchar industry_id PK + varchar industry_name + varchar industry_code UK + varchar parent_industry_id FK int level int sort_order datetime created_at @@ -799,30 +769,28 @@ erDiagram --- -### 14. Attribution (归因) - 核心实体 +### 14. Attribution Entity ```mermaid erDiagram - Attribution ||--o{ TouchpointAttribution : "analyzes" + Attribution ||--o{ TouchpointAttribution : analyzes Attribution { - varchar(64) attribution_id PK - varchar(50) entity_type - varchar(64) entity_id FK - varchar(50) attribution_model - json touchpoint_sequence - json attribution_weights + varchar attribution_id PK + varchar entity_type + varchar entity_id FK + varchar attribution_model datetime created_at datetime updated_at } TouchpointAttribution { - varchar(64) ta_id PK - varchar(64) attribution_id FK - varchar(64) touchpoint_id FK - varchar(64) campaign_id FK - varchar(64) channel_id FK - decimal(5_4) attribution_weight + varchar ta_id PK + varchar attribution_id FK + varchar touchpoint_id FK + varchar campaign_id FK + varchar channel_id FK + decimal attribution_weight int position_in_journey datetime touchpoint_time } @@ -838,33 +806,31 @@ erDiagram --- -### 15. CustomerJourney (客户旅程) - 核心实体 +### 15. CustomerJourney Entity ```mermaid erDiagram - CustomerJourney ||--o{ JourneyStage : "follows" + CustomerJourney ||--o{ JourneyStage : follows CustomerJourney { - varchar(64) journey_id PK - varchar(200) journey_name - varchar(50) entity_type - varchar(64) entity_id FK - varchar(50) journey_stage + varchar journey_id PK + varchar journey_name + varchar entity_type + varchar entity_id FK + varchar journey_stage datetime journey_start_at datetime journey_end_at int total_touchpoints int journey_duration_days - json stage_history datetime created_at datetime updated_at } JourneyStage { - varchar(64) stage_id PK - varchar(100) stage_name + varchar stage_id PK + varchar stage_name int stage_order - varchar(50) stage_category - json milestone_criteria + varchar stage_category datetime created_at } ``` @@ -877,59 +843,59 @@ erDiagram ```mermaid erDiagram - Account ||--o{ AccountContactRelation : "has" - Account ||--|| AccountSummary : "aggregates" - Account ||--o{ Opportunity : "owns" - Account ||--o{ Touchpoint : "receives" - Account ||--o{ AccountChannelIdentity : "identifies in" - - Contact ||--o{ AccountContactRelation : "belongs to" - Contact ||--|| ContactSummary : "aggregates" - Contact ||--o{ Opportunity : "participates" - Contact ||--o{ Touchpoint : "receives" - Contact ||--o{ Event : "generates" - Contact ||--o{ ContactChannelIdentity : "identifies in" - - Lead ||--o| Contact : "converts to" - Lead ||--o| Account : "converts to" - Lead ||--o| Opportunity : "converts to" - Lead ||--|| LeadSummary : "aggregates" - Lead ||--o{ Touchpoint : "receives" - Lead ||--o{ Event : "generates" - Lead ||--o{ LeadChannelIdentity : "identifies in" - - Opportunity ||--o{ OpportunityStageHistory : "tracks" - Opportunity ||--o{ OpportunityProduct : "contains" - - Product ||--o{ OpportunityProduct : "included in" - Product }o--|| ProductCategory : "categorized by" - - Channel ||--o{ Touchpoint : "generates" - Channel ||--o{ Event : "tracks" - Channel ||--o{ ChannelPerformance : "measures" - Channel ||--o{ AccountChannelIdentity : "identifies" - Channel ||--o{ ContactChannelIdentity : "identifies" - Channel ||--o{ LeadChannelIdentity : "identifies" - - Campaign ||--o{ Touchpoint : "drives" - Campaign ||--o{ Lead : "generates" - Campaign ||--o{ CampaignMember : "includes" - Campaign ||--o{ CampaignPerformance : "measures" - - Tag ||--o{ TagRelation : "tags" - Segment ||--o{ SegmentMember : "groups" - - ScoreModel ||--o{ ScoreRecord : "scores" - ScoreRecord ||--o{ ScoreHistory : "history" - - Industry ||--o{ Account : "classifies" - Industry ||--o{ Lead : "classifies" - Industry ||--o{ Industry : "parent of" - - Attribution ||--o{ TouchpointAttribution : "attributes" - Touchpoint ||--o{ TouchpointAttribution : "contributes" - - CustomerJourney ||--o{ JourneyStage : "progresses" + Account ||--o{ AccountContactRelation : has + Account ||--|| AccountSummary : aggregates + Account ||--o{ Opportunity : owns + Account ||--o{ Touchpoint : receives + Account ||--o{ AccountChannelIdentity : identified_by + + Contact ||--o{ AccountContactRelation : belongs_to + Contact ||--|| ContactSummary : aggregates + Contact ||--o{ Opportunity : participates + Contact ||--o{ Touchpoint : receives + Contact ||--o{ Event : generates + Contact ||--o{ ContactChannelIdentity : identified_by + + Lead ||--o| Contact : converts_to + Lead ||--o| Account : converts_to + Lead ||--o| Opportunity : converts_to + Lead ||--|| LeadSummary : aggregates + Lead ||--o{ Touchpoint : receives + Lead ||--o{ Event : generates + Lead ||--o{ LeadChannelIdentity : identified_by + + Opportunity ||--o{ OpportunityStageHistory : tracks + Opportunity ||--o{ OpportunityProduct : contains + + Product ||--o{ OpportunityProduct : included_in + Product }o--|| ProductCategory : categorized_by + + Channel ||--o{ Touchpoint : generates + Channel ||--o{ Event : tracks + Channel ||--o{ ChannelPerformance : measured_by + Channel ||--o{ AccountChannelIdentity : identifies + Channel ||--o{ ContactChannelIdentity : identifies + Channel ||--o{ LeadChannelIdentity : identifies + + Campaign ||--o{ Touchpoint : drives + Campaign ||--o{ Lead : generates + Campaign ||--o{ CampaignMember : includes + Campaign ||--o{ CampaignPerformance : measured_by + + Tag ||--o{ TagRelation : tags + Segment ||--o{ SegmentMember : groups + + ScoreModel ||--o{ ScoreRecord : scores + ScoreRecord ||--o{ ScoreHistory : tracks + + Industry ||--o{ Account : classifies + Industry ||--o{ Lead : classifies + Industry ||--o{ Industry : parent_of + + Attribution ||--o{ TouchpointAttribution : attributes + Touchpoint ||--o{ TouchpointAttribution : contributes + + CustomerJourney ||--o{ JourneyStage : progresses ``` --- @@ -938,57 +904,53 @@ erDiagram ```mermaid graph TB - subgraph Sources["多渠道数据源"] - direction TB - S1["微信公众号
WeChat Official"] - S2["企业微信
Enterprise WeChat"] - S3["官方网站
Website"] - S4["抖音账号
Douyin"] - S5["邮件系统
Email"] - S6["电话系统
Phone"] - S7["线下活动
Offline Event"] + subgraph Sources[Multi-Channel Data Sources] + S1[WeChat Official] + S2[Enterprise WeChat] + S3[Website] + S4[Douyin] + S5[Email System] + S6[Phone System] + S7[Offline Events] end - subgraph Identity["身份识别中心 Identity Resolution"] - direction TB - IR["Identity Resolution Engine
身份识别引擎"] - IM["Identity Matching
身份匹配"] - IG["Identity Grouping
身份聚合"] + subgraph Identity[Identity Resolution Engine] + IR[Identity Matching Rules] + IG[Identity Grouping Logic] end - subgraph Entities["统一客户实体 Unified Entities"] - direction TB - UC["Contact
统一联系人
ID: CNT_001"] - UA["Account
统一企业
ID: ACC_001"] + subgraph Entities[Unified Customer Entities] + UC[Contact
Unified Contact ID] + UA[Account
Unified Account ID] end - subgraph Mapping["渠道身份映射 Channel Identity Mapping"] - direction TB - CI1["ContactChannelIdentity
微信OpenID"] - CI2["ContactChannelIdentity
企业微信UserID"] - CI3["ContactChannelIdentity
网站CookieID"] - CI4["ContactChannelIdentity
抖音UserID"] - AI1["AccountChannelIdentity
企业微信CorpID"] - AI2["AccountChannelIdentity
官网企业ID"] + subgraph Mapping[Channel Identity Mapping] + CI1[ContactChannelIdentity
WeChat OpenID] + CI2[ContactChannelIdentity
Enterprise WeChat UserID] + CI3[ContactChannelIdentity
Website Cookie] + CI4[ContactChannelIdentity
Douyin UserID] + AI1[AccountChannelIdentity
Enterprise WeChat CorpID] + AI2[AccountChannelIdentity
Website Company ID] end - S1 & S2 & S3 & S4 & S5 & S6 & S7 --> IR - IR --> IM - IM --> IG + S1 --> IR + S2 --> IR + S3 --> IR + S4 --> IR + S5 --> IR + S6 --> IR + S7 --> IR + + IR --> IG IG --> UC IG --> UA - UC --> CI1 & CI2 & CI3 & CI4 - UA --> AI1 & AI2 - - classDef sourceClass fill:#e3f2fd,stroke:#1976d2,stroke-width:2px - classDef identityClass fill:#fff3e0,stroke:#f57c00,stroke-width:2px - classDef entityClass fill:#e8f5e9,stroke:#388e3c,stroke-width:2px - classDef mappingClass fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px - - class S1,S2,S3,S4,S5,S6,S7 sourceClass - class IR,IM,IG identityClass - class UC,UA entityClass - class CI1,CI2,CI3,CI4,AI1,AI2 mappingClass + + UC --> CI1 + UC --> CI2 + UC --> CI3 + UC --> CI4 + UA --> AI1 + UA --> AI2 ``` --- @@ -999,34 +961,27 @@ graph TB ```mermaid stateDiagram-v2 - [*] --> NewLead: 捕获线索
Lead Capture + [*] --> NewLead: Lead Capture - state "新线索 New Lead" as NewLead - state "已联系 Contacted" as Contacted - state "已限定 Qualified" as Qualified - state "已转化 Converted" as Converted - state "无效线索 Invalid" as Invalid + NewLead --> Contacted: First Contact + NewLead --> Invalid: Mark Invalid - NewLead --> Contacted: 首次联系
First Contact - NewLead --> Invalid: 标记无效
Mark Invalid + Contacted --> Qualified: Qualification + Contacted --> Invalid: Mark Invalid - Contacted --> Qualified: 资格验证
Qualification - Contacted --> Invalid: 标记无效
Mark Invalid - - Qualified --> Converted: 转化
Convert + Qualified --> Converted: Convert state Converted { - [*] --> CreateContact: 创建联系人
Create Contact - [*] --> CreateAccount: 创建/关联企业
Create/Link Account - [*] --> CreateOpportunity: 创建商机
Create Opportunity - + [*] --> CreateContact + [*] --> CreateAccount + [*] --> CreateOpportunity CreateContact --> [*] CreateAccount --> [*] CreateOpportunity --> [*] } - Converted --> [*]: 完成
Complete - Invalid --> [*]: 结束
End + Converted --> [*] + Invalid --> [*] ``` --- @@ -1035,37 +990,32 @@ stateDiagram-v2 ```mermaid stateDiagram-v2 - [*] --> Lead: 线索阶段
Lead Stage - - state "线索阶段 Lead" as Lead - state "需求确认 Qualification" as Qualification - state "方案设计 Solution Design" as SolutionDesign - state "商务谈判 Negotiation" as Negotiation - state "合同签订 Contract" as Contract - state "赢单 Won" as Won - state "输单 Lost" as Lost + [*] --> Lead: Lead Stage - Lead --> Qualification: 需求分析
Needs Analysis - Lead --> Lost: 丢失
Lost + Lead --> Qualification: Needs Analysis + Lead --> Lost: Lost - Qualification --> SolutionDesign: 方案输出
Solution Output - Qualification --> Lost: 丢失
Lost + Qualification --> SolutionDesign: Solution Design + Qualification --> Lost: Lost - SolutionDesign --> Negotiation: 商务谈判
Negotiation Start - SolutionDesign --> Lost: 丢失
Lost + SolutionDesign --> Negotiation: Negotiation + SolutionDesign --> Lost: Lost - Negotiation --> Contract: 签订合同
Sign Contract - Negotiation --> Lost: 丢失
Lost + Negotiation --> Contract: Sign Contract + Negotiation --> Lost: Lost - Contract --> Won: 成功赢单
Close Won - Contract --> Lost: 失败输单
Close Lost + Contract --> Won: Close Won + Contract --> Lost: Close Lost - Won --> [*]: 完成
Complete - Lost --> [*]: 结束
End + Won --> [*] + Lost --> [*] note right of Won - 更新AccountSummary - Update Account Summary + Update AccountSummary + - total_revenue + - won_opportunities + - win_rate + - health_score end note ``` @@ -1075,77 +1025,139 @@ stateDiagram-v2 ```mermaid stateDiagram-v2 - [*] --> Awareness: 认知阶段
Awareness + [*] --> Awareness: Awareness Stage + + Awareness --> Consideration: Show Interest + Consideration --> Decision: Evaluation + Decision --> Retention: Purchase + Retention --> Expansion: Upsell/Cross-sell + + Retention --> Churn: Inactivity + Expansion --> Churn: Inactivity + + Churn --> Winback: Winback Campaign + Winback --> Retention: Success + Winback --> [*]: Permanent Loss + + Expansion --> [*]: Ongoing +``` + +--- + +### 全渠道数据流转流程 + +```mermaid +sequenceDiagram + participant User + participant Channel + participant Event + participant Identity + participant CDP + participant Lead + participant Contact + participant Account + participant Summary + + User->>Channel: 1. Interaction + Channel->>Event: 2. Log Event + Event->>Identity: 3. Identity Resolution + + alt New User + Identity->>Lead: 4a. Create Lead + Lead->>CDP: 5a. Save Data + CDP->>Summary: 6a. Update LeadSummary + else Identified User + Identity->>Contact: 4b. Link Contact + Contact->>Account: 5b. Link Account + Account->>CDP: 6b. Save Data + CDP->>Summary: 7b. Update Summary + end - state "认知阶段 Awareness" as Awareness - state "考虑阶段 Consideration" as Consideration - state "决策阶段 Decision" as Decision - state "留存阶段 Retention" as Retention - state "扩展阶段 Expansion" as Expansion - state "流失 Churn" as Churn - state "召回 Winback" as Winback + Summary->>CDP: 8. Trigger Rules + CDP->>Channel: 9. Personalized Response + Channel->>User: 10. Return Content +``` + +--- + +### 身份识别匹配流程 + +```mermaid +flowchart TD + Start[Receive Multi-Channel Data] --> Extract[Extract Identity Identifiers] - Awareness --> Consideration: 产生兴趣
Show Interest - Consideration --> Decision: 深度评估
Deep Evaluation - Decision --> Retention: 成交购买
Purchase - Retention --> Expansion: 追加购买
Upsell/Cross-sell + Extract --> Phone{Has
Phone?} + Extract --> Email{Has
Email?} + Extract --> Wechat{Has
WeChat ID?} + Extract --> Corp{Has
Corp Code?} - Retention --> Churn: 停止使用
Stop Using - Expansion --> Churn: 停止使用
Stop Using + Phone -->|Yes| MatchPhone[Match by Phone
Priority: 1] + Email -->|Yes| MatchEmail[Match by Email
Priority: 2] + Wechat -->|Yes| MatchWechat[Match by WeChat
Priority: 3] + Corp -->|Yes| MatchCorp[Match by Corp Code
Priority: 1] - Churn --> Winback: 召回营销
Winback Campaign - Winback --> Retention: 成功召回
Successfully Winback - Winback --> [*]: 永久流失
Permanent Loss + MatchPhone --> FoundContact{Found
Contact?} + MatchEmail --> FoundContact + MatchWechat --> FoundContact - Expansion --> [*]: 持续合作
Ongoing Partnership + MatchCorp --> FoundAccount{Found
Account?} - note right of Retention - health_score > 70 - 健康客户 - end note + FoundContact -->|Yes| MergeContact[Merge to Existing Contact] + FoundContact -->|No| CreateContact[Create New Contact] - note right of Churn - health_score < 30 - 预警流失 - end note + FoundAccount -->|Yes| MergeAccount[Merge to Existing Account] + FoundAccount -->|No| CreateAccount[Create New Account] + + MergeContact --> AddIdentity[Add/Update
ContactChannelIdentity] + CreateContact --> AddIdentity + + MergeAccount --> AddAccountIdentity[Add/Update
AccountChannelIdentity] + CreateAccount --> AddAccountIdentity + + AddIdentity --> UpdateSummary[Update ContactSummary] + AddAccountIdentity --> UpdateAccountSummary[Update AccountSummary] + + UpdateSummary --> End[Complete] + UpdateAccountSummary --> End ``` --- -### 全渠道数据流转流程 +### 汇总数据更新策略 ```mermaid -sequenceDiagram - actor User as 用户
User - participant CH as 渠道
Channel - participant EV as 事件系统
Event System - participant ID as 身份识别
Identity Resolution - participant CDP as CDP核心
CDP Core - participant LD as Lead管理
Lead Management - participant CT as Contact管理
Contact Management - participant AC as Account管理
Account Management - participant SM as 汇总计算
Summary Calculation - - User->>CH: 1. 访问互动
Visit/Interact - CH->>EV: 2. 记录事件
Log Event - EV->>ID: 3. 身份识别
Identify - - alt 新用户 New User - ID->>LD: 4a. 创建Lead
Create Lead - LD->>CDP: 5a. 保存数据
Save Data - CDP->>SM: 6a. 更新LeadSummary
Update Summary - else 已识别用户 Identified User - ID->>CT: 4b. 关联Contact
Link Contact - CT->>AC: 5b. 关联Account
Link Account - AC->>CDP: 6b. 保存数据
Save Data - CDP->>SM: 7b. 更新Summary
Update Summary +flowchart LR + subgraph Triggers[Update Triggers] + T1[Scheduled Job
Hourly] + T2[Real-time Trigger
Critical Events] + T3[Manual Refresh
On Demand] + end + + subgraph Engine[Calculation Engine] + C1[Aggregation] + C2[Scoring] + C3[Rule Engine] + end + + subgraph Summary[Summary Tables] + S1[AccountSummary] + S2[ContactSummary] + S3[LeadSummary] end - SM->>CDP: 8. 触发规则引擎
Trigger Rules - CDP->>CH: 9. 个性化响应
Personalized Response - CH->>User: 10. 返回内容
Return Content + T1 --> C1 + T2 --> C2 + T3 --> C3 - Note over SM: ContactSummary
AccountSummary
LeadSummary + C1 --> S1 + C1 --> S2 + C1 --> S3 + C2 --> S1 + C2 --> S2 + C2 --> S3 + C3 --> S1 + C3 --> S2 + C3 --> S3 ``` --- @@ -1159,24 +1171,21 @@ sequenceDiagram | account_id | VARCHAR | 64 | PK, NOT NULL | 账户唯一标识 | ACC_20231105001 | | account_name | VARCHAR | 200 | NOT NULL | 企业名称 | 阿里巴巴网络技术有限公司 | | unified_social_credit_code | VARCHAR | 18 | UNIQUE | 统一社会信用代码 | 91330000MA27XYZ123 | -| account_type | VARCHAR | 50 | NOT NULL | 客户类型 | CUSTOMER / PARTNER / COMPETITOR / PROSPECT | +| account_type | VARCHAR | 50 | NOT NULL | 客户类型 | CUSTOMER, PARTNER, COMPETITOR, PROSPECT | | industry_id | VARCHAR | 64 | FK | 行业分类外键 | IND_001 | -| account_status | VARCHAR | 50 | NOT NULL | 账户状态 | ACTIVE / DORMANT / CHURNED / BLACKLIST | -| account_level | VARCHAR | 50 | | 客户等级 | STRATEGIC / IMPORTANT / NORMAL | +| account_status | VARCHAR | 50 | NOT NULL | 账户状态 | ACTIVE, DORMANT, CHURNED, BLACKLIST | +| account_level | VARCHAR | 50 | | 客户等级 | STRATEGIC, IMPORTANT, NORMAL | | annual_revenue | DECIMAL | 18,2 | | 年营收(万元) | 50000.00 | | employee_count | INT | | | 员工人数 | 5000 | | company_website | VARCHAR | 500 | | 公司网站 | https://www.alibaba.com | -| company_address | VARCHAR | 500 | | 公司地址 | 浙江省杭州市余杭区文一西路969号 | | province | VARCHAR | 50 | | 省份 | 浙江省 | | city | VARCHAR | 50 | | 城市 | 杭州市 | -| district | VARCHAR | 50 | | 区县 | 余杭区 | -| account_source | VARCHAR | 100 | | 来源渠道 | WEBSITE / EXHIBITION / PARTNER / COLD_CALL | +| account_source | VARCHAR | 100 | | 来源渠道 | WEBSITE, EXHIBITION, PARTNER, COLD_CALL | | primary_channel_id | VARCHAR | 64 | FK | 主渠道ID | CH_001 | | owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | | created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | | updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | -| custom_fields | JSON | | | 自定义扩展字段 | {"crm_id": "CRM001", "region": "east"} | -| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | AWARENESS / CONSIDERATION / DECISION / RETENTION / EXPANSION | +| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | AWARENESS, CONSIDERATION, DECISION, RETENTION, EXPANSION | --- @@ -1195,22 +1204,11 @@ sequenceDiagram | lost_opportunities | INT | | DEFAULT 0 | 输单数量 | 2 | | win_rate | DECIMAL | 5,2 | DEFAULT 0 | 赢单率(%) | 71.43 | | total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 156 | -| active_campaigns | INT | | DEFAULT 0 | 活跃营销活动数 | 3 | | health_score | INT | | DEFAULT 0 | 健康度评分(0-100) | 85 | | last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | -| last_purchase_at | DATETIME | | | 最后购买时间 | 2023-10-15 09:00:00 | -| first_purchase_date | DATE | | | 首次购买日期 | 2022-03-20 | -| latest_opportunity_date | DATE | | | 最新商机日期 | 2023-11-01 | -| days_since_last_contact | INT | | DEFAULT 0 | 距上次联系天数 | 5 | | calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | | updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | -**AccountSummary 设计说明:** -1. 该表用于存储Account的所有汇总统计数据，避免频繁实时计算 -2. 建议通过定时任务(如每小时)或触发器更新 -3. 用于快速查询和展示客户360度视图 -4. 支持客户健康度监控和预警 - --- ### Contact 详细字段设计 @@ -1224,15 +1222,14 @@ sequenceDiagram | wechat_id | VARCHAR | 100 | | 微信ID | wx_zhangwei | | job_title | VARCHAR | 100 | | 职位 | 首席技术官 CTO | | department | VARCHAR | 100 | | 部门 | 技术部 | -| contact_status | VARCHAR | 50 | NOT NULL | 联系人状态 | ACTIVE / INACTIVE / BOUNCED / UNSUBSCRIBED | +| contact_status | VARCHAR | 50 | NOT NULL | 联系人状态 | ACTIVE, INACTIVE, BOUNCED, UNSUBSCRIBED | | primary_account_id | VARCHAR | 64 | FK | 主要关联企业ID | ACC_20231105001 | -| contact_source | VARCHAR | 100 | | 来源 | WEBSITE / FORM / IMPORT / API / REFERRAL | +| contact_source | VARCHAR | 100 | | 来源 | WEBSITE, FORM, IMPORT, API, REFERRAL | | primary_channel_id | VARCHAR | 64 | FK | 主渠道ID | CH_001 | | owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | | created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | | updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | -| custom_fields | JSON | | | 自定义字段 | {"birthday": "1985-01-01", "hobby": "reading"} | -| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | SUBSCRIBER / LEAD / MQL / SQL / OPPORTUNITY / CUSTOMER | +| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | SUBSCRIBER, LEAD, MQL, SQL, OPPORTUNITY, CUSTOMER | | is_decision_maker | BOOLEAN | | DEFAULT FALSE | 是否决策者 | true | | is_verified | BOOLEAN | | DEFAULT FALSE | 是否已验证 | true | @@ -1252,10 +1249,7 @@ sequenceDiagram | content_downloads | INT | | DEFAULT 0 | 内容下载次数 | 8 | | engagement_score | INT | | DEFAULT 0 | 参与度评分(0-100) | 78 | | last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | -| last_email_sent_at | DATETIME | | | 最后邮件发送时间 | 2023-11-04 10:00:00 | -| last_email_opened_at | DATETIME | | | 最后邮件打开时间 | 2023-11-04 15:30:00 | | days_since_last_activity | INT | | DEFAULT 0 | 距上次活跃天数 | 1 | -| campaign_responses | INT | | DEFAULT 0 | 营销活动响应次数 | 5 | | calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | | updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | @@ -1272,12 +1266,12 @@ sequenceDiagram | email | VARCHAR | 200 | | 邮箱地址 | liming@company.com | | wechat_id | VARCHAR | 100 | | 微信ID | wx_liming | | job_title | VARCHAR | 100 | | 职位 | 产品经理 | -| lead_source | VARCHAR | 100 | NOT NULL | 线索来源 | WEBSITE / FORM / CAMPAIGN / COLD_CALL / REFERRAL | +| lead_source | VARCHAR | 100 | NOT NULL | 线索来源 | WEBSITE, FORM, CAMPAIGN, COLD_CALL, REFERRAL | | channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | | campaign_id | VARCHAR | 64 | FK | 营销活动ID | CMP_001 | -| lead_status | VARCHAR | 50 | NOT NULL | 线索状态 | NEW / CONTACTED / QUALIFIED / CONVERTED / DISQUALIFIED | +| lead_status | VARCHAR | 50 | NOT NULL | 线索状态 | NEW, CONTACTED, QUALIFIED, CONVERTED, DISQUALIFIED | | lead_score | INT | | DEFAULT 0 | 线索评分(0-100) | 80 | -| lead_grade | VARCHAR | 10 | | 线索等级 | A / B / C / D | +| lead_grade | VARCHAR | 10 | | 线索等级 | A, B, C, D | | industry_id | VARCHAR | 64 | FK | 行业ID | IND_001 | | province | VARCHAR | 50 | | 省份 | 广东省 | | city | VARCHAR | 50 | | 城市 | 深圳市 | @@ -1289,7 +1283,6 @@ sequenceDiagram | converted_contact_id | VARCHAR | 64 | FK | 转化后联系人ID | CNT_20231110001 | | converted_account_id | VARCHAR | 64 | FK | 转化后企业ID | ACC_20231110001 | | converted_opportunity_id | VARCHAR | 64 | FK | 转化后商机ID | OPP_20231110001 | -| custom_fields | JSON | | | 自定义字段 | {"product_interest": "AI", "budget": "100K"} | | is_qualified | BOOLEAN | | DEFAULT FALSE | 是否为合格线索 | true | --- @@ -1323,21 +1316,19 @@ sequenceDiagram | account_id | VARCHAR | 64 | FK, NOT NULL | 关联企业ID | ACC_20231105001 | | primary_contact_id | VARCHAR | 64 | FK | 主要联系人ID | CNT_20231105001 | | lead_id | VARCHAR | 64 | FK | 来源线索ID | LEAD_20231105001 | -| opportunity_type | VARCHAR | 50 | | 商机类型 | NEW_BUSINESS / UPSELL / RENEWAL / CROSS_SELL | -| opportunity_source | VARCHAR | 100 | | 商机来源 | LEAD_CONVERSION / DIRECT_SALES / PARTNER | +| opportunity_type | VARCHAR | 50 | | 商机类型 | NEW_BUSINESS, UPSELL, RENEWAL, CROSS_SELL | +| opportunity_source | VARCHAR | 100 | | 商机来源 | LEAD_CONVERSION, DIRECT_SALES, PARTNER | | amount | DECIMAL | 18,2 | | 预计金额(元) | 1000000.00 | -| currency | VARCHAR | 10 | | 货币单位 | CNY / USD / EUR | -| stage | VARCHAR | 50 | NOT NULL | 当前阶段 | QUALIFICATION / NEEDS_ANALYSIS / PROPOSAL / NEGOTIATION / CLOSED_WON / CLOSED_LOST | +| currency | VARCHAR | 10 | | 货币单位 | CNY, USD, EUR | +| stage | VARCHAR | 50 | NOT NULL | 当前阶段 | QUALIFICATION, NEEDS_ANALYSIS, PROPOSAL, NEGOTIATION, CLOSED_WON, CLOSED_LOST | | probability | INT | | DEFAULT 0 | 赢单概率(0-100) | 60 | | expected_close_date | DATE | | | 预计成交日期 | 2023-12-31 | | actual_close_date | DATE | | | 实际成交日期 | 2023-12-25 | -| close_reason | VARCHAR | 200 | | 关闭原因 | 价格因素 / 竞争对手 / 预算不足 / 成功签约 | +| close_reason | VARCHAR | 200 | | 关闭原因 | 价格因素, 竞争对手, 预算不足, 成功签约 | | owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | -| product_ids | JSON | | | 关联产品ID数组 | ["PRD_001", "PRD_002"] | | campaign_id | VARCHAR | 64 | FK | 来源活动ID | CMP_001 | | created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | | updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | -| custom_fields | JSON | | | 自定义字段 | {"contract_type": "annual", "payment_terms": "quarterly"} | | days_in_stage | INT | | DEFAULT 0 | 当前阶段停留天数 | 15 | | is_won | BOOLEAN | | DEFAULT FALSE | 是否赢单 | false | | is_lost | BOOLEAN | | DEFAULT FALSE | 是否输单 | false | @@ -1350,15 +1341,13 @@ sequenceDiagram |--------|------|------|------|------|--------| | channel_id | VARCHAR | 64 | PK, NOT NULL | 渠道唯一标识 | CH_001 | | channel_name | VARCHAR | 100 | NOT NULL | 渠道名称 | 官网-产品页 | -| channel_type | VARCHAR | 50 | NOT NULL | 渠道类型 | WEBSITE / WECHAT / DOUYIN / EMAIL / PHONE / OFFLINE / PARTNER | -| channel_category | VARCHAR | 50 | | 渠道分类 | ONLINE / OFFLINE / SOCIAL / DIRECT | +| channel_type | VARCHAR | 50 | NOT NULL | 渠道类型 | WEBSITE, WECHAT, DOUYIN, EMAIL, PHONE, OFFLINE, PARTNER | +| channel_category | VARCHAR | 50 | | 渠道分类 | ONLINE, OFFLINE, SOCIAL, DIRECT | | parent_channel_id | VARCHAR | 64 | FK | 父渠道ID | CH_PARENT_001 | -| channel_status | VARCHAR | 50 | NOT NULL | 渠道状态 | ACTIVE / INACTIVE / TESTING | -| channel_config | JSON | | | 渠道配置信息 | {"api_key": "xxx", "webhook_url": "xxx", "app_id": "xxx"} | +| channel_status | VARCHAR | 50 | NOT NULL | 渠道状态 | ACTIVE, INACTIVE, TESTING | | cost | DECIMAL | 18,2 | | 渠道成本(元/月) | 50000.00 | | created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | | updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | -| custom_fields | JSON | | | 自定义字段 | {"partner_name": "某某合作伙伴", "contract_end": "2024-12-31"} | --- @@ -1367,23 +1356,20 @@ sequenceDiagram | 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | |--------|------|------|------|------|--------| | touchpoint_id | VARCHAR | 64 | PK, NOT NULL | 触点唯一标识 | TP_20231105001 | -| touchpoint_type | VARCHAR | 50 | NOT NULL | 触点类型 | PAGE_VIEW / FORM_SUBMIT / DOWNLOAD / EMAIL / CALL / MEETING / CHAT | +| touchpoint_type | VARCHAR | 50 | NOT NULL | 触点类型 | PAGE_VIEW, FORM_SUBMIT, DOWNLOAD, EMAIL, CALL, MEETING, CHAT | | channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | | campaign_id | VARCHAR | 64 | FK | 营销活动ID | CMP_001 | | contact_id | VARCHAR | 64 | FK | 联系人ID | CNT_20231105001 | | lead_id | VARCHAR | 64 | FK | 线索ID | LEAD_20231105001 | | account_id | VARCHAR | 64 | FK | 企业ID | ACC_20231105001 | | touchpoint_time | DATETIME | | NOT NULL | 触点发生时间 | 2023-11-05 14:30:00 | -| touchpoint_direction | VARCHAR | 20 | | 触点方向 | INBOUND / OUTBOUND | -| touchpoint_status | VARCHAR | 50 | | 触点状态 | COMPLETED / SCHEDULED / CANCELLED | -| content_type | VARCHAR | 50 | | 内容类型 | WHITEPAPER / CASE_STUDY / WEBINAR / DEMO / PROPOSAL | +| touchpoint_direction | VARCHAR | 20 | | 触点方向 | INBOUND, OUTBOUND | +| touchpoint_status | VARCHAR | 50 | | 触点状态 | COMPLETED, SCHEDULED, CANCELLED | +| content_type | VARCHAR | 50 | | 内容类型 | WHITEPAPER, CASE_STUDY, WEBINAR, DEMO, PROPOSAL | | content_id | VARCHAR | 64 | FK | 内容ID | CONTENT_001 | | subject | VARCHAR | 200 | | 主题 | 产品演示会议 | -| description | TEXT | | | 描述详情 | 讨论了AI平台的技术架构和实施方案 | | duration_seconds | INT | | | 持续时长(秒) | 3600 | | owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | -| metadata | JSON | | | 元数据 | {"device": "mobile", "location": "Beijing", "ip": "192.168.1.1"} | -| utm_params | JSON | | | UTM追踪参数 | {"utm_source": "baidu", "utm_medium": "cpc", "utm_campaign": "Q4"} | | created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 14:30:00 | --- @@ -1394,108 +1380,23 @@ sequenceDiagram |--------|------|------|------|------|--------| | event_id | VARCHAR | 64 | PK, NOT NULL | 事件唯一标识 | EVT_20231105001 | | event_name | VARCHAR | 100 | NOT NULL | 事件名称 | page_view | -| event_type | VARCHAR | 50 | NOT NULL | 事件类型 | PAGE_VIEW / CLICK / FORM_START / FORM_SUBMIT / DOWNLOAD / VIDEO_PLAY | +| event_type | VARCHAR | 50 | NOT NULL | 事件类型 | PAGE_VIEW, CLICK, FORM_START, FORM_SUBMIT, DOWNLOAD, VIDEO_PLAY | | channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | | contact_id | VARCHAR | 64 | FK | 联系人ID | CNT_20231105001 | | lead_id | VARCHAR | 64 | FK | 线索ID | LEAD_20231105001 | | account_id | VARCHAR | 64 | FK | 企业ID | ACC_20231105001 | | event_time | DATETIME | | NOT NULL | 事件发生时间 | 2023-11-05 14:35:20 | | session_id | VARCHAR | 64 | | 会话ID | SESSION_20231105001 | -| device_type | VARCHAR | 50 | | 设备类型 | DESKTOP / MOBILE / TABLET | -| browser | VARCHAR | 50 | | 浏览器 | Chrome / Safari / Firefox / Edge | -| os | VARCHAR | 50 | | 操作系统 | Windows 10 / iOS 16 / Android 13 | +| device_type | VARCHAR | 50 | | 设备类型 | DESKTOP, MOBILE, TABLET | +| browser | VARCHAR | 50 | | 浏览器 | Chrome, Safari, Firefox, Edge | +| os | VARCHAR | 50 | | 操作系统 | Windows 10, iOS 16, Android 13 | | ip_address | VARCHAR | 50 | | IP地址 | 192.168.1.1 | | page_url | VARCHAR | 1000 | | 页面URL | https://www.example.com/product/ai-platform | | referrer_url | VARCHAR | 1000 | | 来源URL | https://www.baidu.com/s?wd=AI平台 | -| event_properties | JSON | | | 事件属性 | {"button_text": "申请试用", "form_id": "trial_form", "product": "AI平台"} | -| utm_params | JSON | | | UTM追踪参数 | {"utm_source": "baidu", "utm_campaign": "Q4_campaign", "utm_content": "ad1"} | | created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 14:35:20 | --- -## 全渠道身份映射方案 - -### 身份识别匹配规则 - -```mermaid -flowchart TD - Start([接收多渠道数据]) --> ExtractID[提取身份标识符] - - ExtractID --> CheckPhone{是否有
手机号?} - ExtractID --> CheckEmail{是否有
邮箱?} - ExtractID --> CheckWechat{是否有
微信ID?} - ExtractID --> CheckCorpID{是否有
企业信用代码?} - - CheckPhone -->|是| MatchPhone[手机号匹配
优先级: 1] - CheckEmail -->|是| MatchEmail[邮箱匹配
优先级: 2] - CheckWechat -->|是| MatchWechat[微信ID匹配
优先级: 3] - CheckCorpID -->|是| MatchCorpID[企业信用代码匹配
优先级: 1] - - MatchPhone --> FoundContact{找到匹配
Contact?} - MatchEmail --> FoundContact - MatchWechat --> FoundContact - - MatchCorpID --> FoundAccount{找到匹配
Account?} - - FoundContact -->|是| MergeContact[合并到现有Contact] - FoundContact -->|否| CreateContact[创建新Contact] - - FoundAccount -->|是| MergeAccount[合并到现有Account] - FoundAccount -->|否| CreateAccount[创建新Account] - - MergeContact --> AddChannelIdentity[添加/更新
ContactChannelIdentity] - CreateContact --> AddChannelIdentity - - MergeAccount --> AddAccountIdentity[添加/更新
AccountChannelIdentity] - CreateAccount --> AddAccountIdentity - - AddChannelIdentity --> UpdateSummary[更新ContactSummary] - AddAccountIdentity --> UpdateAccountSummary[更新AccountSummary] - - UpdateSummary --> End([完成]) - UpdateAccountSummary --> End - - style Start fill:#e1f5ff - style End fill:#e8f5e9 - style MergeContact fill:#fff9c4 - style MergeAccount fill:#fff9c4 - style CreateContact fill:#ffe0b2 - style CreateAccount fill:#ffe0b2 -``` - -### 身份聚合优先级 - -```mermaid -graph LR - subgraph Priority["身份匹配优先级 Identity Matching Priority"] - direction TB - P1["优先级1: 手机号
Priority 1: Mobile Phone
唯一性: 高
准确性: 95%"] - P2["优先级2: 邮箱
Priority 2: Email
唯一性: 高
准确性: 90%"] - P3["优先级3: 企业微信ID
Priority 3: Enterprise WeChat
唯一性: 中
准确性: 85%"] - P4["优先级4: 统一社会信用代码
Priority 4: Unified Credit Code
唯一性: 极高
准确性: 99%"] - P5["优先级5: 自定义ID
Priority 5: Custom ID
唯一性: 中
准确性: 70%"] - end - - subgraph Result["匹配结果 Matching Result"] - direction TB - R1["Contact统一身份
Unified Contact"] - R2["Account统一身份
Unified Account"] - end - - P1 & P2 & P3 & P5 --> R1 - P4 --> R2 - - style P1 fill:#c8e6c9 - style P2 fill:#c8e6c9 - style P4 fill:#c8e6c9 - style P3 fill:#fff9c4 - style P5 fill:#ffccbc - style R1 fill:#e1f5ff - style R2 fill:#e1f5ff -``` - ---- - ## 数据字典总结 ### 核心实体统计 @@ -1513,13 +1414,13 @@ graph LR | 性能实体 | ChannelPerformance, CampaignPerformance | 2 | 绩效统计 | | 历史实体 | OpportunityStageHistory | 1 | 变更历史追踪 | | 支撑实体 | Industry, TouchpointAttachment | 2 | 基础支撑数据 | -| **合计** | | **36** | 完整覆盖B2B CDP业务场景 | +| 合计 | | 36 | 完整覆盖B2B CDP业务场景 | --- ## 汇总数据表设计说明 -### 为什么需要汇总数据表 +### 汇总表的设计原则 1. **性能优化** - 避免频繁的多表JOIN和聚合计算 @@ -1536,105 +1437,22 @@ graph LR - 定时批量更新保证数据准确性 - 避免实时计算的结果不一致 -### 汇总数据更新策略 +### 汇总表更新机制 -```mermaid -flowchart LR - subgraph Trigger["触发更新 Update Triggers"] - T1[定时任务
Scheduled Job
每小时] - T2[实时触发
Real-time Trigger
关键事件] - T3[手动刷新
Manual Refresh
按需执行] - end - - subgraph Calculation["计算引擎 Calculation Engine"] - C1[聚合计算
Aggregation] - C2[评分计算
Scoring] - C3[规则引擎
Rule Engine] - end - - subgraph Summary["汇总表 Summary Tables"] - S1[AccountSummary
企业汇总] - S2[ContactSummary
联系人汇总] - S3[LeadSummary
线索汇总] - end - - T1 --> C1 - T2 --> C2 - T3 --> C3 - - C1 & C2 & C3 --> S1 & S2 & S3 - - style T1 fill:#e3f2fd - style T2 fill:#fff3e0 - style T3 fill:#f3e5f5 - style S1 fill:#e8f5e9 - style S2 fill:#e8f5e9 - style S3 fill:#e8f5e9 -``` +**更新触发条件:** +- 定时任务：每小时执行一次全量更新 +- 实时触发：关键业务事件发生时立即更新（如商机赢单、联系人新增） +- 手动刷新：管理员按需触发更新 + +**更新策略:** +- 增量更新：仅更新有变化的记录 +- 全量更新：定期执行完整重算 +- 异步更新：使用消息队列避免阻塞主业务流程 --- ## 技术实现建议 -### 数据库架构 - -```mermaid -flowchart TB - subgraph AppLayer["应用层 Application Layer"] - API[API服务
API Service] - Worker[后台任务
Background Worker] - end - - subgraph DataLayer["数据层 Data Layer"] - direction TB - - subgraph OLTP["OLTP 在线事务处理"] - PG[(PostgreSQL
主数据存储
Account/Contact/Lead
Opportunity)] - end - - subgraph BigData["大数据存储 Big Data Storage"] - CH[(ClickHouse
行为数据
Event/Touchpoint
10亿+记录)] - end - - subgraph Cache["缓存层 Cache Layer"] - Redis[(Redis
热数据缓存
Summary/Score/Tag)] - end - - subgraph Search["搜索引擎 Search Engine"] - ES[(Elasticsearch
全文检索
客户搜索/内容搜索)] - end - - subgraph Analytics["分析层 Analytics Layer"] - DW[(数据仓库
Data Warehouse
BI报表/数据分析)] - end - end - - API --> PG - API --> Redis - API --> ES - - Worker --> PG - Worker --> CH - Worker --> Redis - - PG -.同步.-> ES - PG -.同步.-> Redis - PG -.ETL.-> DW - CH -.ETL.-> DW - - classDef oltp fill:#e8f5e9,stroke:#388e3c,stroke-width:2px - classDef bigdata fill:#fff3e0,stroke:#f57c00,stroke-width:2px - classDef cache fill:#fce4ec,stroke:#c2185b,stroke-width:2px - classDef search fill:#e1f5ff,stroke:#0277bd,stroke-width:2px - classDef analytics fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px - - class PG oltp - class CH bigdata - class Redis cache - class ES search - class DW analytics -``` - ### 索引设计建议 **Account表索引:** @@ -1643,7 +1461,8 @@ flowchart TB PRIMARY KEY (account_id); -- 唯一索引 -CREATE UNIQUE INDEX uk_account_credit_code ON Account(unified_social_credit_code) WHERE unified_social_credit_code IS NOT NULL; +CREATE UNIQUE INDEX uk_account_credit_code ON Account(unified_social_credit_code) +WHERE unified_social_credit_code IS NOT NULL; -- 业务查询索引 CREATE INDEX idx_account_name ON Account(account_name); @@ -1677,8 +1496,10 @@ CREATE INDEX idx_summary_win_rate ON AccountSummary(win_rate DESC); PRIMARY KEY (contact_id); -- 唯一索引 -CREATE UNIQUE INDEX uk_contact_phone ON Contact(mobile_phone) WHERE mobile_phone IS NOT NULL; -CREATE UNIQUE INDEX uk_contact_email ON Contact(email) WHERE email IS NOT NULL; +CREATE UNIQUE INDEX uk_contact_phone ON Contact(mobile_phone) +WHERE mobile_phone IS NOT NULL; +CREATE UNIQUE INDEX uk_contact_email ON Contact(email) +WHERE email IS NOT NULL; -- 业务查询索引 CREATE INDEX idx_contact_name ON Contact(contact_name); From 936caac1f3e8c1c553cae99930247d83b826daef Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 14:51:01 +0000 Subject: [PATCH 04/10] Refactor: Improve B2B CDP entity design documentation Co-authored-by: 13585811473 <13585811473@163.com> --- docs/b2b-cdp-entity-design.md | 1272 +++++++++++++-------------------- 1 file changed, 510 insertions(+), 762 deletions(-) diff --git a/docs/b2b-cdp-entity-design.md b/docs/b2b-cdp-entity-design.md index 70a16cb8d..fe3cbb0cf 100644 --- a/docs/b2b-cdp-entity-design.md +++ b/docs/b2b-cdp-entity-design.md @@ -1,170 +1,103 @@ -# B2B CDP 实体设计详细方案 +# B2B CDP Entity Design Specification -## 目录 -- [整体架构](#整体架构) -- [核心实体设计](#核心实体设计) -- [实体关系图](#实体关系图) -- [业务流程](#业务流程) -- [数据模型详细设计](#数据模型详细设计) +## Table of Contents +- [Architecture Overview](#architecture-overview) +- [Core Entity Design](#core-entity-design) +- [Entity Relationship](#entity-relationship) +- [Business Processes](#business-processes) +- [Data Model Design](#data-model-design) --- -## 整体架构 +## Architecture Overview -### 实体分层架构图 +### Layered Architecture ```mermaid graph TB - subgraph Layer1[Data Source Layer] - DS1[Official Website] - DS2[WeChat Ecosystem] - DS3[Social Media] - DS4[Offline Events] - DS5[Phone & Email] - DS6[CRM System] - DS7[3rd Party Platform] - end - - subgraph Layer2[Channel Layer] - Channel[Channel Management] - end - - subgraph Layer3[Interaction Layer] - Touchpoint[Touchpoint] - Event[Event] - Campaign[Campaign] - end - - subgraph Layer4[Customer Entity Layer] - Contact[Contact] - Lead[Lead] - Account[Account] - end - - subgraph Layer5[Summary Layer] - ContactSummary[ContactSummary] - LeadSummary[LeadSummary] - AccountSummary[AccountSummary] - end - - subgraph Layer6[Business Layer] - Opportunity[Opportunity] - Product[Product] - end - - subgraph Layer7[Insight Layer] - Segment[Segment] - Tag[Tag] - Score[Score Model] - Journey[Customer Journey] - Attribution[Attribution] - end - - DS1 --> Channel - DS2 --> Channel - DS3 --> Channel - DS4 --> Channel - DS5 --> Channel - DS6 --> Channel - DS7 --> Channel - - Channel --> Touchpoint - Channel --> Event - Campaign --> Touchpoint - - Touchpoint --> Contact - Touchpoint --> Lead - Event --> Contact - Event --> Lead - - Contact --> Account - Lead --> Contact - - Contact --> ContactSummary - Lead --> LeadSummary - Account --> AccountSummary - - Lead --> Opportunity - Account --> Opportunity - Opportunity --> Product - - Contact --> Segment - Account --> Segment - Contact --> Tag - Account --> Tag - Contact --> Score - Account --> Score - Lead --> Score - Contact --> Journey - Campaign --> Attribution - Opportunity --> Attribution + DS1[Official Website] --> CH[Channel Layer] + DS2[WeChat Ecosystem] --> CH + DS3[Social Media] --> CH + DS4[Offline Events] --> CH + DS5[Phone & Email] --> CH + DS6[CRM System] --> CH + DS7[3rd Party Platform] --> CH + + CH --> TP[Touchpoint] + CH --> EV[Event] + CM[Campaign] --> TP + + TP --> CT[Contact] + TP --> LD[Lead] + EV --> CT + EV --> LD + + CT --> AC[Account] + LD --> CT + + CT --> CTS[ContactSummary] + LD --> LDS[LeadSummary] + AC --> ACS[AccountSummary] + + LD --> OP[Opportunity] + AC --> OP + OP --> PR[Product] + + CT --> SG[Segment] + AC --> SG + CT --> TG[Tag] + AC --> TG + CT --> SC[ScoreModel] + AC --> SC + LD --> SC + CT --> JN[CustomerJourney] + CM --> AT[Attribution] + OP --> AT ``` --- -### 数据库架构设计 +### Technology Architecture ```mermaid graph LR - subgraph ApplicationLayer[Application Layer] - API[API Gateway] - Worker[Background Jobs] - end + API[API Gateway] --> PG[(PostgreSQL)] + API --> RD[(Redis Cache)] + API --> ES[(Elasticsearch)] - subgraph DatabaseLayer[Database Layer] - PostgreSQL[(PostgreSQL
OLTP Database
Account/Contact/Lead
Opportunity)] - ClickHouse[(ClickHouse
OLAP Database
Event/Touchpoint)] - Redis[(Redis
Cache Layer
Summary/Score)] - end + WK[Background Worker] --> PG + WK --> CH[(ClickHouse)] + WK --> RD - subgraph AnalyticsLayer[Analytics Layer] - Elasticsearch[(Elasticsearch
Search Engine)] - DataWarehouse[(Data Warehouse
BI & Analytics)] - end - - API --> PostgreSQL - API --> Redis - API --> Elasticsearch - Worker --> PostgreSQL - Worker --> ClickHouse - Worker --> Redis - PostgreSQL --> Elasticsearch - PostgreSQL --> DataWarehouse - ClickHouse --> DataWarehouse + PG --> DW[(Data Warehouse)] + CH --> DW ``` --- -## 核心实体设计 +## Core Entity Design -### 1. Account Entity +### Account Entity Model ```mermaid erDiagram Account ||--o{ AccountChannelIdentity : has Account ||--|| AccountSummary : aggregates Account ||--o{ AccountRelation : parent - Account ||--o{ AccountRelation : child Account { varchar account_id PK varchar account_name varchar unified_social_credit_code UK varchar account_type - varchar industry_id FK varchar account_status varchar account_level decimal annual_revenue int employee_count - varchar company_website - varchar province - varchar city - varchar account_source - varchar primary_channel_id FK + varchar industry_id FK varchar owner_user_id FK datetime created_at datetime updated_at - varchar lifecycle_stage } AccountChannelIdentity { @@ -172,56 +105,43 @@ erDiagram varchar account_id FK varchar channel_id FK varchar channel_account_id - varchar identity_type boolean is_verified datetime first_seen_at - datetime last_seen_at } AccountSummary { varchar summary_id PK - varchar account_id FK UK + varchar account_id FK int total_contacts int total_opportunities - int total_leads decimal total_revenue decimal lifetime_value int won_opportunities - int lost_opportunities decimal win_rate - int total_touchpoints int health_score datetime last_activity_at datetime calculated_at } - - AccountRelation { - varchar relation_id PK - varchar parent_account_id FK - varchar child_account_id FK - varchar relation_type - datetime created_at - } ``` -**Account 核心字段说明:** -- `account_id`: 全局唯一账户标识 -- `unified_social_credit_code`: 统一社会信用代码（企业唯一标识） -- `account_type`: 客户类型（CUSTOMER, PARTNER, COMPETITOR, PROSPECT） -- `account_status`: 账户状态（ACTIVE, DORMANT, CHURNED, BLACKLIST） -- `account_level`: 客户等级（STRATEGIC, IMPORTANT, NORMAL） -- `lifecycle_stage`: 生命周期阶段（AWARENESS, CONSIDERATION, DECISION, RETENTION, EXPANSION） - -**AccountSummary 核心指标:** -- `total_contacts`: 关联联系人总数 -- `total_revenue`: 累计收入金额 -- `lifetime_value`: 客户生命周期价值 -- `win_rate`: 赢单率（百分比） -- `health_score`: 健康度评分（0-100） +**Key Fields:** +- `account_id` - Unique account identifier +- `unified_social_credit_code` - Enterprise unique identifier +- `account_type` - CUSTOMER, PARTNER, COMPETITOR, PROSPECT +- `account_status` - ACTIVE, DORMANT, CHURNED, BLACKLIST +- `account_level` - STRATEGIC, IMPORTANT, NORMAL +- `lifecycle_stage` - AWARENESS, CONSIDERATION, DECISION, RETENTION, EXPANSION + +**AccountSummary Metrics:** +- `total_contacts` - Number of associated contacts +- `total_revenue` - Cumulative revenue +- `lifetime_value` - Customer lifetime value +- `win_rate` - Win rate percentage +- `health_score` - Health score (0-100) --- -### 2. Contact Entity +### Contact Entity Model ```mermaid erDiagram @@ -239,14 +159,9 @@ erDiagram varchar department varchar contact_status varchar primary_account_id FK - varchar contact_source - varchar primary_channel_id FK varchar owner_user_id FK - datetime created_at - datetime updated_at - varchar lifecycle_stage boolean is_decision_maker - boolean is_verified + datetime created_at } ContactChannelIdentity { @@ -254,25 +169,19 @@ erDiagram varchar contact_id FK varchar channel_id FK varchar channel_user_id - varchar identity_type - boolean is_primary boolean is_verified datetime first_seen_at - datetime last_seen_at } ContactSummary { varchar summary_id PK - varchar contact_id FK UK + varchar contact_id FK int total_touchpoints int total_events int email_opens int email_clicks - int form_submissions - int content_downloads int engagement_score datetime last_activity_at - int days_since_last_activity datetime calculated_at } @@ -283,24 +192,21 @@ erDiagram varchar role_in_account varchar decision_level boolean is_primary_contact - varchar relationship_status - date relation_start_date - datetime created_at } ``` -**Contact 核心字段说明:** -- `decision_level`: 决策层级（DECISION_MAKER, INFLUENCER, USER, GATEKEEPER） -- `lifecycle_stage`: 生命周期阶段（SUBSCRIBER, LEAD, MQL, SQL, OPPORTUNITY, CUSTOMER） +**Key Fields:** +- `decision_level` - DECISION_MAKER, INFLUENCER, USER, GATEKEEPER +- `lifecycle_stage` - SUBSCRIBER, LEAD, MQL, SQL, OPPORTUNITY, CUSTOMER -**ContactSummary 核心指标:** -- `engagement_score`: 参与度评分（0-100） -- `email_opens`: 邮件打开次数 -- `days_since_last_activity`: 距上次活跃天数 +**ContactSummary Metrics:** +- `engagement_score` - Engagement score (0-100) +- `email_opens` - Email open count +- `days_since_last_activity` - Days since last activity --- -### 3. Lead Entity +### Lead Entity Model ```mermaid erDiagram @@ -313,26 +219,17 @@ erDiagram varchar company_name varchar mobile_phone varchar email - varchar wechat_id - varchar job_title varchar lead_source varchar channel_id FK varchar campaign_id FK varchar lead_status int lead_score varchar lead_grade - varchar industry_id FK - varchar province - varchar city varchar owner_user_id FK datetime created_at - datetime updated_at - datetime last_contacted_at datetime converted_at varchar converted_contact_id FK varchar converted_account_id FK - varchar converted_opportunity_id FK - boolean is_qualified } LeadChannelIdentity { @@ -345,28 +242,26 @@ erDiagram LeadSummary { varchar summary_id PK - varchar lead_id FK UK + varchar lead_id FK int total_touchpoints int total_events int form_submissions int content_downloads - int page_views int days_in_pipeline - int contact_attempts datetime last_activity_at datetime calculated_at } ``` -**Lead 核心字段说明:** -- `lead_status`: 线索状态（NEW, CONTACTED, QUALIFIED, CONVERTED, DISQUALIFIED） -- `lead_score`: 线索评分（0-100） -- `lead_grade`: 线索等级（A, B, C, D） -- `is_qualified`: 是否为合格线索（MQL/SQL） +**Key Fields:** +- `lead_status` - NEW, CONTACTED, QUALIFIED, CONVERTED, DISQUALIFIED +- `lead_score` - Lead score (0-100) +- `lead_grade` - A, B, C, D +- `is_qualified` - MQL/SQL indicator --- -### 4. Opportunity Entity +### Opportunity Entity Model ```mermaid erDiagram @@ -380,19 +275,13 @@ erDiagram varchar primary_contact_id FK varchar lead_id FK varchar opportunity_type - varchar opportunity_source decimal amount - varchar currency varchar stage int probability date expected_close_date date actual_close_date - varchar close_reason varchar owner_user_id FK - varchar campaign_id FK datetime created_at - datetime updated_at - int days_in_stage boolean is_won boolean is_lost } @@ -403,8 +292,6 @@ erDiagram varchar from_stage varchar to_stage datetime changed_at - varchar changed_by_user_id FK - varchar change_reason int duration_days } @@ -415,18 +302,17 @@ erDiagram int quantity decimal unit_price decimal total_price - decimal discount } ``` -**Opportunity 核心字段说明:** -- `stage`: 商机阶段（LEAD, QUALIFICATION, NEEDS_ANALYSIS, PROPOSAL, NEGOTIATION, CONTRACT, CLOSED_WON, CLOSED_LOST） -- `probability`: 赢单概率（0-100） -- `opportunity_type`: 商机类型（NEW_BUSINESS, UPSELL, RENEWAL, CROSS_SELL） +**Key Fields:** +- `stage` - LEAD, QUALIFICATION, NEEDS_ANALYSIS, PROPOSAL, NEGOTIATION, CONTRACT, CLOSED_WON, CLOSED_LOST +- `probability` - Win probability (0-100) +- `opportunity_type` - NEW_BUSINESS, UPSELL, RENEWAL, CROSS_SELL --- -### 5. Channel Entity +### Channel Entity Model ```mermaid erDiagram @@ -441,7 +327,6 @@ erDiagram varchar channel_status decimal cost datetime created_at - datetime updated_at } ChannelPerformance { @@ -452,24 +337,20 @@ erDiagram int contact_count int opportunity_count decimal revenue - decimal cost decimal roi - int conversion_count decimal conversion_rate - datetime created_at } ``` -**渠道类型 (channel_type):** +**Channel Types:** - WEBSITE, SEO, SEM - WECHAT, ENTERPRISE_WECHAT, DOUYIN - EMAIL, PHONE -- OFFLINE_EVENT, EXHIBITION -- PARTNER +- OFFLINE_EVENT, EXHIBITION, PARTNER --- -### 6. Campaign Entity +### Campaign Entity Model ```mermaid erDiagram @@ -487,7 +368,6 @@ erDiagram decimal actual_cost varchar owner_user_id FK datetime created_at - datetime updated_at } CampaignPerformance { @@ -497,12 +377,10 @@ erDiagram int impressions int clicks int leads_generated - int opportunities_generated decimal revenue decimal roi decimal cpl decimal cpa - datetime created_at } CampaignMember { @@ -512,22 +390,12 @@ erDiagram varchar member_ref_id FK varchar member_status datetime joined_at - datetime responded_at - varchar response_status } ``` -**活动类型 (campaign_type):** -- WEBINAR - 网络研讨会 -- CONFERENCE - 线下会议 -- EXHIBITION - 展会 -- EMAIL_MARKETING - 邮件营销 -- CONTENT_MARKETING - 内容营销 -- PRODUCT_TRIAL - 产品试用 - --- -### 7. Touchpoint Entity +### Touchpoint Entity Model ```mermaid erDiagram @@ -543,12 +411,8 @@ erDiagram varchar account_id FK datetime touchpoint_time varchar touchpoint_direction - varchar touchpoint_status varchar content_type - varchar content_id FK - varchar subject int duration_seconds - varchar owner_user_id FK datetime created_at } @@ -559,19 +423,17 @@ erDiagram varchar file_url varchar file_type int file_size - datetime uploaded_at } ``` -**触点类型 (touchpoint_type):** +**Touchpoint Types:** - PAGE_VIEW, FORM_SUBMIT, DOWNLOAD - EMAIL_OPEN, EMAIL_CLICK - CALL, MEETING, CHAT -- SOCIAL_INTERACTION --- -### 8. Event Entity +### Event Entity Model ```mermaid erDiagram @@ -587,15 +449,13 @@ erDiagram varchar session_id varchar device_type varchar browser - varchar os - varchar ip_address varchar page_url varchar referrer_url datetime created_at } ``` -**事件类型 (event_type):** +**Event Types:** - PAGE_VIEW, BUTTON_CLICK - FORM_START, FORM_SUBMIT - FILE_DOWNLOAD, VIDEO_PLAY @@ -603,7 +463,7 @@ erDiagram --- -### 9. Product Entity +### Product Entity Model ```mermaid erDiagram @@ -614,13 +474,10 @@ erDiagram varchar product_name varchar product_code UK varchar product_category_id FK - varchar product_type varchar product_status decimal list_price varchar currency - varchar pricing_model datetime created_at - datetime updated_at } ProductCategory { @@ -629,13 +486,12 @@ erDiagram varchar parent_category_id FK int level int sort_order - datetime created_at } ``` --- -### 10. Tag Entity +### Tag Entity Model ```mermaid erDiagram @@ -647,9 +503,7 @@ erDiagram varchar tag_category varchar tag_type varchar description - varchar color datetime created_at - varchar created_by_user_id FK } TagRelation { @@ -658,21 +512,19 @@ erDiagram varchar entity_type varchar entity_id FK datetime tagged_at - varchar tagged_by_user_id FK boolean is_auto_tagged - varchar tag_source } ``` -**标签类型 (tag_type):** -- BEHAVIOR - 行为标签 -- PROFILE - 画像标签 -- BUSINESS - 业务标签 -- INTEREST - 兴趣标签 +**Tag Types:** +- BEHAVIOR - Behavioral tags +- PROFILE - Profile tags +- BUSINESS - Business tags +- INTEREST - Interest tags --- -### 11. Segment Entity +### Segment Entity Model ```mermaid erDiagram @@ -686,9 +538,7 @@ erDiagram int member_count boolean is_dynamic datetime last_calculated_at - varchar created_by_user_id FK datetime created_at - datetime updated_at } SegmentMember { @@ -697,14 +547,13 @@ erDiagram varchar entity_type varchar entity_id FK datetime joined_at - datetime left_at boolean is_active } ``` --- -### 12. Score Entity +### Score Entity Model ```mermaid erDiagram @@ -719,7 +568,6 @@ erDiagram int max_score varchar status datetime created_at - datetime updated_at } ScoreRecord { @@ -730,7 +578,6 @@ erDiagram int score varchar grade datetime calculated_at - datetime updated_at } ScoreHistory { @@ -743,14 +590,14 @@ erDiagram } ``` -**评分类型 (model_type):** -- LEAD_SCORING - 线索评分 -- ACCOUNT_HEALTH - 企业健康度评分 -- CONTACT_ENGAGEMENT - 联系人参与度评分 +**Score Types:** +- LEAD_SCORING - Lead scoring +- ACCOUNT_HEALTH - Account health scoring +- CONTACT_ENGAGEMENT - Contact engagement scoring --- -### 13. Industry Entity +### Industry Entity Model ```mermaid erDiagram @@ -763,13 +610,12 @@ erDiagram varchar parent_industry_id FK int level int sort_order - datetime created_at } ``` --- -### 14. Attribution Entity +### Attribution Entity Model ```mermaid erDiagram @@ -781,7 +627,6 @@ erDiagram varchar entity_id FK varchar attribution_model datetime created_at - datetime updated_at } TouchpointAttribution { @@ -792,21 +637,20 @@ erDiagram varchar channel_id FK decimal attribution_weight int position_in_journey - datetime touchpoint_time } ``` -**归因模型 (attribution_model):** -- FIRST_TOUCH - 首次触点归因 -- LAST_TOUCH - 末次触点归因 -- LINEAR - 线性归因 -- TIME_DECAY - 时间衰减归因 -- U_SHAPED - U型归因 -- W_SHAPED - W型归因 +**Attribution Models:** +- FIRST_TOUCH - First touch attribution +- LAST_TOUCH - Last touch attribution +- LINEAR - Linear attribution +- TIME_DECAY - Time decay attribution +- U_SHAPED - U-shaped attribution +- W_SHAPED - W-shaped attribution --- -### 15. CustomerJourney Entity +### CustomerJourney Entity Model ```mermaid erDiagram @@ -821,9 +665,7 @@ erDiagram datetime journey_start_at datetime journey_end_at int total_touchpoints - int journey_duration_days datetime created_at - datetime updated_at } JourneyStage { @@ -831,15 +673,14 @@ erDiagram varchar stage_name int stage_order varchar stage_category - datetime created_at } ``` --- -## 实体关系图 +## Entity Relationship -### 核心实体关系总览 +### Core Entity Relationships ```mermaid erDiagram @@ -847,22 +688,18 @@ erDiagram Account ||--|| AccountSummary : aggregates Account ||--o{ Opportunity : owns Account ||--o{ Touchpoint : receives - Account ||--o{ AccountChannelIdentity : identified_by - Contact ||--o{ AccountContactRelation : belongs_to + Contact ||--o{ AccountContactRelation : belongs Contact ||--|| ContactSummary : aggregates Contact ||--o{ Opportunity : participates Contact ||--o{ Touchpoint : receives Contact ||--o{ Event : generates - Contact ||--o{ ContactChannelIdentity : identified_by Lead ||--o| Contact : converts_to Lead ||--o| Account : converts_to Lead ||--o| Opportunity : converts_to Lead ||--|| LeadSummary : aggregates Lead ||--o{ Touchpoint : receives - Lead ||--o{ Event : generates - Lead ||--o{ LeadChannelIdentity : identified_by Opportunity ||--o{ OpportunityStageHistory : tracks Opportunity ||--o{ OpportunityProduct : contains @@ -873,95 +710,57 @@ erDiagram Channel ||--o{ Touchpoint : generates Channel ||--o{ Event : tracks Channel ||--o{ ChannelPerformance : measured_by - Channel ||--o{ AccountChannelIdentity : identifies - Channel ||--o{ ContactChannelIdentity : identifies - Channel ||--o{ LeadChannelIdentity : identifies Campaign ||--o{ Touchpoint : drives Campaign ||--o{ Lead : generates Campaign ||--o{ CampaignMember : includes - Campaign ||--o{ CampaignPerformance : measured_by Tag ||--o{ TagRelation : tags Segment ||--o{ SegmentMember : groups ScoreModel ||--o{ ScoreRecord : scores - ScoreRecord ||--o{ ScoreHistory : tracks Industry ||--o{ Account : classifies Industry ||--o{ Lead : classifies - Industry ||--o{ Industry : parent_of Attribution ||--o{ TouchpointAttribution : attributes - Touchpoint ||--o{ TouchpointAttribution : contributes - - CustomerJourney ||--o{ JourneyStage : progresses ``` --- -### 全渠道身份统一架构 +### Identity Resolution Architecture ```mermaid graph TB - subgraph Sources[Multi-Channel Data Sources] - S1[WeChat Official] - S2[Enterprise WeChat] - S3[Website] - S4[Douyin] - S5[Email System] - S6[Phone System] - S7[Offline Events] - end - - subgraph Identity[Identity Resolution Engine] - IR[Identity Matching Rules] - IG[Identity Grouping Logic] - end - - subgraph Entities[Unified Customer Entities] - UC[Contact
Unified Contact ID] - UA[Account
Unified Account ID] - end - - subgraph Mapping[Channel Identity Mapping] - CI1[ContactChannelIdentity
WeChat OpenID] - CI2[ContactChannelIdentity
Enterprise WeChat UserID] - CI3[ContactChannelIdentity
Website Cookie] - CI4[ContactChannelIdentity
Douyin UserID] - AI1[AccountChannelIdentity
Enterprise WeChat CorpID] - AI2[AccountChannelIdentity
Website Company ID] - end - - S1 --> IR - S2 --> IR - S3 --> IR - S4 --> IR - S5 --> IR - S6 --> IR - S7 --> IR - - IR --> IG - IG --> UC - IG --> UA - - UC --> CI1 - UC --> CI2 - UC --> CI3 - UC --> CI4 - UA --> AI1 - UA --> AI2 + S1[WeChat Official] --> IR[Identity Resolution Engine] + S2[Enterprise WeChat] --> IR + S3[Website] --> IR + S4[Douyin] --> IR + S5[Email] --> IR + S6[Phone] --> IR + S7[Offline Events] --> IR + + IR --> IM[Identity Matching] + IM --> UC[Unified Contact] + IM --> UA[Unified Account] + + UC --> CI1[ContactChannelIdentity WeChat] + UC --> CI2[ContactChannelIdentity Website] + UC --> CI3[ContactChannelIdentity Email] + + UA --> AI1[AccountChannelIdentity CorpID] + UA --> AI2[AccountChannelIdentity WebsiteID] ``` --- -## 业务流程 +## Business Processes -### 线索到商机转化流程 +### Lead to Opportunity Conversion ```mermaid stateDiagram-v2 - [*] --> NewLead: Lead Capture + [*] --> NewLead NewLead --> Contacted: First Contact NewLead --> Invalid: Mark Invalid @@ -975,9 +774,6 @@ stateDiagram-v2 [*] --> CreateContact [*] --> CreateAccount [*] --> CreateOpportunity - CreateContact --> [*] - CreateAccount --> [*] - CreateOpportunity --> [*] } Converted --> [*] @@ -986,65 +782,57 @@ stateDiagram-v2 --- -### 商机阶段流转流程 +### Opportunity Stage Progression ```mermaid stateDiagram-v2 - [*] --> Lead: Lead Stage + [*] --> LeadStage - Lead --> Qualification: Needs Analysis - Lead --> Lost: Lost + LeadStage --> Qualification + LeadStage --> Lost - Qualification --> SolutionDesign: Solution Design - Qualification --> Lost: Lost + Qualification --> SolutionDesign + Qualification --> Lost - SolutionDesign --> Negotiation: Negotiation - SolutionDesign --> Lost: Lost + SolutionDesign --> Negotiation + SolutionDesign --> Lost - Negotiation --> Contract: Sign Contract - Negotiation --> Lost: Lost + Negotiation --> Contract + Negotiation --> Lost - Contract --> Won: Close Won - Contract --> Lost: Close Lost + Contract --> Won + Contract --> Lost Won --> [*] Lost --> [*] - - note right of Won - Update AccountSummary - - total_revenue - - won_opportunities - - win_rate - - health_score - end note ``` --- -### 客户生命周期管理 +### Customer Lifecycle ```mermaid stateDiagram-v2 - [*] --> Awareness: Awareness Stage + [*] --> Awareness - Awareness --> Consideration: Show Interest - Consideration --> Decision: Evaluation - Decision --> Retention: Purchase - Retention --> Expansion: Upsell/Cross-sell + Awareness --> Consideration + Consideration --> Decision + Decision --> Retention + Retention --> Expansion - Retention --> Churn: Inactivity - Expansion --> Churn: Inactivity + Retention --> Churn + Expansion --> Churn - Churn --> Winback: Winback Campaign - Winback --> Retention: Success - Winback --> [*]: Permanent Loss + Churn --> Winback + Winback --> Retention + Winback --> [*] - Expansion --> [*]: Ongoing + Expansion --> [*] ``` --- -### 全渠道数据流转流程 +### Multi-Channel Data Flow ```mermaid sequenceDiagram @@ -1052,484 +840,444 @@ sequenceDiagram participant Channel participant Event participant Identity - participant CDP participant Lead participant Contact participant Account participant Summary - User->>Channel: 1. Interaction - Channel->>Event: 2. Log Event - Event->>Identity: 3. Identity Resolution + User->>Channel: Interaction + Channel->>Event: Log Event + Event->>Identity: Identify User alt New User - Identity->>Lead: 4a. Create Lead - Lead->>CDP: 5a. Save Data - CDP->>Summary: 6a. Update LeadSummary - else Identified User - Identity->>Contact: 4b. Link Contact - Contact->>Account: 5b. Link Account - Account->>CDP: 6b. Save Data - CDP->>Summary: 7b. Update Summary + Identity->>Lead: Create Lead + Lead->>Summary: Update LeadSummary + else Known User + Identity->>Contact: Link Contact + Contact->>Account: Link Account + Account->>Summary: Update Summary end - Summary->>CDP: 8. Trigger Rules - CDP->>Channel: 9. Personalized Response - Channel->>User: 10. Return Content + Summary->>Channel: Trigger Response + Channel->>User: Return Content ``` --- -### 身份识别匹配流程 +### Identity Matching Process ```mermaid flowchart TD - Start[Receive Multi-Channel Data] --> Extract[Extract Identity Identifiers] - - Extract --> Phone{Has
Phone?} - Extract --> Email{Has
Email?} - Extract --> Wechat{Has
WeChat ID?} - Extract --> Corp{Has
Corp Code?} - - Phone -->|Yes| MatchPhone[Match by Phone
Priority: 1] - Email -->|Yes| MatchEmail[Match by Email
Priority: 2] - Wechat -->|Yes| MatchWechat[Match by WeChat
Priority: 3] - Corp -->|Yes| MatchCorp[Match by Corp Code
Priority: 1] - - MatchPhone --> FoundContact{Found
Contact?} - MatchEmail --> FoundContact - MatchWechat --> FoundContact + Start[Receive Data] --> Extract[Extract Identifiers] - MatchCorp --> FoundAccount{Found
Account?} + Extract --> Phone{Has Phone?} + Extract --> Email{Has Email?} + Extract --> WeChat{Has WeChat?} - FoundContact -->|Yes| MergeContact[Merge to Existing Contact] - FoundContact -->|No| CreateContact[Create New Contact] + Phone -->|Yes| MatchPhone[Match by Phone Priority 1] + Email -->|Yes| MatchEmail[Match by Email Priority 2] + WeChat -->|Yes| MatchWeChat[Match by WeChat Priority 3] - FoundAccount -->|Yes| MergeAccount[Merge to Existing Account] - FoundAccount -->|No| CreateAccount[Create New Account] + MatchPhone --> Found{Found?} + MatchEmail --> Found + MatchWeChat --> Found - MergeContact --> AddIdentity[Add/Update
ContactChannelIdentity] - CreateContact --> AddIdentity + Found -->|Yes| Merge[Merge to Existing] + Found -->|No| Create[Create New] - MergeAccount --> AddAccountIdentity[Add/Update
AccountChannelIdentity] - CreateAccount --> AddAccountIdentity - - AddIdentity --> UpdateSummary[Update ContactSummary] - AddAccountIdentity --> UpdateAccountSummary[Update AccountSummary] + Merge --> AddIdentity[Add Channel Identity] + Create --> AddIdentity + AddIdentity --> UpdateSummary[Update Summary] UpdateSummary --> End[Complete] - UpdateAccountSummary --> End ``` --- -### 汇总数据更新策略 +### Summary Update Strategy ```mermaid flowchart LR - subgraph Triggers[Update Triggers] - T1[Scheduled Job
Hourly] - T2[Real-time Trigger
Critical Events] - T3[Manual Refresh
On Demand] - end - - subgraph Engine[Calculation Engine] - C1[Aggregation] - C2[Scoring] - C3[Rule Engine] - end - - subgraph Summary[Summary Tables] - S1[AccountSummary] - S2[ContactSummary] - S3[LeadSummary] - end + T1[Scheduled Job Hourly] --> Engine[Calculation Engine] + T2[Real-time Trigger] --> Engine + T3[Manual Refresh] --> Engine - T1 --> C1 - T2 --> C2 - T3 --> C3 - - C1 --> S1 - C1 --> S2 - C1 --> S3 - C2 --> S1 - C2 --> S2 - C2 --> S3 - C3 --> S1 - C3 --> S2 - C3 --> S3 + Engine --> S1[AccountSummary] + Engine --> S2[ContactSummary] + Engine --> S3[LeadSummary] ``` --- -## 数据模型详细设计 - -### Account 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| account_id | VARCHAR | 64 | PK, NOT NULL | 账户唯一标识 | ACC_20231105001 | -| account_name | VARCHAR | 200 | NOT NULL | 企业名称 | 阿里巴巴网络技术有限公司 | -| unified_social_credit_code | VARCHAR | 18 | UNIQUE | 统一社会信用代码 | 91330000MA27XYZ123 | -| account_type | VARCHAR | 50 | NOT NULL | 客户类型 | CUSTOMER, PARTNER, COMPETITOR, PROSPECT | -| industry_id | VARCHAR | 64 | FK | 行业分类外键 | IND_001 | -| account_status | VARCHAR | 50 | NOT NULL | 账户状态 | ACTIVE, DORMANT, CHURNED, BLACKLIST | -| account_level | VARCHAR | 50 | | 客户等级 | STRATEGIC, IMPORTANT, NORMAL | -| annual_revenue | DECIMAL | 18,2 | | 年营收(万元) | 50000.00 | -| employee_count | INT | | | 员工人数 | 5000 | -| company_website | VARCHAR | 500 | | 公司网站 | https://www.alibaba.com | -| province | VARCHAR | 50 | | 省份 | 浙江省 | -| city | VARCHAR | 50 | | 城市 | 杭州市 | -| account_source | VARCHAR | 100 | | 来源渠道 | WEBSITE, EXHIBITION, PARTNER, COLD_CALL | -| primary_channel_id | VARCHAR | 64 | FK | 主渠道ID | CH_001 | -| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | -| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | -| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | AWARENESS, CONSIDERATION, DECISION, RETENTION, EXPANSION | +## Data Model Design + +### Account Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| account_id | VARCHAR | 64 | PK, NOT NULL | Unique account ID | ACC_20231105001 | +| account_name | VARCHAR | 200 | NOT NULL | Account name | Alibaba Group | +| unified_social_credit_code | VARCHAR | 18 | UNIQUE | Unified credit code | 91330000MA27XYZ123 | +| account_type | VARCHAR | 50 | NOT NULL | Account type | CUSTOMER, PARTNER, PROSPECT | +| industry_id | VARCHAR | 64 | FK | Industry ID | IND_001 | +| account_status | VARCHAR | 50 | NOT NULL | Account status | ACTIVE, DORMANT, CHURNED | +| account_level | VARCHAR | 50 | | Account level | STRATEGIC, IMPORTANT, NORMAL | +| annual_revenue | DECIMAL | 18,2 | | Annual revenue | 50000.00 | +| employee_count | INT | | | Employee count | 5000 | +| company_website | VARCHAR | 500 | | Website URL | https://www.alibaba.com | +| province | VARCHAR | 50 | | Province | Zhejiang | +| city | VARCHAR | 50 | | City | Hangzhou | +| account_source | VARCHAR | 100 | | Source | WEBSITE, EXHIBITION, PARTNER | +| primary_channel_id | VARCHAR | 64 | FK | Primary channel | CH_001 | +| owner_user_id | VARCHAR | 64 | FK | Owner user ID | USER_001 | +| created_at | DATETIME | | NOT NULL | Created timestamp | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | Updated timestamp | 2023-11-05 15:20:00 | +| lifecycle_stage | VARCHAR | 50 | | Lifecycle stage | AWARENESS, RETENTION, EXPANSION | --- -### AccountSummary 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | ACCS_20231105001 | -| account_id | VARCHAR | 64 | FK, UNIQUE, NOT NULL | 账户ID | ACC_20231105001 | -| total_contacts | INT | | DEFAULT 0 | 关联联系人总数 | 25 | -| total_opportunities | INT | | DEFAULT 0 | 商机总数 | 8 | -| total_leads | INT | | DEFAULT 0 | 线索总数 | 45 | -| total_revenue | DECIMAL | 18,2 | DEFAULT 0 | 累计收入(元) | 5000000.00 | -| lifetime_value | DECIMAL | 18,2 | DEFAULT 0 | 生命周期价值(元) | 8000000.00 | -| won_opportunities | INT | | DEFAULT 0 | 赢单数量 | 5 | -| lost_opportunities | INT | | DEFAULT 0 | 输单数量 | 2 | -| win_rate | DECIMAL | 5,2 | DEFAULT 0 | 赢单率(%) | 71.43 | -| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 156 | -| health_score | INT | | DEFAULT 0 | 健康度评分(0-100) | 85 | -| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | -| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | -| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | +### AccountSummary Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | Summary ID | ACCS_20231105001 | +| account_id | VARCHAR | 64 | FK, UNIQUE | Account ID | ACC_20231105001 | +| total_contacts | INT | | DEFAULT 0 | Total contacts | 25 | +| total_opportunities | INT | | DEFAULT 0 | Total opportunities | 8 | +| total_leads | INT | | DEFAULT 0 | Total leads | 45 | +| total_revenue | DECIMAL | 18,2 | DEFAULT 0 | Total revenue | 5000000.00 | +| lifetime_value | DECIMAL | 18,2 | DEFAULT 0 | Lifetime value | 8000000.00 | +| won_opportunities | INT | | DEFAULT 0 | Won count | 5 | +| lost_opportunities | INT | | DEFAULT 0 | Lost count | 2 | +| win_rate | DECIMAL | 5,2 | DEFAULT 0 | Win rate % | 71.43 | +| total_touchpoints | INT | | DEFAULT 0 | Total touchpoints | 156 | +| health_score | INT | | DEFAULT 0 | Health score 0-100 | 85 | +| last_activity_at | DATETIME | | | Last activity time | 2023-11-05 14:30:00 | +| calculated_at | DATETIME | | NOT NULL | Calculated time | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 16:00:00 | --- -### Contact 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| contact_id | VARCHAR | 64 | PK, NOT NULL | 联系人唯一标识 | CNT_20231105001 | -| contact_name | VARCHAR | 100 | NOT NULL | 联系人姓名 | 张伟 | -| mobile_phone | VARCHAR | 20 | UNIQUE | 手机号 | 13800138000 | -| email | VARCHAR | 200 | UNIQUE | 邮箱地址 | zhangwei@company.com | -| wechat_id | VARCHAR | 100 | | 微信ID | wx_zhangwei | -| job_title | VARCHAR | 100 | | 职位 | 首席技术官 CTO | -| department | VARCHAR | 100 | | 部门 | 技术部 | -| contact_status | VARCHAR | 50 | NOT NULL | 联系人状态 | ACTIVE, INACTIVE, BOUNCED, UNSUBSCRIBED | -| primary_account_id | VARCHAR | 64 | FK | 主要关联企业ID | ACC_20231105001 | -| contact_source | VARCHAR | 100 | | 来源 | WEBSITE, FORM, IMPORT, API, REFERRAL | -| primary_channel_id | VARCHAR | 64 | FK | 主渠道ID | CH_001 | -| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | -| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | -| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | SUBSCRIBER, LEAD, MQL, SQL, OPPORTUNITY, CUSTOMER | -| is_decision_maker | BOOLEAN | | DEFAULT FALSE | 是否决策者 | true | -| is_verified | BOOLEAN | | DEFAULT FALSE | 是否已验证 | true | +### Contact Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| contact_id | VARCHAR | 64 | PK, NOT NULL | Contact ID | CNT_20231105001 | +| contact_name | VARCHAR | 100 | NOT NULL | Contact name | Zhang Wei | +| mobile_phone | VARCHAR | 20 | UNIQUE | Mobile phone | 13800138000 | +| email | VARCHAR | 200 | UNIQUE | Email address | zhangwei@company.com | +| wechat_id | VARCHAR | 100 | | WeChat ID | wx_zhangwei | +| job_title | VARCHAR | 100 | | Job title | CTO | +| department | VARCHAR | 100 | | Department | Technology | +| contact_status | VARCHAR | 50 | NOT NULL | Status | ACTIVE, INACTIVE, UNSUBSCRIBED | +| primary_account_id | VARCHAR | 64 | FK | Primary account | ACC_20231105001 | +| contact_source | VARCHAR | 100 | | Source | WEBSITE, FORM, IMPORT | +| primary_channel_id | VARCHAR | 64 | FK | Primary channel | CH_001 | +| owner_user_id | VARCHAR | 64 | FK | Owner user | USER_001 | +| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 15:20:00 | +| lifecycle_stage | VARCHAR | 50 | | Lifecycle stage | LEAD, MQL, SQL, CUSTOMER | +| is_decision_maker | BOOLEAN | | DEFAULT FALSE | Is decision maker | true | +| is_verified | BOOLEAN | | DEFAULT FALSE | Is verified | true | --- -### ContactSummary 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | CNTS_20231105001 | -| contact_id | VARCHAR | 64 | FK, UNIQUE, NOT NULL | 联系人ID | CNT_20231105001 | -| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 87 | -| total_events | INT | | DEFAULT 0 | 总事件数 | 234 | -| email_opens | INT | | DEFAULT 0 | 邮件打开次数 | 45 | -| email_clicks | INT | | DEFAULT 0 | 邮件点击次数 | 23 | -| form_submissions | INT | | DEFAULT 0 | 表单提交次数 | 12 | -| content_downloads | INT | | DEFAULT 0 | 内容下载次数 | 8 | -| engagement_score | INT | | DEFAULT 0 | 参与度评分(0-100) | 78 | -| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | -| days_since_last_activity | INT | | DEFAULT 0 | 距上次活跃天数 | 1 | -| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | -| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | +### ContactSummary Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | Summary ID | CNTS_20231105001 | +| contact_id | VARCHAR | 64 | FK, UNIQUE | Contact ID | CNT_20231105001 | +| total_touchpoints | INT | | DEFAULT 0 | Total touchpoints | 87 | +| total_events | INT | | DEFAULT 0 | Total events | 234 | +| email_opens | INT | | DEFAULT 0 | Email opens | 45 | +| email_clicks | INT | | DEFAULT 0 | Email clicks | 23 | +| form_submissions | INT | | DEFAULT 0 | Form submissions | 12 | +| content_downloads | INT | | DEFAULT 0 | Content downloads | 8 | +| engagement_score | INT | | DEFAULT 0 | Engagement score 0-100 | 78 | +| last_activity_at | DATETIME | | | Last activity | 2023-11-05 14:30:00 | +| days_since_last_activity | INT | | DEFAULT 0 | Days since activity | 1 | +| calculated_at | DATETIME | | NOT NULL | Calculated time | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 16:00:00 | --- -### Lead 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| lead_id | VARCHAR | 64 | PK, NOT NULL | 线索唯一标识 | LEAD_20231105001 | -| lead_name | VARCHAR | 100 | NOT NULL | 线索姓名 | 李明 | -| company_name | VARCHAR | 200 | | 公司名称 | 腾讯科技有限公司 | -| mobile_phone | VARCHAR | 20 | | 手机号 | 13900139000 | -| email | VARCHAR | 200 | | 邮箱地址 | liming@company.com | -| wechat_id | VARCHAR | 100 | | 微信ID | wx_liming | -| job_title | VARCHAR | 100 | | 职位 | 产品经理 | -| lead_source | VARCHAR | 100 | NOT NULL | 线索来源 | WEBSITE, FORM, CAMPAIGN, COLD_CALL, REFERRAL | -| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | -| campaign_id | VARCHAR | 64 | FK | 营销活动ID | CMP_001 | -| lead_status | VARCHAR | 50 | NOT NULL | 线索状态 | NEW, CONTACTED, QUALIFIED, CONVERTED, DISQUALIFIED | -| lead_score | INT | | DEFAULT 0 | 线索评分(0-100) | 80 | -| lead_grade | VARCHAR | 10 | | 线索等级 | A, B, C, D | -| industry_id | VARCHAR | 64 | FK | 行业ID | IND_001 | -| province | VARCHAR | 50 | | 省份 | 广东省 | -| city | VARCHAR | 50 | | 城市 | 深圳市 | -| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | -| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | -| last_contacted_at | DATETIME | | | 最后联系时间 | 2023-11-05 14:00:00 | -| converted_at | DATETIME | | | 转化时间 | 2023-11-10 09:00:00 | -| converted_contact_id | VARCHAR | 64 | FK | 转化后联系人ID | CNT_20231110001 | -| converted_account_id | VARCHAR | 64 | FK | 转化后企业ID | ACC_20231110001 | -| converted_opportunity_id | VARCHAR | 64 | FK | 转化后商机ID | OPP_20231110001 | -| is_qualified | BOOLEAN | | DEFAULT FALSE | 是否为合格线索 | true | +### Lead Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| lead_id | VARCHAR | 64 | PK, NOT NULL | Lead ID | LEAD_20231105001 | +| lead_name | VARCHAR | 100 | NOT NULL | Lead name | Li Ming | +| company_name | VARCHAR | 200 | | Company name | Tencent Technology | +| mobile_phone | VARCHAR | 20 | | Mobile phone | 13900139000 | +| email | VARCHAR | 200 | | Email | liming@company.com | +| wechat_id | VARCHAR | 100 | | WeChat ID | wx_liming | +| job_title | VARCHAR | 100 | | Job title | Product Manager | +| lead_source | VARCHAR | 100 | NOT NULL | Lead source | WEBSITE, FORM, CAMPAIGN | +| channel_id | VARCHAR | 64 | FK | Channel ID | CH_001 | +| campaign_id | VARCHAR | 64 | FK | Campaign ID | CMP_001 | +| lead_status | VARCHAR | 50 | NOT NULL | Status | NEW, CONTACTED, QUALIFIED | +| lead_score | INT | | DEFAULT 0 | Lead score 0-100 | 80 | +| lead_grade | VARCHAR | 10 | | Lead grade | A, B, C, D | +| industry_id | VARCHAR | 64 | FK | Industry ID | IND_001 | +| province | VARCHAR | 50 | | Province | Guangdong | +| city | VARCHAR | 50 | | City | Shenzhen | +| owner_user_id | VARCHAR | 64 | FK | Owner user | USER_001 | +| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 15:20:00 | +| last_contacted_at | DATETIME | | | Last contact | 2023-11-05 14:00:00 | +| converted_at | DATETIME | | | Converted time | 2023-11-10 09:00:00 | +| converted_contact_id | VARCHAR | 64 | FK | Converted contact | CNT_20231110001 | +| converted_account_id | VARCHAR | 64 | FK | Converted account | ACC_20231110001 | +| converted_opportunity_id | VARCHAR | 64 | FK | Converted opportunity | OPP_20231110001 | +| is_qualified | BOOLEAN | | DEFAULT FALSE | Is qualified MQL/SQL | true | --- -### LeadSummary 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | LEADS_20231105001 | -| lead_id | VARCHAR | 64 | FK, UNIQUE, NOT NULL | 线索ID | LEAD_20231105001 | -| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 12 | -| total_events | INT | | DEFAULT 0 | 总事件数 | 45 | -| form_submissions | INT | | DEFAULT 0 | 表单提交次数 | 3 | -| content_downloads | INT | | DEFAULT 0 | 内容下载次数 | 2 | -| page_views | INT | | DEFAULT 0 | 页面浏览次数 | 28 | -| days_in_pipeline | INT | | DEFAULT 0 | 在管道中天数 | 7 | -| contact_attempts | INT | | DEFAULT 0 | 联系尝试次数 | 4 | -| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | -| last_contact_attempt_at | DATETIME | | | 最后联系尝试时间 | 2023-11-05 11:00:00 | -| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | -| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | +### LeadSummary Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | Summary ID | LEADS_20231105001 | +| lead_id | VARCHAR | 64 | FK, UNIQUE | Lead ID | LEAD_20231105001 | +| total_touchpoints | INT | | DEFAULT 0 | Total touchpoints | 12 | +| total_events | INT | | DEFAULT 0 | Total events | 45 | +| form_submissions | INT | | DEFAULT 0 | Form submissions | 3 | +| content_downloads | INT | | DEFAULT 0 | Content downloads | 2 | +| page_views | INT | | DEFAULT 0 | Page views | 28 | +| days_in_pipeline | INT | | DEFAULT 0 | Days in pipeline | 7 | +| contact_attempts | INT | | DEFAULT 0 | Contact attempts | 4 | +| last_activity_at | DATETIME | | | Last activity | 2023-11-05 14:30:00 | +| last_contact_attempt_at | DATETIME | | | Last contact attempt | 2023-11-05 11:00:00 | +| calculated_at | DATETIME | | NOT NULL | Calculated time | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 16:00:00 | --- -### Opportunity 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| opportunity_id | VARCHAR | 64 | PK, NOT NULL | 商机唯一标识 | OPP_20231105001 | -| opportunity_name | VARCHAR | 200 | NOT NULL | 商机名称 | 腾讯科技-AI平台采购项目 | -| account_id | VARCHAR | 64 | FK, NOT NULL | 关联企业ID | ACC_20231105001 | -| primary_contact_id | VARCHAR | 64 | FK | 主要联系人ID | CNT_20231105001 | -| lead_id | VARCHAR | 64 | FK | 来源线索ID | LEAD_20231105001 | -| opportunity_type | VARCHAR | 50 | | 商机类型 | NEW_BUSINESS, UPSELL, RENEWAL, CROSS_SELL | -| opportunity_source | VARCHAR | 100 | | 商机来源 | LEAD_CONVERSION, DIRECT_SALES, PARTNER | -| amount | DECIMAL | 18,2 | | 预计金额(元) | 1000000.00 | -| currency | VARCHAR | 10 | | 货币单位 | CNY, USD, EUR | -| stage | VARCHAR | 50 | NOT NULL | 当前阶段 | QUALIFICATION, NEEDS_ANALYSIS, PROPOSAL, NEGOTIATION, CLOSED_WON, CLOSED_LOST | -| probability | INT | | DEFAULT 0 | 赢单概率(0-100) | 60 | -| expected_close_date | DATE | | | 预计成交日期 | 2023-12-31 | -| actual_close_date | DATE | | | 实际成交日期 | 2023-12-25 | -| close_reason | VARCHAR | 200 | | 关闭原因 | 价格因素, 竞争对手, 预算不足, 成功签约 | -| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | -| campaign_id | VARCHAR | 64 | FK | 来源活动ID | CMP_001 | -| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | -| days_in_stage | INT | | DEFAULT 0 | 当前阶段停留天数 | 15 | -| is_won | BOOLEAN | | DEFAULT FALSE | 是否赢单 | false | -| is_lost | BOOLEAN | | DEFAULT FALSE | 是否输单 | false | +### Opportunity Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| opportunity_id | VARCHAR | 64 | PK, NOT NULL | Opportunity ID | OPP_20231105001 | +| opportunity_name | VARCHAR | 200 | NOT NULL | Opportunity name | Tencent AI Platform Project | +| account_id | VARCHAR | 64 | FK, NOT NULL | Account ID | ACC_20231105001 | +| primary_contact_id | VARCHAR | 64 | FK | Primary contact | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | Source lead | LEAD_20231105001 | +| opportunity_type | VARCHAR | 50 | | Type | NEW_BUSINESS, UPSELL, RENEWAL | +| opportunity_source | VARCHAR | 100 | | Source | LEAD_CONVERSION, DIRECT_SALES | +| amount | DECIMAL | 18,2 | | Amount | 1000000.00 | +| currency | VARCHAR | 10 | | Currency | CNY, USD, EUR | +| stage | VARCHAR | 50 | NOT NULL | Stage | QUALIFICATION, PROPOSAL, CONTRACT | +| probability | INT | | DEFAULT 0 | Probability 0-100 | 60 | +| expected_close_date | DATE | | | Expected close | 2023-12-31 | +| actual_close_date | DATE | | | Actual close | 2023-12-25 | +| close_reason | VARCHAR | 200 | | Close reason | Price, Competitor, Budget, Success | +| owner_user_id | VARCHAR | 64 | FK | Owner user | USER_001 | +| campaign_id | VARCHAR | 64 | FK | Campaign | CMP_001 | +| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 15:20:00 | +| days_in_stage | INT | | DEFAULT 0 | Days in stage | 15 | +| is_won | BOOLEAN | | DEFAULT FALSE | Is won | false | +| is_lost | BOOLEAN | | DEFAULT FALSE | Is lost | false | --- -### Channel 详细字段设计 +### Channel Table Schema -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| channel_id | VARCHAR | 64 | PK, NOT NULL | 渠道唯一标识 | CH_001 | -| channel_name | VARCHAR | 100 | NOT NULL | 渠道名称 | 官网-产品页 | -| channel_type | VARCHAR | 50 | NOT NULL | 渠道类型 | WEBSITE, WECHAT, DOUYIN, EMAIL, PHONE, OFFLINE, PARTNER | -| channel_category | VARCHAR | 50 | | 渠道分类 | ONLINE, OFFLINE, SOCIAL, DIRECT | -| parent_channel_id | VARCHAR | 64 | FK | 父渠道ID | CH_PARENT_001 | -| channel_status | VARCHAR | 50 | NOT NULL | 渠道状态 | ACTIVE, INACTIVE, TESTING | -| cost | DECIMAL | 18,2 | | 渠道成本(元/月) | 50000.00 | -| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| channel_id | VARCHAR | 64 | PK, NOT NULL | Channel ID | CH_001 | +| channel_name | VARCHAR | 100 | NOT NULL | Channel name | Website Product Page | +| channel_type | VARCHAR | 50 | NOT NULL | Type | WEBSITE, WECHAT, EMAIL, PHONE | +| channel_category | VARCHAR | 50 | | Category | ONLINE, OFFLINE, SOCIAL | +| parent_channel_id | VARCHAR | 64 | FK | Parent channel | CH_PARENT_001 | +| channel_status | VARCHAR | 50 | NOT NULL | Status | ACTIVE, INACTIVE, TESTING | +| cost | DECIMAL | 18,2 | | Cost per month | 50000.00 | +| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 15:20:00 | --- -### Touchpoint 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| touchpoint_id | VARCHAR | 64 | PK, NOT NULL | 触点唯一标识 | TP_20231105001 | -| touchpoint_type | VARCHAR | 50 | NOT NULL | 触点类型 | PAGE_VIEW, FORM_SUBMIT, DOWNLOAD, EMAIL, CALL, MEETING, CHAT | -| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | -| campaign_id | VARCHAR | 64 | FK | 营销活动ID | CMP_001 | -| contact_id | VARCHAR | 64 | FK | 联系人ID | CNT_20231105001 | -| lead_id | VARCHAR | 64 | FK | 线索ID | LEAD_20231105001 | -| account_id | VARCHAR | 64 | FK | 企业ID | ACC_20231105001 | -| touchpoint_time | DATETIME | | NOT NULL | 触点发生时间 | 2023-11-05 14:30:00 | -| touchpoint_direction | VARCHAR | 20 | | 触点方向 | INBOUND, OUTBOUND | -| touchpoint_status | VARCHAR | 50 | | 触点状态 | COMPLETED, SCHEDULED, CANCELLED | -| content_type | VARCHAR | 50 | | 内容类型 | WHITEPAPER, CASE_STUDY, WEBINAR, DEMO, PROPOSAL | -| content_id | VARCHAR | 64 | FK | 内容ID | CONTENT_001 | -| subject | VARCHAR | 200 | | 主题 | 产品演示会议 | -| duration_seconds | INT | | | 持续时长(秒) | 3600 | -| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | -| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 14:30:00 | +### Touchpoint Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| touchpoint_id | VARCHAR | 64 | PK, NOT NULL | Touchpoint ID | TP_20231105001 | +| touchpoint_type | VARCHAR | 50 | NOT NULL | Type | PAGE_VIEW, FORM_SUBMIT, EMAIL | +| channel_id | VARCHAR | 64 | FK | Channel | CH_001 | +| campaign_id | VARCHAR | 64 | FK | Campaign | CMP_001 | +| contact_id | VARCHAR | 64 | FK | Contact | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | Lead | LEAD_20231105001 | +| account_id | VARCHAR | 64 | FK | Account | ACC_20231105001 | +| touchpoint_time | DATETIME | | NOT NULL | Touchpoint time | 2023-11-05 14:30:00 | +| touchpoint_direction | VARCHAR | 20 | | Direction | INBOUND, OUTBOUND | +| touchpoint_status | VARCHAR | 50 | | Status | COMPLETED, SCHEDULED, CANCELLED | +| content_type | VARCHAR | 50 | | Content type | WHITEPAPER, WEBINAR, DEMO | +| content_id | VARCHAR | 64 | FK | Content ID | CONTENT_001 | +| subject | VARCHAR | 200 | | Subject | Product Demo Meeting | +| duration_seconds | INT | | | Duration seconds | 3600 | +| owner_user_id | VARCHAR | 64 | FK | Owner user | USER_001 | +| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 14:30:00 | --- -### Event 详细字段设计 - -| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | -|--------|------|------|------|------|--------| -| event_id | VARCHAR | 64 | PK, NOT NULL | 事件唯一标识 | EVT_20231105001 | -| event_name | VARCHAR | 100 | NOT NULL | 事件名称 | page_view | -| event_type | VARCHAR | 50 | NOT NULL | 事件类型 | PAGE_VIEW, CLICK, FORM_START, FORM_SUBMIT, DOWNLOAD, VIDEO_PLAY | -| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | -| contact_id | VARCHAR | 64 | FK | 联系人ID | CNT_20231105001 | -| lead_id | VARCHAR | 64 | FK | 线索ID | LEAD_20231105001 | -| account_id | VARCHAR | 64 | FK | 企业ID | ACC_20231105001 | -| event_time | DATETIME | | NOT NULL | 事件发生时间 | 2023-11-05 14:35:20 | -| session_id | VARCHAR | 64 | | 会话ID | SESSION_20231105001 | -| device_type | VARCHAR | 50 | | 设备类型 | DESKTOP, MOBILE, TABLET | -| browser | VARCHAR | 50 | | 浏览器 | Chrome, Safari, Firefox, Edge | -| os | VARCHAR | 50 | | 操作系统 | Windows 10, iOS 16, Android 13 | -| ip_address | VARCHAR | 50 | | IP地址 | 192.168.1.1 | -| page_url | VARCHAR | 1000 | | 页面URL | https://www.example.com/product/ai-platform | -| referrer_url | VARCHAR | 1000 | | 来源URL | https://www.baidu.com/s?wd=AI平台 | -| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 14:35:20 | +### Event Table Schema + +| Field | Type | Length | Constraint | Description | Example | +|-------|------|--------|------------|-------------|---------| +| event_id | VARCHAR | 64 | PK, NOT NULL | Event ID | EVT_20231105001 | +| event_name | VARCHAR | 100 | NOT NULL | Event name | page_view | +| event_type | VARCHAR | 50 | NOT NULL | Type | PAGE_VIEW, CLICK, FORM_SUBMIT | +| channel_id | VARCHAR | 64 | FK | Channel | CH_001 | +| contact_id | VARCHAR | 64 | FK | Contact | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | Lead | LEAD_20231105001 | +| account_id | VARCHAR | 64 | FK | Account | ACC_20231105001 | +| event_time | DATETIME | | NOT NULL | Event time | 2023-11-05 14:35:20 | +| session_id | VARCHAR | 64 | | Session | SESSION_20231105001 | +| device_type | VARCHAR | 50 | | Device | DESKTOP, MOBILE, TABLET | +| browser | VARCHAR | 50 | | Browser | Chrome, Safari, Firefox | +| os | VARCHAR | 50 | | OS | Windows 10, iOS 16 | +| ip_address | VARCHAR | 50 | | IP | 192.168.1.1 | +| page_url | VARCHAR | 1000 | | Page URL | https://example.com/product | +| referrer_url | VARCHAR | 1000 | | Referrer URL | https://baidu.com/search | +| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 14:35:20 | --- -## 数据字典总结 - -### 核心实体统计 - -| 实体分类 | 实体名称 | 数量 | 说明 | -|---------|---------|------|------| -| 客户主体实体 | Account, Contact, Lead | 3 | 核心客户数据 | -| 汇总数据实体 | AccountSummary, ContactSummary, LeadSummary | 3 | 统计汇总数据 | -| 业务实体 | Opportunity, Product, ProductCategory | 3 | 业务交易数据 | -| 营销实体 | Campaign, Channel | 2 | 营销活动管理 | -| 交互实体 | Touchpoint, Event | 2 | 客户互动行为 | -| 关系实体 | AccountContactRelation, AccountRelation, OpportunityProduct, CampaignMember, TagRelation, SegmentMember | 6 | 实体关系映射 | -| 身份实体 | AccountChannelIdentity, ContactChannelIdentity, LeadChannelIdentity | 3 | 全渠道身份映射 | -| 分析实体 | Segment, Tag, ScoreModel, ScoreRecord, ScoreHistory, Attribution, TouchpointAttribution, CustomerJourney, JourneyStage | 9 | 数据分析洞察 | -| 性能实体 | ChannelPerformance, CampaignPerformance | 2 | 绩效统计 | -| 历史实体 | OpportunityStageHistory | 1 | 变更历史追踪 | -| 支撑实体 | Industry, TouchpointAttachment | 2 | 基础支撑数据 | -| 合计 | | 36 | 完整覆盖B2B CDP业务场景 | +## Entity Summary + +### Entity Statistics + +| Category | Entity Name | Count | Description | +|----------|-------------|-------|-------------| +| Customer Entities | Account, Contact, Lead | 3 | Core customer data | +| Summary Entities | AccountSummary, ContactSummary, LeadSummary | 3 | Aggregated statistics | +| Business Entities | Opportunity, Product, ProductCategory | 3 | Business transaction data | +| Marketing Entities | Campaign, Channel | 2 | Marketing management | +| Interaction Entities | Touchpoint, Event | 2 | Customer interaction data | +| Relationship Entities | AccountContactRelation, AccountRelation, OpportunityProduct, CampaignMember, TagRelation, SegmentMember | 6 | Entity relationships | +| Identity Entities | AccountChannelIdentity, ContactChannelIdentity, LeadChannelIdentity | 3 | Multi-channel identity | +| Analytics Entities | Segment, Tag, ScoreModel, ScoreRecord, ScoreHistory, Attribution, TouchpointAttribution, CustomerJourney, JourneyStage | 9 | Analytics and insights | +| Performance Entities | ChannelPerformance, CampaignPerformance | 2 | Performance metrics | +| History Entities | OpportunityStageHistory | 1 | Change history tracking | +| Support Entities | Industry, TouchpointAttachment | 2 | Supporting data | +| Total | | 36 | Complete B2B CDP coverage | --- -## 汇总数据表设计说明 +## Summary Table Design -### 汇总表的设计原则 +### Design Principles -1. **性能优化** - - 避免频繁的多表JOIN和聚合计算 - - 提升客户360度视图的查询速度 - - 降低数据库负载 +**Performance Optimization** +- Avoid frequent multi-table JOINs and aggregations +- Improve query performance for 360-degree customer view +- Reduce database load -2. **业务需求** - - 快速展示客户健康度评分 - - 实时显示客户价值指标 - - 支持客户预警和监控 +**Business Requirements** +- Quickly display customer health scores +- Real-time customer value metrics +- Support customer alerting and monitoring -3. **数据一致性** - - 统一的计算口径 - - 定时批量更新保证数据准确性 - - 避免实时计算的结果不一致 +**Data Consistency** +- Unified calculation methodology +- Scheduled batch updates ensure accuracy +- Avoid real-time calculation inconsistencies -### 汇总表更新机制 +### Update Mechanisms -**更新触发条件:** -- 定时任务：每小时执行一次全量更新 -- 实时触发：关键业务事件发生时立即更新（如商机赢单、联系人新增） -- 手动刷新：管理员按需触发更新 +**Update Triggers:** +- Scheduled Jobs: Hourly full update +- Real-time Triggers: Immediate update on critical events +- Manual Refresh: On-demand administrator trigger -**更新策略:** -- 增量更新：仅更新有变化的记录 -- 全量更新：定期执行完整重算 -- 异步更新：使用消息队列避免阻塞主业务流程 +**Update Strategy:** +- Incremental Update: Only changed records +- Full Update: Periodic complete recalculation +- Async Update: Message queue to avoid blocking --- -## 技术实现建议 +## Index Design -### 索引设计建议 +### Account Table Indexes -**Account表索引:** ```sql --- 主键索引 +-- Primary key PRIMARY KEY (account_id); --- 唯一索引 +-- Unique indexes CREATE UNIQUE INDEX uk_account_credit_code ON Account(unified_social_credit_code) WHERE unified_social_credit_code IS NOT NULL; --- 业务查询索引 +-- Query indexes CREATE INDEX idx_account_name ON Account(account_name); CREATE INDEX idx_account_status ON Account(account_status); CREATE INDEX idx_account_owner ON Account(owner_user_id); CREATE INDEX idx_account_created ON Account(created_at DESC); CREATE INDEX idx_account_industry ON Account(industry_id); --- 组合索引 +-- Composite indexes CREATE INDEX idx_account_type_status ON Account(account_type, account_status); CREATE INDEX idx_account_location ON Account(province, city); -CREATE INDEX idx_account_lifecycle ON Account(lifecycle_stage, account_status); ``` -**AccountSummary表索引:** +### AccountSummary Table Indexes + ```sql --- 主键和唯一索引 +-- Primary and unique PRIMARY KEY (summary_id); CREATE UNIQUE INDEX uk_summary_account ON AccountSummary(account_id); --- 查询索引 +-- Query indexes CREATE INDEX idx_summary_health_score ON AccountSummary(health_score DESC); CREATE INDEX idx_summary_last_activity ON AccountSummary(last_activity_at DESC); CREATE INDEX idx_summary_total_revenue ON AccountSummary(total_revenue DESC); CREATE INDEX idx_summary_win_rate ON AccountSummary(win_rate DESC); ``` -**Contact表索引:** +### Contact Table Indexes + ```sql --- 主键索引 +-- Primary key PRIMARY KEY (contact_id); --- 唯一索引 +-- Unique indexes CREATE UNIQUE INDEX uk_contact_phone ON Contact(mobile_phone) WHERE mobile_phone IS NOT NULL; CREATE UNIQUE INDEX uk_contact_email ON Contact(email) WHERE email IS NOT NULL; --- 业务查询索引 +-- Query indexes CREATE INDEX idx_contact_name ON Contact(contact_name); CREATE INDEX idx_contact_account ON Contact(primary_account_id); CREATE INDEX idx_contact_status ON Contact(contact_status); CREATE INDEX idx_contact_wechat ON Contact(wechat_id); --- 组合索引 +-- Composite indexes CREATE INDEX idx_contact_phone_email ON Contact(mobile_phone, email); -CREATE INDEX idx_contact_lifecycle ON Contact(lifecycle_stage, contact_status); ``` -**Event表索引 (ClickHouse):** +### Event Table Indexes (ClickHouse) + ```sql --- 排序键设计 +-- Order by key ORDER BY (channel_id, event_time, contact_id, event_type); --- 分区键设计 +-- Partition key PARTITION BY toYYYYMM(event_time); --- 采样表达式 +-- Sample expression SAMPLE BY cityHash64(event_id); ``` -**Touchpoint表索引:** +### Touchpoint Table Indexes + ```sql --- 主键索引 +-- Primary key PRIMARY KEY (touchpoint_id); --- 查询索引 +-- Query indexes CREATE INDEX idx_touchpoint_contact ON Touchpoint(contact_id, touchpoint_time DESC); CREATE INDEX idx_touchpoint_lead ON Touchpoint(lead_id, touchpoint_time DESC); CREATE INDEX idx_touchpoint_account ON Touchpoint(account_id, touchpoint_time DESC); @@ -1541,32 +1289,32 @@ CREATE INDEX idx_touchpoint_type ON Touchpoint(touchpoint_type, touchpoint_time --- -## 总结 - -本B2B CDP实体设计方案提供: - -**核心能力** -- 36个实体，完整覆盖B2B CDP业务场景 -- 全渠道身份映射方案，支持跨渠道客户识别 -- 汇总数据表设计，优化查询性能 -- 完整的客户生命周期管理 -- 从线索到商机的完整转化流程 - -**关键特性** -- 灵活的标签和分群能力 -- 多维度的归因分析能力 -- 详细的字段设计和数据字典 -- 专业的数据库架构建议 -- 完整的索引优化方案 - -**适用场景** -- 全渠道客户数据整合 -- 客户360度画像分析 -- 精准营销和客户分群 -- 销售线索管理和转化 -- 客户旅程分析 -- 营销归因分析 -- 客户价值评估 -- 客户健康度监控 - -根据实际业务需要，可以选择性实现部分实体，并在后续迭代中逐步完善。 +## Conclusion + +This B2B CDP entity design provides: + +**Core Capabilities** +- 36 entities covering complete B2B CDP scenarios +- Multi-channel identity resolution +- Summary tables for performance optimization +- Complete customer lifecycle management +- Lead to opportunity conversion tracking + +**Key Features** +- Flexible tagging and segmentation +- Multi-dimensional attribution analysis +- Detailed field design and data dictionary +- Professional database architecture +- Complete index optimization + +**Use Cases** +- Multi-channel customer data integration +- 360-degree customer profile +- Precision marketing and segmentation +- Sales lead management and conversion +- Customer journey analysis +- Marketing attribution analysis +- Customer value assessment +- Customer health monitoring + +Implementation can be phased based on business needs and refined iteratively. From eee5e31c812e8dc4c4d7583adda9b2a68c7c600a Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 14:56:35 +0000 Subject: [PATCH 05/10] feat: Add B2B CDP entity design specification Co-authored-by: 13585811473 <13585811473@163.com> --- docs/b2b-cdp-entity-design.md | 1041 +++++++++++++++++---------------- 1 file changed, 524 insertions(+), 517 deletions(-) diff --git a/docs/b2b-cdp-entity-design.md b/docs/b2b-cdp-entity-design.md index fe3cbb0cf..aae6b98dc 100644 --- a/docs/b2b-cdp-entity-design.md +++ b/docs/b2b-cdp-entity-design.md @@ -1,83 +1,83 @@ -# B2B CDP Entity Design Specification +# B2B CDP 实体设计详细方案 -## Table of Contents -- [Architecture Overview](#architecture-overview) -- [Core Entity Design](#core-entity-design) -- [Entity Relationship](#entity-relationship) -- [Business Processes](#business-processes) -- [Data Model Design](#data-model-design) +## 目录 +- [整体架构](#整体架构) +- [核心实体设计](#核心实体设计) +- [实体关系图](#实体关系图) +- [业务流程](#业务流程) +- [数据模型详细设计](#数据模型详细设计) --- -## Architecture Overview +## 整体架构 -### Layered Architecture +### 实体分层架构图 ```mermaid graph TB - DS1[Official Website] --> CH[Channel Layer] - DS2[WeChat Ecosystem] --> CH - DS3[Social Media] --> CH - DS4[Offline Events] --> CH - DS5[Phone & Email] --> CH - DS6[CRM System] --> CH - DS7[3rd Party Platform] --> CH - - CH --> TP[Touchpoint] - CH --> EV[Event] - CM[Campaign] --> TP - - TP --> CT[Contact] - TP --> LD[Lead] + DS1[官方网站] --> CH[渠道层] + DS2[微信生态] --> CH + DS3[社交媒体] --> CH + DS4[线下活动] --> CH + DS5[电话邮件] --> CH + DS6[CRM系统] --> CH + DS7[第三方平台] --> CH + + CH --> TP[触点层Touchpoint] + CH --> EV[事件层Event] + CM[营销活动Campaign] --> TP + + TP --> CT[联系人Contact] + TP --> LD[线索Lead] EV --> CT EV --> LD - CT --> AC[Account] + CT --> AC[企业账户Account] LD --> CT CT --> CTS[ContactSummary] LD --> LDS[LeadSummary] AC --> ACS[AccountSummary] - LD --> OP[Opportunity] + LD --> OP[商机Opportunity] AC --> OP - OP --> PR[Product] + OP --> PR[产品Product] - CT --> SG[Segment] + CT --> SG[客户分群Segment] AC --> SG - CT --> TG[Tag] + CT --> TG[标签Tag] AC --> TG - CT --> SC[ScoreModel] + CT --> SC[评分模型Score] AC --> SC LD --> SC - CT --> JN[CustomerJourney] - CM --> AT[Attribution] + CT --> JN[客户旅程Journey] + CM --> AT[归因Attribution] OP --> AT ``` --- -### Technology Architecture +### 技术架构图 ```mermaid graph LR - API[API Gateway] --> PG[(PostgreSQL)] - API --> RD[(Redis Cache)] - API --> ES[(Elasticsearch)] + API[API网关] --> PG[(PostgreSQL
主数据库)] + API --> RD[(Redis
缓存层)] + API --> ES[(Elasticsearch
搜索引擎)] - WK[Background Worker] --> PG - WK --> CH[(ClickHouse)] + WK[后台任务] --> PG + WK --> CH[(ClickHouse
行为数据库)] WK --> RD - PG --> DW[(Data Warehouse)] + PG --> DW[(数据仓库
BI分析)] CH --> DW ``` --- -## Core Entity Design +## 核心实体设计 -### Account Entity Model +### 1. Account 企业账户实体 ```mermaid erDiagram @@ -124,24 +124,23 @@ erDiagram } ``` -**Key Fields:** -- `account_id` - Unique account identifier -- `unified_social_credit_code` - Enterprise unique identifier -- `account_type` - CUSTOMER, PARTNER, COMPETITOR, PROSPECT -- `account_status` - ACTIVE, DORMANT, CHURNED, BLACKLIST -- `account_level` - STRATEGIC, IMPORTANT, NORMAL -- `lifecycle_stage` - AWARENESS, CONSIDERATION, DECISION, RETENTION, EXPANSION - -**AccountSummary Metrics:** -- `total_contacts` - Number of associated contacts -- `total_revenue` - Cumulative revenue -- `lifetime_value` - Customer lifetime value -- `win_rate` - Win rate percentage -- `health_score` - Health score (0-100) +**核心字段说明：** +- `account_id` - 账户唯一标识 +- `unified_social_credit_code` - 统一社会信用代码（企业唯一标识） +- `account_type` - 客户类型（CUSTOMER客户、PARTNER合作伙伴、COMPETITOR竞争对手、PROSPECT潜在客户） +- `account_status` - 账户状态（ACTIVE活跃、DORMANT休眠、CHURNED流失、BLACKLIST黑名单） +- `account_level` - 客户等级（STRATEGIC战略级、IMPORTANT重要级、NORMAL普通级） + +**AccountSummary 汇总指标：** +- `total_contacts` - 关联联系人总数 +- `total_revenue` - 累计收入金额 +- `lifetime_value` - 客户生命周期价值 +- `win_rate` - 赢单率百分比 +- `health_score` - 健康度评分（0-100） --- -### Contact Entity Model +### 2. Contact 联系人实体 ```mermaid erDiagram @@ -195,18 +194,18 @@ erDiagram } ``` -**Key Fields:** -- `decision_level` - DECISION_MAKER, INFLUENCER, USER, GATEKEEPER -- `lifecycle_stage` - SUBSCRIBER, LEAD, MQL, SQL, OPPORTUNITY, CUSTOMER +**核心字段说明：** +- `decision_level` - 决策层级（DECISION_MAKER决策者、INFLUENCER影响者、USER使用者、GATEKEEPER把关者） +- `lifecycle_stage` - 生命周期阶段（SUBSCRIBER订阅者、LEAD线索、MQL市场合格线索、SQL销售合格线索、OPPORTUNITY商机、CUSTOMER客户） -**ContactSummary Metrics:** -- `engagement_score` - Engagement score (0-100) -- `email_opens` - Email open count -- `days_since_last_activity` - Days since last activity +**ContactSummary 汇总指标：** +- `engagement_score` - 参与度评分（0-100） +- `email_opens` - 邮件打开次数 +- `days_since_last_activity` - 距上次活跃天数 --- -### Lead Entity Model +### 3. Lead 线索实体 ```mermaid erDiagram @@ -253,15 +252,15 @@ erDiagram } ``` -**Key Fields:** -- `lead_status` - NEW, CONTACTED, QUALIFIED, CONVERTED, DISQUALIFIED -- `lead_score` - Lead score (0-100) -- `lead_grade` - A, B, C, D -- `is_qualified` - MQL/SQL indicator +**核心字段说明：** +- `lead_status` - 线索状态（NEW新建、CONTACTED已联系、QUALIFIED已限定、CONVERTED已转化、DISQUALIFIED无效） +- `lead_score` - 线索评分（0-100） +- `lead_grade` - 线索等级（A、B、C、D） +- `is_qualified` - 是否为合格线索（MQL市场合格线索/SQL销售合格线索） --- -### Opportunity Entity Model +### 4. Opportunity 商机实体 ```mermaid erDiagram @@ -305,14 +304,14 @@ erDiagram } ``` -**Key Fields:** -- `stage` - LEAD, QUALIFICATION, NEEDS_ANALYSIS, PROPOSAL, NEGOTIATION, CONTRACT, CLOSED_WON, CLOSED_LOST -- `probability` - Win probability (0-100) -- `opportunity_type` - NEW_BUSINESS, UPSELL, RENEWAL, CROSS_SELL +**核心字段说明：** +- `stage` - 商机阶段（LEAD线索、QUALIFICATION资格确认、NEEDS_ANALYSIS需求分析、PROPOSAL方案提议、NEGOTIATION商务谈判、CONTRACT合同签订、CLOSED_WON赢单、CLOSED_LOST输单） +- `probability` - 赢单概率（0-100） +- `opportunity_type` - 商机类型（NEW_BUSINESS新客户、UPSELL追加销售、RENEWAL续约、CROSS_SELL交叉销售） --- -### Channel Entity Model +### 5. Channel 渠道实体 ```mermaid erDiagram @@ -342,15 +341,15 @@ erDiagram } ``` -**Channel Types:** -- WEBSITE, SEO, SEM -- WECHAT, ENTERPRISE_WECHAT, DOUYIN -- EMAIL, PHONE -- OFFLINE_EVENT, EXHIBITION, PARTNER +**渠道类型：** +- WEBSITE官方网站、SEO搜索引擎优化、SEM搜索引擎营销 +- WECHAT微信、ENTERPRISE_WECHAT企业微信、DOUYIN抖音 +- EMAIL邮件营销、PHONE电话 +- OFFLINE_EVENT线下活动、EXHIBITION展会、PARTNER合作伙伴 --- -### Campaign Entity Model +### 6. Campaign 营销活动实体 ```mermaid erDiagram @@ -393,9 +392,17 @@ erDiagram } ``` +**活动类型：** +- WEBINAR网络研讨会 +- CONFERENCE线下会议 +- EXHIBITION展会 +- EMAIL_MARKETING邮件营销 +- CONTENT_MARKETING内容营销 +- PRODUCT_TRIAL产品试用 + --- -### Touchpoint Entity Model +### 7. Touchpoint 触点实体 ```mermaid erDiagram @@ -426,14 +433,14 @@ erDiagram } ``` -**Touchpoint Types:** -- PAGE_VIEW, FORM_SUBMIT, DOWNLOAD -- EMAIL_OPEN, EMAIL_CLICK -- CALL, MEETING, CHAT +**触点类型：** +- PAGE_VIEW页面浏览、FORM_SUBMIT表单提交、DOWNLOAD下载 +- EMAIL_OPEN邮件打开、EMAIL_CLICK邮件点击 +- CALL电话、MEETING会议、CHAT聊天 --- -### Event Entity Model +### 8. Event 行为事件实体 ```mermaid erDiagram @@ -455,15 +462,15 @@ erDiagram } ``` -**Event Types:** -- PAGE_VIEW, BUTTON_CLICK -- FORM_START, FORM_SUBMIT -- FILE_DOWNLOAD, VIDEO_PLAY -- PRODUCT_TRIAL, SEARCH +**事件类型：** +- PAGE_VIEW页面浏览、BUTTON_CLICK按钮点击 +- FORM_START表单开始、FORM_SUBMIT表单提交 +- FILE_DOWNLOAD文件下载、VIDEO_PLAY视频播放 +- PRODUCT_TRIAL产品试用、SEARCH搜索 --- -### Product Entity Model +### 9. Product 产品实体 ```mermaid erDiagram @@ -491,7 +498,7 @@ erDiagram --- -### Tag Entity Model +### 10. Tag 标签实体 ```mermaid erDiagram @@ -516,15 +523,15 @@ erDiagram } ``` -**Tag Types:** -- BEHAVIOR - Behavioral tags -- PROFILE - Profile tags -- BUSINESS - Business tags -- INTEREST - Interest tags +**标签类型：** +- BEHAVIOR行为标签 +- PROFILE画像标签 +- BUSINESS业务标签 +- INTEREST兴趣标签 --- -### Segment Entity Model +### 11. Segment 客户分群实体 ```mermaid erDiagram @@ -553,7 +560,7 @@ erDiagram --- -### Score Entity Model +### 12. Score 评分实体 ```mermaid erDiagram @@ -590,14 +597,14 @@ erDiagram } ``` -**Score Types:** -- LEAD_SCORING - Lead scoring -- ACCOUNT_HEALTH - Account health scoring -- CONTACT_ENGAGEMENT - Contact engagement scoring +**评分类型：** +- LEAD_SCORING线索评分 +- ACCOUNT_HEALTH企业健康度评分 +- CONTACT_ENGAGEMENT联系人参与度评分 --- -### Industry Entity Model +### 13. Industry 行业实体 ```mermaid erDiagram @@ -615,7 +622,7 @@ erDiagram --- -### Attribution Entity Model +### 14. Attribution 归因实体 ```mermaid erDiagram @@ -640,17 +647,17 @@ erDiagram } ``` -**Attribution Models:** -- FIRST_TOUCH - First touch attribution -- LAST_TOUCH - Last touch attribution -- LINEAR - Linear attribution -- TIME_DECAY - Time decay attribution -- U_SHAPED - U-shaped attribution -- W_SHAPED - W-shaped attribution +**归因模型：** +- FIRST_TOUCH首次触点归因 +- LAST_TOUCH末次触点归因 +- LINEAR线性归因 +- TIME_DECAY时间衰减归因 +- U_SHAPED U型归因 +- W_SHAPED W型归因 --- -### CustomerJourney Entity Model +### 15. CustomerJourney 客户旅程实体 ```mermaid erDiagram @@ -678,9 +685,9 @@ erDiagram --- -## Entity Relationship +## 实体关系图 -### Core Entity Relationships +### 核心实体关系总览 ```mermaid erDiagram @@ -728,556 +735,556 @@ erDiagram --- -### Identity Resolution Architecture +### 全渠道身份统一架构 ```mermaid graph TB - S1[WeChat Official] --> IR[Identity Resolution Engine] - S2[Enterprise WeChat] --> IR - S3[Website] --> IR - S4[Douyin] --> IR - S5[Email] --> IR - S6[Phone] --> IR - S7[Offline Events] --> IR - - IR --> IM[Identity Matching] - IM --> UC[Unified Contact] - IM --> UA[Unified Account] - - UC --> CI1[ContactChannelIdentity WeChat] - UC --> CI2[ContactChannelIdentity Website] - UC --> CI3[ContactChannelIdentity Email] - - UA --> AI1[AccountChannelIdentity CorpID] - UA --> AI2[AccountChannelIdentity WebsiteID] + S1[微信公众号] --> IR[身份识别引擎] + S2[企业微信] --> IR + S3[官方网站] --> IR + S4[抖音] --> IR + S5[邮件系统] --> IR + S6[电话系统] --> IR + S7[线下活动] --> IR + + IR --> IM[身份匹配] + IM --> UC[统一联系人Contact] + IM --> UA[统一企业账户Account] + + UC --> CI1[ContactChannelIdentity
微信OpenID] + UC --> CI2[ContactChannelIdentity
企业微信UserID] + UC --> CI3[ContactChannelIdentity
网站CookieID] + + UA --> AI1[AccountChannelIdentity
企业微信CorpID] + UA --> AI2[AccountChannelIdentity
网站企业ID] ``` --- -## Business Processes +## 业务流程 -### Lead to Opportunity Conversion +### 线索到商机转化流程 ```mermaid stateDiagram-v2 - [*] --> NewLead + [*] --> 新线索 - NewLead --> Contacted: First Contact - NewLead --> Invalid: Mark Invalid + 新线索 --> 已联系: 首次联系 + 新线索 --> 无效线索: 标记无效 - Contacted --> Qualified: Qualification - Contacted --> Invalid: Mark Invalid + 已联系 --> 已限定: 资格验证 + 已联系 --> 无效线索: 标记无效 - Qualified --> Converted: Convert + 已限定 --> 已转化: 转化 - state Converted { - [*] --> CreateContact - [*] --> CreateAccount - [*] --> CreateOpportunity + state 已转化 { + [*] --> 创建联系人 + [*] --> 创建企业账户 + [*] --> 创建商机 } - Converted --> [*] - Invalid --> [*] + 已转化 --> [*] + 无效线索 --> [*] ``` --- -### Opportunity Stage Progression +### 商机阶段流转流程 ```mermaid stateDiagram-v2 - [*] --> LeadStage + [*] --> 线索阶段 - LeadStage --> Qualification - LeadStage --> Lost + 线索阶段 --> 需求确认: 需求分析 + 线索阶段 --> 输单: 丢失 - Qualification --> SolutionDesign - Qualification --> Lost + 需求确认 --> 方案设计: 方案输出 + 需求确认 --> 输单: 丢失 - SolutionDesign --> Negotiation - SolutionDesign --> Lost + 方案设计 --> 商务谈判: 开始谈判 + 方案设计 --> 输单: 丢失 - Negotiation --> Contract - Negotiation --> Lost + 商务谈判 --> 合同签订: 签订合同 + 商务谈判 --> 输单: 丢失 - Contract --> Won - Contract --> Lost + 合同签订 --> 赢单: 成功签约 + 合同签订 --> 输单: 失败 - Won --> [*] - Lost --> [*] + 赢单 --> [*] + 输单 --> [*] ``` --- -### Customer Lifecycle +### 客户生命周期管理 ```mermaid stateDiagram-v2 - [*] --> Awareness + [*] --> 认知阶段 - Awareness --> Consideration - Consideration --> Decision - Decision --> Retention - Retention --> Expansion + 认知阶段 --> 考虑阶段: 产生兴趣 + 考虑阶段 --> 决策阶段: 深度评估 + 决策阶段 --> 留存阶段: 成交购买 + 留存阶段 --> 扩展阶段: 追加购买 - Retention --> Churn - Expansion --> Churn + 留存阶段 --> 流失: 停止使用 + 扩展阶段 --> 流失: 停止使用 - Churn --> Winback - Winback --> Retention - Winback --> [*] + 流失 --> 召回: 召回营销 + 召回 --> 留存阶段: 成功召回 + 召回 --> [*]: 永久流失 - Expansion --> [*] + 扩展阶段 --> [*]: 持续合作 ``` --- -### Multi-Channel Data Flow +### 全渠道数据流转流程 ```mermaid sequenceDiagram - participant User - participant Channel - participant Event - participant Identity - participant Lead - participant Contact - participant Account - participant Summary - - User->>Channel: Interaction - Channel->>Event: Log Event - Event->>Identity: Identify User - - alt New User - Identity->>Lead: Create Lead - Lead->>Summary: Update LeadSummary - else Known User - Identity->>Contact: Link Contact - Contact->>Account: Link Account - Account->>Summary: Update Summary + participant 用户 + participant 渠道 + participant 事件系统 + participant 身份识别 + participant 线索管理 + participant 联系人管理 + participant 企业账户管理 + participant 汇总计算 + + 用户->>渠道: 1.访问互动 + 渠道->>事件系统: 2.记录事件 + 事件系统->>身份识别: 3.身份识别 + + alt 新用户 + 身份识别->>线索管理: 4a.创建Lead + 线索管理->>汇总计算: 5a.更新LeadSummary + else 已识别用户 + 身份识别->>联系人管理: 4b.关联Contact + 联系人管理->>企业账户管理: 5b.关联Account + 企业账户管理->>汇总计算: 6b.更新Summary end - Summary->>Channel: Trigger Response - Channel->>User: Return Content + 汇总计算->>渠道: 7.触发规则引擎 + 渠道->>用户: 8.个性化响应 ``` --- -### Identity Matching Process +### 身份识别匹配流程 ```mermaid flowchart TD - Start[Receive Data] --> Extract[Extract Identifiers] + 开始[接收多渠道数据] --> 提取[提取身份标识符] - Extract --> Phone{Has Phone?} - Extract --> Email{Has Email?} - Extract --> WeChat{Has WeChat?} + 提取 --> 手机{有手机号?} + 提取 --> 邮箱{有邮箱?} + 提取 --> 微信{有微信ID?} - Phone -->|Yes| MatchPhone[Match by Phone Priority 1] - Email -->|Yes| MatchEmail[Match by Email Priority 2] - WeChat -->|Yes| MatchWeChat[Match by WeChat Priority 3] + 手机 -->|是| 手机匹配[按手机号匹配
优先级1] + 邮箱 -->|是| 邮箱匹配[按邮箱匹配
优先级2] + 微信 -->|是| 微信匹配[按微信ID匹配
优先级3] - MatchPhone --> Found{Found?} - MatchEmail --> Found - MatchWeChat --> Found + 手机匹配 --> 找到{找到匹配?} + 邮箱匹配 --> 找到 + 微信匹配 --> 找到 - Found -->|Yes| Merge[Merge to Existing] - Found -->|No| Create[Create New] + 找到 -->|是| 合并[合并到现有Contact] + 找到 -->|否| 创建[创建新Contact] - Merge --> AddIdentity[Add Channel Identity] - Create --> AddIdentity + 合并 --> 添加身份[添加更新
ContactChannelIdentity] + 创建 --> 添加身份 - AddIdentity --> UpdateSummary[Update Summary] - UpdateSummary --> End[Complete] + 添加身份 --> 更新汇总[更新ContactSummary] + 更新汇总 --> 完成[完成] ``` --- -### Summary Update Strategy +### 汇总数据更新策略 ```mermaid flowchart LR - T1[Scheduled Job Hourly] --> Engine[Calculation Engine] - T2[Real-time Trigger] --> Engine - T3[Manual Refresh] --> Engine + 定时任务[定时任务
每小时执行] --> 计算引擎[计算引擎] + 实时触发[实时触发
关键事件] --> 计算引擎 + 手动刷新[手动刷新
按需执行] --> 计算引擎 - Engine --> S1[AccountSummary] - Engine --> S2[ContactSummary] - Engine --> S3[LeadSummary] + 计算引擎 --> 企业汇总[AccountSummary] + 计算引擎 --> 联系人汇总[ContactSummary] + 计算引擎 --> 线索汇总[LeadSummary] ``` --- -## Data Model Design - -### Account Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| account_id | VARCHAR | 64 | PK, NOT NULL | Unique account ID | ACC_20231105001 | -| account_name | VARCHAR | 200 | NOT NULL | Account name | Alibaba Group | -| unified_social_credit_code | VARCHAR | 18 | UNIQUE | Unified credit code | 91330000MA27XYZ123 | -| account_type | VARCHAR | 50 | NOT NULL | Account type | CUSTOMER, PARTNER, PROSPECT | -| industry_id | VARCHAR | 64 | FK | Industry ID | IND_001 | -| account_status | VARCHAR | 50 | NOT NULL | Account status | ACTIVE, DORMANT, CHURNED | -| account_level | VARCHAR | 50 | | Account level | STRATEGIC, IMPORTANT, NORMAL | -| annual_revenue | DECIMAL | 18,2 | | Annual revenue | 50000.00 | -| employee_count | INT | | | Employee count | 5000 | -| company_website | VARCHAR | 500 | | Website URL | https://www.alibaba.com | -| province | VARCHAR | 50 | | Province | Zhejiang | -| city | VARCHAR | 50 | | City | Hangzhou | -| account_source | VARCHAR | 100 | | Source | WEBSITE, EXHIBITION, PARTNER | -| primary_channel_id | VARCHAR | 64 | FK | Primary channel | CH_001 | -| owner_user_id | VARCHAR | 64 | FK | Owner user ID | USER_001 | -| created_at | DATETIME | | NOT NULL | Created timestamp | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | Updated timestamp | 2023-11-05 15:20:00 | -| lifecycle_stage | VARCHAR | 50 | | Lifecycle stage | AWARENESS, RETENTION, EXPANSION | +## 数据模型详细设计 + +### Account 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| account_id | VARCHAR | 64 | PK, NOT NULL | 账户唯一标识 | ACC_20231105001 | +| account_name | VARCHAR | 200 | NOT NULL | 企业名称 | 阿里巴巴网络技术有限公司 | +| unified_social_credit_code | VARCHAR | 18 | UNIQUE | 统一社会信用代码 | 91330000MA27XYZ123 | +| account_type | VARCHAR | 50 | NOT NULL | 客户类型 | CUSTOMER、PARTNER、PROSPECT | +| industry_id | VARCHAR | 64 | FK | 行业分类外键 | IND_001 | +| account_status | VARCHAR | 50 | NOT NULL | 账户状态 | ACTIVE、DORMANT、CHURNED | +| account_level | VARCHAR | 50 | | 客户等级 | STRATEGIC、IMPORTANT、NORMAL | +| annual_revenue | DECIMAL | 18,2 | | 年营收（万元） | 50000.00 | +| employee_count | INT | | | 员工人数 | 5000 | +| company_website | VARCHAR | 500 | | 公司网站 | https://www.alibaba.com | +| province | VARCHAR | 50 | | 省份 | 浙江省 | +| city | VARCHAR | 50 | | 城市 | 杭州市 | +| account_source | VARCHAR | 100 | | 来源渠道 | WEBSITE、EXHIBITION、PARTNER | +| primary_channel_id | VARCHAR | 64 | FK | 主渠道ID | CH_001 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | AWARENESS、RETENTION、EXPANSION | --- -### AccountSummary Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| summary_id | VARCHAR | 64 | PK, NOT NULL | Summary ID | ACCS_20231105001 | -| account_id | VARCHAR | 64 | FK, UNIQUE | Account ID | ACC_20231105001 | -| total_contacts | INT | | DEFAULT 0 | Total contacts | 25 | -| total_opportunities | INT | | DEFAULT 0 | Total opportunities | 8 | -| total_leads | INT | | DEFAULT 0 | Total leads | 45 | -| total_revenue | DECIMAL | 18,2 | DEFAULT 0 | Total revenue | 5000000.00 | -| lifetime_value | DECIMAL | 18,2 | DEFAULT 0 | Lifetime value | 8000000.00 | -| won_opportunities | INT | | DEFAULT 0 | Won count | 5 | -| lost_opportunities | INT | | DEFAULT 0 | Lost count | 2 | -| win_rate | DECIMAL | 5,2 | DEFAULT 0 | Win rate % | 71.43 | -| total_touchpoints | INT | | DEFAULT 0 | Total touchpoints | 156 | -| health_score | INT | | DEFAULT 0 | Health score 0-100 | 85 | -| last_activity_at | DATETIME | | | Last activity time | 2023-11-05 14:30:00 | -| calculated_at | DATETIME | | NOT NULL | Calculated time | 2023-11-05 16:00:00 | -| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 16:00:00 | +### AccountSummary 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | ACCS_20231105001 | +| account_id | VARCHAR | 64 | FK, UNIQUE | 账户ID | ACC_20231105001 | +| total_contacts | INT | | DEFAULT 0 | 关联联系人总数 | 25 | +| total_opportunities | INT | | DEFAULT 0 | 商机总数 | 8 | +| total_leads | INT | | DEFAULT 0 | 线索总数 | 45 | +| total_revenue | DECIMAL | 18,2 | DEFAULT 0 | 累计收入（元） | 5000000.00 | +| lifetime_value | DECIMAL | 18,2 | DEFAULT 0 | 生命周期价值（元） | 8000000.00 | +| won_opportunities | INT | | DEFAULT 0 | 赢单数量 | 5 | +| lost_opportunities | INT | | DEFAULT 0 | 输单数量 | 2 | +| win_rate | DECIMAL | 5,2 | DEFAULT 0 | 赢单率（%） | 71.43 | +| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 156 | +| health_score | INT | | DEFAULT 0 | 健康度评分（0-100） | 85 | +| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | +| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | --- -### Contact Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| contact_id | VARCHAR | 64 | PK, NOT NULL | Contact ID | CNT_20231105001 | -| contact_name | VARCHAR | 100 | NOT NULL | Contact name | Zhang Wei | -| mobile_phone | VARCHAR | 20 | UNIQUE | Mobile phone | 13800138000 | -| email | VARCHAR | 200 | UNIQUE | Email address | zhangwei@company.com | -| wechat_id | VARCHAR | 100 | | WeChat ID | wx_zhangwei | -| job_title | VARCHAR | 100 | | Job title | CTO | -| department | VARCHAR | 100 | | Department | Technology | -| contact_status | VARCHAR | 50 | NOT NULL | Status | ACTIVE, INACTIVE, UNSUBSCRIBED | -| primary_account_id | VARCHAR | 64 | FK | Primary account | ACC_20231105001 | -| contact_source | VARCHAR | 100 | | Source | WEBSITE, FORM, IMPORT | -| primary_channel_id | VARCHAR | 64 | FK | Primary channel | CH_001 | -| owner_user_id | VARCHAR | 64 | FK | Owner user | USER_001 | -| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 15:20:00 | -| lifecycle_stage | VARCHAR | 50 | | Lifecycle stage | LEAD, MQL, SQL, CUSTOMER | -| is_decision_maker | BOOLEAN | | DEFAULT FALSE | Is decision maker | true | -| is_verified | BOOLEAN | | DEFAULT FALSE | Is verified | true | +### Contact 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| contact_id | VARCHAR | 64 | PK, NOT NULL | 联系人唯一标识 | CNT_20231105001 | +| contact_name | VARCHAR | 100 | NOT NULL | 联系人姓名 | 张伟 | +| mobile_phone | VARCHAR | 20 | UNIQUE | 手机号 | 13800138000 | +| email | VARCHAR | 200 | UNIQUE | 邮箱地址 | zhangwei@company.com | +| wechat_id | VARCHAR | 100 | | 微信ID | wx_zhangwei | +| job_title | VARCHAR | 100 | | 职位 | 首席技术官 | +| department | VARCHAR | 100 | | 部门 | 技术部 | +| contact_status | VARCHAR | 50 | NOT NULL | 联系人状态 | ACTIVE、INACTIVE、UNSUBSCRIBED | +| primary_account_id | VARCHAR | 64 | FK | 主要关联企业ID | ACC_20231105001 | +| contact_source | VARCHAR | 100 | | 来源 | WEBSITE、FORM、IMPORT | +| primary_channel_id | VARCHAR | 64 | FK | 主渠道ID | CH_001 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| lifecycle_stage | VARCHAR | 50 | | 生命周期阶段 | LEAD、MQL、SQL、CUSTOMER | +| is_decision_maker | BOOLEAN | | DEFAULT FALSE | 是否决策者 | true | +| is_verified | BOOLEAN | | DEFAULT FALSE | 是否已验证 | true | --- -### ContactSummary Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| summary_id | VARCHAR | 64 | PK, NOT NULL | Summary ID | CNTS_20231105001 | -| contact_id | VARCHAR | 64 | FK, UNIQUE | Contact ID | CNT_20231105001 | -| total_touchpoints | INT | | DEFAULT 0 | Total touchpoints | 87 | -| total_events | INT | | DEFAULT 0 | Total events | 234 | -| email_opens | INT | | DEFAULT 0 | Email opens | 45 | -| email_clicks | INT | | DEFAULT 0 | Email clicks | 23 | -| form_submissions | INT | | DEFAULT 0 | Form submissions | 12 | -| content_downloads | INT | | DEFAULT 0 | Content downloads | 8 | -| engagement_score | INT | | DEFAULT 0 | Engagement score 0-100 | 78 | -| last_activity_at | DATETIME | | | Last activity | 2023-11-05 14:30:00 | -| days_since_last_activity | INT | | DEFAULT 0 | Days since activity | 1 | -| calculated_at | DATETIME | | NOT NULL | Calculated time | 2023-11-05 16:00:00 | -| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 16:00:00 | +### ContactSummary 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | CNTS_20231105001 | +| contact_id | VARCHAR | 64 | FK, UNIQUE | 联系人ID | CNT_20231105001 | +| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 87 | +| total_events | INT | | DEFAULT 0 | 总事件数 | 234 | +| email_opens | INT | | DEFAULT 0 | 邮件打开次数 | 45 | +| email_clicks | INT | | DEFAULT 0 | 邮件点击次数 | 23 | +| form_submissions | INT | | DEFAULT 0 | 表单提交次数 | 12 | +| content_downloads | INT | | DEFAULT 0 | 内容下载次数 | 8 | +| engagement_score | INT | | DEFAULT 0 | 参与度评分（0-100） | 78 | +| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | +| days_since_last_activity | INT | | DEFAULT 0 | 距上次活跃天数 | 1 | +| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | --- -### Lead Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| lead_id | VARCHAR | 64 | PK, NOT NULL | Lead ID | LEAD_20231105001 | -| lead_name | VARCHAR | 100 | NOT NULL | Lead name | Li Ming | -| company_name | VARCHAR | 200 | | Company name | Tencent Technology | -| mobile_phone | VARCHAR | 20 | | Mobile phone | 13900139000 | -| email | VARCHAR | 200 | | Email | liming@company.com | -| wechat_id | VARCHAR | 100 | | WeChat ID | wx_liming | -| job_title | VARCHAR | 100 | | Job title | Product Manager | -| lead_source | VARCHAR | 100 | NOT NULL | Lead source | WEBSITE, FORM, CAMPAIGN | -| channel_id | VARCHAR | 64 | FK | Channel ID | CH_001 | -| campaign_id | VARCHAR | 64 | FK | Campaign ID | CMP_001 | -| lead_status | VARCHAR | 50 | NOT NULL | Status | NEW, CONTACTED, QUALIFIED | -| lead_score | INT | | DEFAULT 0 | Lead score 0-100 | 80 | -| lead_grade | VARCHAR | 10 | | Lead grade | A, B, C, D | -| industry_id | VARCHAR | 64 | FK | Industry ID | IND_001 | -| province | VARCHAR | 50 | | Province | Guangdong | -| city | VARCHAR | 50 | | City | Shenzhen | -| owner_user_id | VARCHAR | 64 | FK | Owner user | USER_001 | -| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 15:20:00 | -| last_contacted_at | DATETIME | | | Last contact | 2023-11-05 14:00:00 | -| converted_at | DATETIME | | | Converted time | 2023-11-10 09:00:00 | -| converted_contact_id | VARCHAR | 64 | FK | Converted contact | CNT_20231110001 | -| converted_account_id | VARCHAR | 64 | FK | Converted account | ACC_20231110001 | -| converted_opportunity_id | VARCHAR | 64 | FK | Converted opportunity | OPP_20231110001 | -| is_qualified | BOOLEAN | | DEFAULT FALSE | Is qualified MQL/SQL | true | +### Lead 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| lead_id | VARCHAR | 64 | PK, NOT NULL | 线索唯一标识 | LEAD_20231105001 | +| lead_name | VARCHAR | 100 | NOT NULL | 线索姓名 | 李明 | +| company_name | VARCHAR | 200 | | 公司名称 | 腾讯科技有限公司 | +| mobile_phone | VARCHAR | 20 | | 手机号 | 13900139000 | +| email | VARCHAR | 200 | | 邮箱地址 | liming@company.com | +| wechat_id | VARCHAR | 100 | | 微信ID | wx_liming | +| job_title | VARCHAR | 100 | | 职位 | 产品经理 | +| lead_source | VARCHAR | 100 | NOT NULL | 线索来源 | WEBSITE、FORM、CAMPAIGN | +| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | +| campaign_id | VARCHAR | 64 | FK | 营销活动ID | CMP_001 | +| lead_status | VARCHAR | 50 | NOT NULL | 线索状态 | NEW、CONTACTED、QUALIFIED | +| lead_score | INT | | DEFAULT 0 | 线索评分（0-100） | 80 | +| lead_grade | VARCHAR | 10 | | 线索等级 | A、B、C、D | +| industry_id | VARCHAR | 64 | FK | 行业ID | IND_001 | +| province | VARCHAR | 50 | | 省份 | 广东省 | +| city | VARCHAR | 50 | | 城市 | 深圳市 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| last_contacted_at | DATETIME | | | 最后联系时间 | 2023-11-05 14:00:00 | +| converted_at | DATETIME | | | 转化时间 | 2023-11-10 09:00:00 | +| converted_contact_id | VARCHAR | 64 | FK | 转化后联系人ID | CNT_20231110001 | +| converted_account_id | VARCHAR | 64 | FK | 转化后企业ID | ACC_20231110001 | +| converted_opportunity_id | VARCHAR | 64 | FK | 转化后商机ID | OPP_20231110001 | +| is_qualified | BOOLEAN | | DEFAULT FALSE | 是否为合格线索 | true | --- -### LeadSummary Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| summary_id | VARCHAR | 64 | PK, NOT NULL | Summary ID | LEADS_20231105001 | -| lead_id | VARCHAR | 64 | FK, UNIQUE | Lead ID | LEAD_20231105001 | -| total_touchpoints | INT | | DEFAULT 0 | Total touchpoints | 12 | -| total_events | INT | | DEFAULT 0 | Total events | 45 | -| form_submissions | INT | | DEFAULT 0 | Form submissions | 3 | -| content_downloads | INT | | DEFAULT 0 | Content downloads | 2 | -| page_views | INT | | DEFAULT 0 | Page views | 28 | -| days_in_pipeline | INT | | DEFAULT 0 | Days in pipeline | 7 | -| contact_attempts | INT | | DEFAULT 0 | Contact attempts | 4 | -| last_activity_at | DATETIME | | | Last activity | 2023-11-05 14:30:00 | -| last_contact_attempt_at | DATETIME | | | Last contact attempt | 2023-11-05 11:00:00 | -| calculated_at | DATETIME | | NOT NULL | Calculated time | 2023-11-05 16:00:00 | -| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 16:00:00 | +### LeadSummary 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| summary_id | VARCHAR | 64 | PK, NOT NULL | 汇总记录唯一标识 | LEADS_20231105001 | +| lead_id | VARCHAR | 64 | FK, UNIQUE | 线索ID | LEAD_20231105001 | +| total_touchpoints | INT | | DEFAULT 0 | 总触点数 | 12 | +| total_events | INT | | DEFAULT 0 | 总事件数 | 45 | +| form_submissions | INT | | DEFAULT 0 | 表单提交次数 | 3 | +| content_downloads | INT | | DEFAULT 0 | 内容下载次数 | 2 | +| page_views | INT | | DEFAULT 0 | 页面浏览次数 | 28 | +| days_in_pipeline | INT | | DEFAULT 0 | 在管道中天数 | 7 | +| contact_attempts | INT | | DEFAULT 0 | 联系尝试次数 | 4 | +| last_activity_at | DATETIME | | | 最后活跃时间 | 2023-11-05 14:30:00 | +| last_contact_attempt_at | DATETIME | | | 最后联系尝试时间 | 2023-11-05 11:00:00 | +| calculated_at | DATETIME | | NOT NULL | 计算时间 | 2023-11-05 16:00:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 16:00:00 | --- -### Opportunity Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| opportunity_id | VARCHAR | 64 | PK, NOT NULL | Opportunity ID | OPP_20231105001 | -| opportunity_name | VARCHAR | 200 | NOT NULL | Opportunity name | Tencent AI Platform Project | -| account_id | VARCHAR | 64 | FK, NOT NULL | Account ID | ACC_20231105001 | -| primary_contact_id | VARCHAR | 64 | FK | Primary contact | CNT_20231105001 | -| lead_id | VARCHAR | 64 | FK | Source lead | LEAD_20231105001 | -| opportunity_type | VARCHAR | 50 | | Type | NEW_BUSINESS, UPSELL, RENEWAL | -| opportunity_source | VARCHAR | 100 | | Source | LEAD_CONVERSION, DIRECT_SALES | -| amount | DECIMAL | 18,2 | | Amount | 1000000.00 | -| currency | VARCHAR | 10 | | Currency | CNY, USD, EUR | -| stage | VARCHAR | 50 | NOT NULL | Stage | QUALIFICATION, PROPOSAL, CONTRACT | -| probability | INT | | DEFAULT 0 | Probability 0-100 | 60 | -| expected_close_date | DATE | | | Expected close | 2023-12-31 | -| actual_close_date | DATE | | | Actual close | 2023-12-25 | -| close_reason | VARCHAR | 200 | | Close reason | Price, Competitor, Budget, Success | -| owner_user_id | VARCHAR | 64 | FK | Owner user | USER_001 | -| campaign_id | VARCHAR | 64 | FK | Campaign | CMP_001 | -| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 15:20:00 | -| days_in_stage | INT | | DEFAULT 0 | Days in stage | 15 | -| is_won | BOOLEAN | | DEFAULT FALSE | Is won | false | -| is_lost | BOOLEAN | | DEFAULT FALSE | Is lost | false | +### Opportunity 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| opportunity_id | VARCHAR | 64 | PK, NOT NULL | 商机唯一标识 | OPP_20231105001 | +| opportunity_name | VARCHAR | 200 | NOT NULL | 商机名称 | 腾讯科技AI平台采购项目 | +| account_id | VARCHAR | 64 | FK, NOT NULL | 关联企业ID | ACC_20231105001 | +| primary_contact_id | VARCHAR | 64 | FK | 主要联系人ID | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | 来源线索ID | LEAD_20231105001 | +| opportunity_type | VARCHAR | 50 | | 商机类型 | NEW_BUSINESS、UPSELL、RENEWAL | +| opportunity_source | VARCHAR | 100 | | 商机来源 | LEAD_CONVERSION、DIRECT_SALES | +| amount | DECIMAL | 18,2 | | 预计金额（元） | 1000000.00 | +| currency | VARCHAR | 10 | | 货币单位 | CNY、USD、EUR | +| stage | VARCHAR | 50 | NOT NULL | 当前阶段 | QUALIFICATION、PROPOSAL、CONTRACT | +| probability | INT | | DEFAULT 0 | 赢单概率（0-100） | 60 | +| expected_close_date | DATE | | | 预计成交日期 | 2023-12-31 | +| actual_close_date | DATE | | | 实际成交日期 | 2023-12-25 | +| close_reason | VARCHAR | 200 | | 关闭原因 | 价格、竞争对手、预算、成功 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| campaign_id | VARCHAR | 64 | FK | 来源活动ID | CMP_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | +| days_in_stage | INT | | DEFAULT 0 | 当前阶段停留天数 | 15 | +| is_won | BOOLEAN | | DEFAULT FALSE | 是否赢单 | false | +| is_lost | BOOLEAN | | DEFAULT FALSE | 是否输单 | false | --- -### Channel Table Schema +### Channel 详细字段设计 -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| channel_id | VARCHAR | 64 | PK, NOT NULL | Channel ID | CH_001 | -| channel_name | VARCHAR | 100 | NOT NULL | Channel name | Website Product Page | -| channel_type | VARCHAR | 50 | NOT NULL | Type | WEBSITE, WECHAT, EMAIL, PHONE | -| channel_category | VARCHAR | 50 | | Category | ONLINE, OFFLINE, SOCIAL | -| parent_channel_id | VARCHAR | 64 | FK | Parent channel | CH_PARENT_001 | -| channel_status | VARCHAR | 50 | NOT NULL | Status | ACTIVE, INACTIVE, TESTING | -| cost | DECIMAL | 18,2 | | Cost per month | 50000.00 | -| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 10:30:00 | -| updated_at | DATETIME | | NOT NULL | Updated time | 2023-11-05 15:20:00 | +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| channel_id | VARCHAR | 64 | PK, NOT NULL | 渠道唯一标识 | CH_001 | +| channel_name | VARCHAR | 100 | NOT NULL | 渠道名称 | 官网产品页 | +| channel_type | VARCHAR | 50 | NOT NULL | 渠道类型 | WEBSITE、WECHAT、EMAIL、PHONE | +| channel_category | VARCHAR | 50 | | 渠道分类 | ONLINE、OFFLINE、SOCIAL | +| parent_channel_id | VARCHAR | 64 | FK | 父渠道ID | CH_PARENT_001 | +| channel_status | VARCHAR | 50 | NOT NULL | 渠道状态 | ACTIVE、INACTIVE、TESTING | +| cost | DECIMAL | 18,2 | | 渠道成本（元/月） | 50000.00 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 10:30:00 | +| updated_at | DATETIME | | NOT NULL | 更新时间 | 2023-11-05 15:20:00 | --- -### Touchpoint Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| touchpoint_id | VARCHAR | 64 | PK, NOT NULL | Touchpoint ID | TP_20231105001 | -| touchpoint_type | VARCHAR | 50 | NOT NULL | Type | PAGE_VIEW, FORM_SUBMIT, EMAIL | -| channel_id | VARCHAR | 64 | FK | Channel | CH_001 | -| campaign_id | VARCHAR | 64 | FK | Campaign | CMP_001 | -| contact_id | VARCHAR | 64 | FK | Contact | CNT_20231105001 | -| lead_id | VARCHAR | 64 | FK | Lead | LEAD_20231105001 | -| account_id | VARCHAR | 64 | FK | Account | ACC_20231105001 | -| touchpoint_time | DATETIME | | NOT NULL | Touchpoint time | 2023-11-05 14:30:00 | -| touchpoint_direction | VARCHAR | 20 | | Direction | INBOUND, OUTBOUND | -| touchpoint_status | VARCHAR | 50 | | Status | COMPLETED, SCHEDULED, CANCELLED | -| content_type | VARCHAR | 50 | | Content type | WHITEPAPER, WEBINAR, DEMO | -| content_id | VARCHAR | 64 | FK | Content ID | CONTENT_001 | -| subject | VARCHAR | 200 | | Subject | Product Demo Meeting | -| duration_seconds | INT | | | Duration seconds | 3600 | -| owner_user_id | VARCHAR | 64 | FK | Owner user | USER_001 | -| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 14:30:00 | +### Touchpoint 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| touchpoint_id | VARCHAR | 64 | PK, NOT NULL | 触点唯一标识 | TP_20231105001 | +| touchpoint_type | VARCHAR | 50 | NOT NULL | 触点类型 | PAGE_VIEW、FORM_SUBMIT、EMAIL | +| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | +| campaign_id | VARCHAR | 64 | FK | 营销活动ID | CMP_001 | +| contact_id | VARCHAR | 64 | FK | 联系人ID | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | 线索ID | LEAD_20231105001 | +| account_id | VARCHAR | 64 | FK | 企业ID | ACC_20231105001 | +| touchpoint_time | DATETIME | | NOT NULL | 触点发生时间 | 2023-11-05 14:30:00 | +| touchpoint_direction | VARCHAR | 20 | | 触点方向 | INBOUND、OUTBOUND | +| touchpoint_status | VARCHAR | 50 | | 触点状态 | COMPLETED、SCHEDULED、CANCELLED | +| content_type | VARCHAR | 50 | | 内容类型 | WHITEPAPER、WEBINAR、DEMO | +| content_id | VARCHAR | 64 | FK | 内容ID | CONTENT_001 | +| subject | VARCHAR | 200 | | 主题 | 产品演示会议 | +| duration_seconds | INT | | | 持续时长（秒） | 3600 | +| owner_user_id | VARCHAR | 64 | FK | 负责人ID | USER_001 | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 14:30:00 | --- -### Event Table Schema - -| Field | Type | Length | Constraint | Description | Example | -|-------|------|--------|------------|-------------|---------| -| event_id | VARCHAR | 64 | PK, NOT NULL | Event ID | EVT_20231105001 | -| event_name | VARCHAR | 100 | NOT NULL | Event name | page_view | -| event_type | VARCHAR | 50 | NOT NULL | Type | PAGE_VIEW, CLICK, FORM_SUBMIT | -| channel_id | VARCHAR | 64 | FK | Channel | CH_001 | -| contact_id | VARCHAR | 64 | FK | Contact | CNT_20231105001 | -| lead_id | VARCHAR | 64 | FK | Lead | LEAD_20231105001 | -| account_id | VARCHAR | 64 | FK | Account | ACC_20231105001 | -| event_time | DATETIME | | NOT NULL | Event time | 2023-11-05 14:35:20 | -| session_id | VARCHAR | 64 | | Session | SESSION_20231105001 | -| device_type | VARCHAR | 50 | | Device | DESKTOP, MOBILE, TABLET | -| browser | VARCHAR | 50 | | Browser | Chrome, Safari, Firefox | -| os | VARCHAR | 50 | | OS | Windows 10, iOS 16 | -| ip_address | VARCHAR | 50 | | IP | 192.168.1.1 | -| page_url | VARCHAR | 1000 | | Page URL | https://example.com/product | -| referrer_url | VARCHAR | 1000 | | Referrer URL | https://baidu.com/search | -| created_at | DATETIME | | NOT NULL | Created time | 2023-11-05 14:35:20 | +### Event 详细字段设计 + +| 字段名 | 类型 | 长度 | 约束 | 说明 | 示例值 | +|--------|------|------|------|------|--------| +| event_id | VARCHAR | 64 | PK, NOT NULL | 事件唯一标识 | EVT_20231105001 | +| event_name | VARCHAR | 100 | NOT NULL | 事件名称 | page_view | +| event_type | VARCHAR | 50 | NOT NULL | 事件类型 | PAGE_VIEW、CLICK、FORM_SUBMIT | +| channel_id | VARCHAR | 64 | FK | 渠道ID | CH_001 | +| contact_id | VARCHAR | 64 | FK | 联系人ID | CNT_20231105001 | +| lead_id | VARCHAR | 64 | FK | 线索ID | LEAD_20231105001 | +| account_id | VARCHAR | 64 | FK | 企业ID | ACC_20231105001 | +| event_time | DATETIME | | NOT NULL | 事件发生时间 | 2023-11-05 14:35:20 | +| session_id | VARCHAR | 64 | | 会话ID | SESSION_20231105001 | +| device_type | VARCHAR | 50 | | 设备类型 | DESKTOP、MOBILE、TABLET | +| browser | VARCHAR | 50 | | 浏览器 | Chrome、Safari、Firefox | +| os | VARCHAR | 50 | | 操作系统 | Windows 10、iOS 16 | +| ip_address | VARCHAR | 50 | | IP地址 | 192.168.1.1 | +| page_url | VARCHAR | 1000 | | 页面URL | https://example.com/product | +| referrer_url | VARCHAR | 1000 | | 来源URL | https://baidu.com/search | +| created_at | DATETIME | | NOT NULL | 创建时间 | 2023-11-05 14:35:20 | --- -## Entity Summary - -### Entity Statistics - -| Category | Entity Name | Count | Description | -|----------|-------------|-------|-------------| -| Customer Entities | Account, Contact, Lead | 3 | Core customer data | -| Summary Entities | AccountSummary, ContactSummary, LeadSummary | 3 | Aggregated statistics | -| Business Entities | Opportunity, Product, ProductCategory | 3 | Business transaction data | -| Marketing Entities | Campaign, Channel | 2 | Marketing management | -| Interaction Entities | Touchpoint, Event | 2 | Customer interaction data | -| Relationship Entities | AccountContactRelation, AccountRelation, OpportunityProduct, CampaignMember, TagRelation, SegmentMember | 6 | Entity relationships | -| Identity Entities | AccountChannelIdentity, ContactChannelIdentity, LeadChannelIdentity | 3 | Multi-channel identity | -| Analytics Entities | Segment, Tag, ScoreModel, ScoreRecord, ScoreHistory, Attribution, TouchpointAttribution, CustomerJourney, JourneyStage | 9 | Analytics and insights | -| Performance Entities | ChannelPerformance, CampaignPerformance | 2 | Performance metrics | -| History Entities | OpportunityStageHistory | 1 | Change history tracking | -| Support Entities | Industry, TouchpointAttachment | 2 | Supporting data | -| Total | | 36 | Complete B2B CDP coverage | +## 实体统计汇总 + +### 实体分类统计 + +| 实体分类 | 实体名称 | 数量 | 说明 | +|---------|---------|------|------| +| 客户主体实体 | Account、Contact、Lead | 3 | 核心客户数据 | +| 汇总数据实体 | AccountSummary、ContactSummary、LeadSummary | 3 | 统计汇总数据 | +| 业务实体 | Opportunity、Product、ProductCategory | 3 | 业务交易数据 | +| 营销实体 | Campaign、Channel | 2 | 营销活动管理 | +| 交互实体 | Touchpoint、Event | 2 | 客户互动数据 | +| 关系实体 | AccountContactRelation、AccountRelation、OpportunityProduct、CampaignMember、TagRelation、SegmentMember | 6 | 实体关系映射 | +| 身份实体 | AccountChannelIdentity、ContactChannelIdentity、LeadChannelIdentity | 3 | 全渠道身份映射 | +| 分析实体 | Segment、Tag、ScoreModel、ScoreRecord、ScoreHistory、Attribution、TouchpointAttribution、CustomerJourney、JourneyStage | 9 | 数据分析洞察 | +| 性能实体 | ChannelPerformance、CampaignPerformance | 2 | 绩效统计 | +| 历史实体 | OpportunityStageHistory | 1 | 变更历史追踪 | +| 支撑实体 | Industry、TouchpointAttachment | 2 | 基础支撑数据 | +| 合计 | | 36 | 完整覆盖B2B CDP业务场景 | --- -## Summary Table Design +## 汇总表设计说明 -### Design Principles +### 设计原则 -**Performance Optimization** -- Avoid frequent multi-table JOINs and aggregations -- Improve query performance for 360-degree customer view -- Reduce database load +**性能优化** +- 避免频繁的多表JOIN和聚合计算 +- 提升客户360度视图的查询速度 +- 降低数据库负载 -**Business Requirements** -- Quickly display customer health scores -- Real-time customer value metrics -- Support customer alerting and monitoring +**业务需求** +- 快速展示客户健康度评分 +- 实时显示客户价值指标 +- 支持客户预警和监控 -**Data Consistency** -- Unified calculation methodology -- Scheduled batch updates ensure accuracy -- Avoid real-time calculation inconsistencies +**数据一致性** +- 统一的计算口径 +- 定时批量更新保证数据准确性 +- 避免实时计算的结果不一致 -### Update Mechanisms +### 更新机制 -**Update Triggers:** -- Scheduled Jobs: Hourly full update -- Real-time Triggers: Immediate update on critical events -- Manual Refresh: On-demand administrator trigger +**更新触发条件：** +- 定时任务：每小时执行一次全量更新 +- 实时触发：关键业务事件发生时立即更新（如商机赢单、联系人新增） +- 手动刷新：管理员按需触发更新 -**Update Strategy:** -- Incremental Update: Only changed records -- Full Update: Periodic complete recalculation -- Async Update: Message queue to avoid blocking +**更新策略：** +- 增量更新：仅更新有变化的记录 +- 全量更新：定期执行完整重算 +- 异步更新：使用消息队列避免阻塞主业务流程 --- -## Index Design +## 索引设计建议 -### Account Table Indexes +### Account表索引 ```sql --- Primary key +-- 主键索引 PRIMARY KEY (account_id); --- Unique indexes +-- 唯一索引 CREATE UNIQUE INDEX uk_account_credit_code ON Account(unified_social_credit_code) WHERE unified_social_credit_code IS NOT NULL; --- Query indexes +-- 业务查询索引 CREATE INDEX idx_account_name ON Account(account_name); CREATE INDEX idx_account_status ON Account(account_status); CREATE INDEX idx_account_owner ON Account(owner_user_id); CREATE INDEX idx_account_created ON Account(created_at DESC); CREATE INDEX idx_account_industry ON Account(industry_id); --- Composite indexes +-- 组合索引 CREATE INDEX idx_account_type_status ON Account(account_type, account_status); CREATE INDEX idx_account_location ON Account(province, city); ``` -### AccountSummary Table Indexes +### AccountSummary表索引 ```sql --- Primary and unique +-- 主键和唯一索引 PRIMARY KEY (summary_id); CREATE UNIQUE INDEX uk_summary_account ON AccountSummary(account_id); --- Query indexes +-- 查询索引 CREATE INDEX idx_summary_health_score ON AccountSummary(health_score DESC); CREATE INDEX idx_summary_last_activity ON AccountSummary(last_activity_at DESC); CREATE INDEX idx_summary_total_revenue ON AccountSummary(total_revenue DESC); CREATE INDEX idx_summary_win_rate ON AccountSummary(win_rate DESC); ``` -### Contact Table Indexes +### Contact表索引 ```sql --- Primary key +-- 主键索引 PRIMARY KEY (contact_id); --- Unique indexes +-- 唯一索引 CREATE UNIQUE INDEX uk_contact_phone ON Contact(mobile_phone) WHERE mobile_phone IS NOT NULL; CREATE UNIQUE INDEX uk_contact_email ON Contact(email) WHERE email IS NOT NULL; --- Query indexes +-- 业务查询索引 CREATE INDEX idx_contact_name ON Contact(contact_name); CREATE INDEX idx_contact_account ON Contact(primary_account_id); CREATE INDEX idx_contact_status ON Contact(contact_status); CREATE INDEX idx_contact_wechat ON Contact(wechat_id); --- Composite indexes +-- 组合索引 CREATE INDEX idx_contact_phone_email ON Contact(mobile_phone, email); ``` -### Event Table Indexes (ClickHouse) +### Event表索引（ClickHouse） ```sql --- Order by key +-- 排序键设计 ORDER BY (channel_id, event_time, contact_id, event_type); --- Partition key +-- 分区键设计 PARTITION BY toYYYYMM(event_time); --- Sample expression +-- 采样表达式 SAMPLE BY cityHash64(event_id); ``` -### Touchpoint Table Indexes +### Touchpoint表索引 ```sql --- Primary key +-- 主键索引 PRIMARY KEY (touchpoint_id); --- Query indexes +-- 查询索引 CREATE INDEX idx_touchpoint_contact ON Touchpoint(contact_id, touchpoint_time DESC); CREATE INDEX idx_touchpoint_lead ON Touchpoint(lead_id, touchpoint_time DESC); CREATE INDEX idx_touchpoint_account ON Touchpoint(account_id, touchpoint_time DESC); @@ -1289,32 +1296,32 @@ CREATE INDEX idx_touchpoint_type ON Touchpoint(touchpoint_type, touchpoint_time --- -## Conclusion - -This B2B CDP entity design provides: - -**Core Capabilities** -- 36 entities covering complete B2B CDP scenarios -- Multi-channel identity resolution -- Summary tables for performance optimization -- Complete customer lifecycle management -- Lead to opportunity conversion tracking - -**Key Features** -- Flexible tagging and segmentation -- Multi-dimensional attribution analysis -- Detailed field design and data dictionary -- Professional database architecture -- Complete index optimization - -**Use Cases** -- Multi-channel customer data integration -- 360-degree customer profile -- Precision marketing and segmentation -- Sales lead management and conversion -- Customer journey analysis -- Marketing attribution analysis -- Customer value assessment -- Customer health monitoring - -Implementation can be phased based on business needs and refined iteratively. +## 总结 + +本B2B CDP实体设计方案提供： + +**核心能力** +- 36个实体，完整覆盖B2B CDP业务场景 +- 全渠道身份映射方案，支持跨渠道客户识别 +- 汇总数据表设计，优化查询性能 +- 完整的客户生命周期管理 +- 从线索到商机的完整转化流程 + +**关键特性** +- 灵活的标签和分群能力 +- 多维度的归因分析能力 +- 详细的字段设计和数据字典 +- 专业的数据库架构建议 +- 完整的索引优化方案 + +**适用场景** +- 全渠道客户数据整合 +- 客户360度画像分析 +- 精准营销和客户分群 +- 销售线索管理和转化 +- 客户旅程分析 +- 营销归因分析 +- 客户价值评估 +- 客户健康度监控 + +根据实际业务需要，可以选择性实现部分实体，并在后续迭代中逐步完善。 From 5e908bc835b08c85edf078f2f67f7bae41a920d5 Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 15:15:01 +0000 Subject: [PATCH 06/10] feat: Add account entity design analysis document Co-authored-by: 13585811473 <13585811473@163.com> --- docs/account-entity-design-analysis.md | 1031 ++++++++++++++++++++++++ 1 file changed, 1031 insertions(+) create mode 100644 docs/account-entity-design-analysis.md diff --git a/docs/account-entity-design-analysis.md b/docs/account-entity-design-analysis.md new file mode 100644 index 000000000..ccbb33fe8 --- /dev/null +++ b/docs/account-entity-design-analysis.md @@ -0,0 +1,1031 @@ +# Account 实体多渠道整合设计方案详解 + +## 目录 +- [设计背景](#设计背景) +- [方案对比](#方案对比) +- [推荐方案详解](#推荐方案详解) +- [实现示例](#实现示例) + +--- + +## 设计背景 + +### 多渠道场景特点 + +在B2B CDP场景下，一个企业账户（Account）的数据来源于多个渠道： + +```mermaid +graph TB + A[企业账户Account
统一社会信用代码91330000MA27XYZ123
阿里巴巴网络技术有限公司] + + A --> B1[官网渠道
注册时间2023-01-15
企业ID: WEB_A001] + A --> B2[企业微信渠道
认证时间2023-02-20
CorpID: wx123456] + A --> B3[线下展会渠道
获取时间2023-03-10
展会编号: EXH_001] + A --> B4[合作伙伴渠道
推荐时间2023-04-05
伙伴ID: PT_A001] + + A --> C1[25个联系人Contact] + A --> C2[8个商机Opportunity] + A --> C3[45个线索Lead] + A --> C4[156个触点Touchpoint] + A --> C5[500万累计收入] + A --> C6[85分健康度评分] +``` + +### 核心设计问题 + +**问题1：基础字段 vs 汇总字段** +- 基础字段：`account_name`、`industry_id`、`province`、`city` 等基本信息 +- 汇总字段：`total_contacts`、`total_opportunities`、`total_revenue`、`health_score` 等统计数据 + +**问题2：实时性 vs 性能** +- 实时计算：每次查询都JOIN多张表计算，准确但慢 +- 预先汇总：定期计算存储，快速但有延迟 + +**问题3：数据一致性** +- 如何保证汇总数据与明细数据一致 +- 如何处理并发更新问题 + +--- + +## 方案对比 + +### 方案一：所有字段都在Account表中（不推荐） + +```mermaid +erDiagram + Account { + varchar account_id PK "主键" + varchar account_name "企业名称" + varchar unified_social_credit_code UK "统一信用代码" + varchar account_type "客户类型" + varchar account_status "账户状态" + varchar industry_id FK "行业ID" + varchar province "省份" + varchar city "城市" + datetime created_at "创建时间" + datetime updated_at "更新时间" + int total_contacts "联系人总数-汇总" + int total_opportunities "商机总数-汇总" + int total_leads "线索总数-汇总" + decimal total_revenue "累计收入-汇总" + decimal lifetime_value "生命周期价值-汇总" + int won_opportunities "赢单数-汇总" + int lost_opportunities "输单数-汇总" + decimal win_rate "赢单率-汇总" + int total_touchpoints "触点总数-汇总" + int health_score "健康度评分-汇总" + datetime last_activity_at "最后活跃时间-汇总" + datetime last_purchase_at "最后购买时间-汇总" + int days_since_last_contact "距上次联系天数-汇总" + } +``` + +**优点：** +- 表结构简单，只有一张表 +- 查询简单，不需要JOIN + +**缺点：** +- 字段过多，单表臃肿（20+个字段） +- 基础字段和汇总字段混在一起，职责不清晰 +- 汇总字段更新频繁，影响主表性能 +- 无法记录历史汇总数据 +- 扩展性差，增加新的汇总维度需要修改表结构 + +**性能问题：** +```sql +-- 每次更新汇总数据都需要锁定Account表 +UPDATE Account +SET total_contacts = (SELECT COUNT(*) FROM Contact WHERE primary_account_id = 'ACC_001'), + total_opportunities = (SELECT COUNT(*) FROM Opportunity WHERE account_id = 'ACC_001'), + total_revenue = (SELECT SUM(amount) FROM Opportunity WHERE account_id = 'ACC_001' AND is_won = true), + health_score = ... -- 复杂计算 +WHERE account_id = 'ACC_001'; + +-- 问题： +-- 1. 多个子查询，性能差 +-- 2. 频繁更新导致行锁竞争 +-- 3. 基础信息查询也会被汇总字段更新阻塞 +``` + +--- + +### 方案二：基础字段和汇总字段分离（推荐） + +```mermaid +erDiagram + Account ||--|| AccountSummary : aggregates + Account ||--o{ AccountChannelIdentity : has + + Account { + varchar account_id PK "主键" + varchar account_name "企业名称" + varchar unified_social_credit_code UK "统一信用代码" + varchar account_type "客户类型" + varchar account_status "账户状态" + varchar account_level "客户等级" + decimal annual_revenue "年营收-企业自身" + int employee_count "员工数-企业自身" + varchar industry_id FK "行业ID" + varchar province "省份" + varchar city "城市" + varchar district "区县" + varchar company_website "企业官网" + varchar account_source "来源渠道" + varchar primary_channel_id FK "主要渠道" + varchar owner_user_id FK "负责人" + datetime created_at "创建时间" + datetime updated_at "更新时间" + varchar lifecycle_stage "生命周期阶段" + } + + AccountSummary { + varchar summary_id PK "主键" + varchar account_id FK UK "账户ID-唯一" + int total_contacts "联系人总数" + int total_opportunities "商机总数" + int total_leads "线索总数" + decimal total_revenue "累计收入" + decimal lifetime_value "生命周期价值" + int won_opportunities "赢单数量" + int lost_opportunities "输单数量" + decimal win_rate "赢单率" + int total_touchpoints "触点总数" + int active_campaigns "活跃营销活动数" + int health_score "健康度评分" + datetime last_activity_at "最后活跃时间" + datetime last_purchase_at "最后购买时间" + date first_purchase_date "首次购买日期" + date latest_opportunity_date "最新商机日期" + int days_since_last_contact "距上次联系天数" + datetime calculated_at "计算时间" + datetime updated_at "更新时间" + } + + AccountChannelIdentity { + varchar identity_id PK "主键" + varchar account_id FK "账户ID" + varchar channel_id FK "渠道ID" + varchar channel_account_id "渠道内账户ID" + varchar identity_type "身份类型" + boolean is_verified "是否已验证" + datetime first_seen_at "首次发现时间" + datetime last_seen_at "最后发现时间" + json additional_info "附加信息" + } +``` + +**优点：** +1. **职责清晰** + - Account表：存储企业基础信息（相对稳定） + - AccountSummary表：存储汇总统计数据（频繁变化） + - AccountChannelIdentity表：存储多渠道身份映射 + +2. **性能优化** + - 基础信息查询不受汇总数据更新影响 + - 汇总数据可以异步计算更新 + - 可以针对不同表设置不同的缓存策略 + +3. **扩展性强** + - 增加新的汇总维度只需修改AccountSummary表 + - 可以按时间维度存储历史汇总数据 + - 支持多版本汇总模型 + +4. **数据一致性** + - 通过`calculated_at`字段明确知道汇总数据的计算时间 + - 可以通过对比`calculated_at`和明细数据的`updated_at`判断是否需要重新计算 + +**缺点：** +- 需要JOIN查询才能获取完整数据（但可以通过缓存优化） +- 需要维护汇总计算逻辑 + +--- + +## 推荐方案详解 + +### 1. 表结构设计 + +#### Account 基础表 + +**设计原则：** +- 只存储企业的基础属性信息 +- 这些信息相对稳定，变化频率低 +- 直接来源于企业自身，不是统计计算得出 + +```sql +CREATE TABLE Account ( + -- 主键和唯一标识 + account_id VARCHAR(64) PRIMARY KEY COMMENT '账户唯一标识', + account_name VARCHAR(200) NOT NULL COMMENT '企业名称', + unified_social_credit_code VARCHAR(18) UNIQUE COMMENT '统一社会信用代码', + + -- 分类和状态 + account_type VARCHAR(50) NOT NULL COMMENT '客户类型：CUSTOMER/PARTNER/PROSPECT/COMPETITOR', + account_status VARCHAR(50) NOT NULL COMMENT '账户状态：ACTIVE/DORMANT/CHURNED/BLACKLIST', + account_level VARCHAR(50) COMMENT '客户等级：STRATEGIC/IMPORTANT/NORMAL', + lifecycle_stage VARCHAR(50) COMMENT '生命周期阶段：AWARENESS/CONSIDERATION/DECISION/RETENTION/EXPANSION', + + -- 企业基本信息（来自企业自身，非统计） + annual_revenue DECIMAL(18,2) COMMENT '企业年营收（万元）-来自企业工商信息', + employee_count INT COMMENT '企业员工数-来自企业工商信息', + industry_id VARCHAR(64) COMMENT '行业分类ID', + + -- 地址信息 + province VARCHAR(50) COMMENT '省份', + city VARCHAR(50) COMMENT '城市', + district VARCHAR(50) COMMENT '区县', + company_address VARCHAR(500) COMMENT '详细地址', + company_website VARCHAR(500) COMMENT '企业官网', + + -- 来源和归属 + account_source VARCHAR(100) COMMENT '来源：WEBSITE/EXHIBITION/PARTNER/COLD_CALL', + primary_channel_id VARCHAR(64) COMMENT '主要渠道ID', + owner_user_id VARCHAR(64) COMMENT '负责销售人员ID', + + -- 时间戳 + created_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP COMMENT '创建时间', + updated_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT '更新时间', + + -- 扩展字段 + custom_fields JSON COMMENT '自定义扩展字段', + + -- 索引 + INDEX idx_account_name (account_name), + INDEX idx_account_status (account_status), + INDEX idx_account_type_status (account_type, account_status), + INDEX idx_account_owner (owner_user_id), + INDEX idx_account_industry (industry_id), + INDEX idx_account_city (province, city), + INDEX idx_account_created (created_at) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='企业账户基础信息表'; +``` + +#### AccountSummary 汇总表 + +**设计原则：** +- 存储所有统计计算的汇总数据 +- 数据从其他表聚合计算得出 +- 可以定期重新计算保证准确性 + +```sql +CREATE TABLE AccountSummary ( + -- 主键 + summary_id VARCHAR(64) PRIMARY KEY COMMENT '汇总记录ID', + account_id VARCHAR(64) NOT NULL UNIQUE COMMENT '账户ID-唯一', + + -- 关联数据统计 + total_contacts INT DEFAULT 0 COMMENT '关联联系人总数', + total_opportunities INT DEFAULT 0 COMMENT '商机总数', + total_leads INT DEFAULT 0 COMMENT '线索总数', + + -- 收入统计 + total_revenue DECIMAL(18,2) DEFAULT 0 COMMENT '累计成交收入（元）', + lifetime_value DECIMAL(18,2) DEFAULT 0 COMMENT '客户生命周期价值LTV（元）', + current_year_revenue DECIMAL(18,2) DEFAULT 0 COMMENT '本年度收入（元）', + last_year_revenue DECIMAL(18,2) DEFAULT 0 COMMENT '去年收入（元）', + + -- 商机统计 + won_opportunities INT DEFAULT 0 COMMENT '赢单商机数', + lost_opportunities INT DEFAULT 0 COMMENT '输单商机数', + open_opportunities INT DEFAULT 0 COMMENT '进行中商机数', + win_rate DECIMAL(5,2) DEFAULT 0 COMMENT '赢单率（%）', + average_deal_size DECIMAL(18,2) DEFAULT 0 COMMENT '平均成交金额（元）', + + -- 互动统计 + total_touchpoints INT DEFAULT 0 COMMENT '总触点数', + total_events INT DEFAULT 0 COMMENT '总事件数', + active_campaigns INT DEFAULT 0 COMMENT '参与的活跃营销活动数', + + -- 健康度和活跃度 + health_score INT DEFAULT 0 COMMENT '健康度评分（0-100）', + engagement_score INT DEFAULT 0 COMMENT '参与度评分（0-100）', + churn_risk_score INT DEFAULT 0 COMMENT '流失风险评分（0-100，越高风险越大）', + + -- 时间维度统计 + last_activity_at DATETIME COMMENT '最后活跃时间（任何互动）', + last_touchpoint_at DATETIME COMMENT '最后触点时间', + last_purchase_at DATETIME COMMENT '最后购买时间', + last_contact_at DATETIME COMMENT '最后主动联系时间', + first_purchase_date DATE COMMENT '首次购买日期', + latest_opportunity_date DATE COMMENT '最新商机日期', + + -- 天数统计 + days_since_last_activity INT DEFAULT 0 COMMENT '距上次活跃天数', + days_since_last_contact INT DEFAULT 0 COMMENT '距上次联系天数', + days_since_last_purchase INT DEFAULT 0 COMMENT '距上次购买天数', + customer_tenure_days INT DEFAULT 0 COMMENT '成为客户的天数', + + -- 元数据 + calculated_at DATETIME NOT NULL COMMENT '汇总计算时间', + updated_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT '更新时间', + + -- 外键和索引 + FOREIGN KEY (account_id) REFERENCES Account(account_id) ON DELETE CASCADE, + INDEX idx_summary_health_score (health_score DESC), + INDEX idx_summary_last_activity (last_activity_at DESC), + INDEX idx_summary_total_revenue (total_revenue DESC), + INDEX idx_summary_win_rate (win_rate DESC), + INDEX idx_summary_churn_risk (churn_risk_score DESC), + INDEX idx_summary_calculated (calculated_at) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='企业账户汇总统计表'; +``` + +#### AccountChannelIdentity 多渠道身份映射表 + +**设计原则：** +- 记录同一个企业在不同渠道的身份标识 +- 支持身份识别和合并 +- 记录首次和最后发现时间 + +```sql +CREATE TABLE AccountChannelIdentity ( + -- 主键 + identity_id VARCHAR(64) PRIMARY KEY COMMENT '身份记录ID', + + -- 关联 + account_id VARCHAR(64) NOT NULL COMMENT '统一账户ID', + channel_id VARCHAR(64) NOT NULL COMMENT '渠道ID', + + -- 渠道内身份 + channel_account_id VARCHAR(200) NOT NULL COMMENT '在该渠道内的账户ID', + identity_type VARCHAR(50) COMMENT '身份类型：CORP_ID/WEBSITE_ID/PARTNER_ID', + + -- 验证状态 + is_verified BOOLEAN DEFAULT FALSE COMMENT '是否已验证', + verification_method VARCHAR(50) COMMENT '验证方式：MANUAL/AUTO/CREDIT_CODE', + + -- 时间记录 + first_seen_at DATETIME NOT NULL COMMENT '首次在该渠道发现时间', + last_seen_at DATETIME NOT NULL COMMENT '最后在该渠道活跃时间', + + -- 附加信息 + additional_info JSON COMMENT '渠道特有的附加信息', + + -- 索引 + UNIQUE KEY uk_account_channel (account_id, channel_id), + INDEX idx_channel_account (channel_id, channel_account_id), + INDEX idx_account (account_id), + INDEX idx_first_seen (first_seen_at), + + FOREIGN KEY (account_id) REFERENCES Account(account_id) ON DELETE CASCADE, + FOREIGN KEY (channel_id) REFERENCES Channel(channel_id) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='企业账户多渠道身份映射表'; +``` + +--- + +### 2. 数据关系示意图 + +```mermaid +graph TB + subgraph 基础数据层 + A[Account基础表
account_id: ACC_001
account_name: 阿里巴巴
unified_social_credit_code: 9133...123] + end + + subgraph 渠道身份层 + CI1[AccountChannelIdentity
channel: 官网
channel_account_id: WEB_A001
first_seen: 2023-01-15] + CI2[AccountChannelIdentity
channel: 企业微信
channel_account_id: wx123456
first_seen: 2023-02-20] + CI3[AccountChannelIdentity
channel: 线下展会
channel_account_id: EXH_001
first_seen: 2023-03-10] + end + + subgraph 关联明细数据层 + C[Contact表
25条联系人记录] + O[Opportunity表
8条商机记录
5条赢单，2条输单] + L[Lead表
45条线索记录] + T[Touchpoint表
156条触点记录] + end + + subgraph 汇总数据层 + AS[AccountSummary汇总表
summary_id: ACCS_001
account_id: ACC_001
total_contacts: 25
total_opportunities: 8
won_opportunities: 5
total_revenue: 5000000
win_rate: 62.5
health_score: 85
calculated_at: 2023-11-05 16:00] + end + + A --> CI1 + A --> CI2 + A --> CI3 + + A --> C + A --> O + A --> L + A --> T + + C --> AS + O --> AS + L --> AS + T --> AS + + A -.1对1关系.-> AS +``` + +--- + +### 3. 汇总计算策略 + +#### 计算触发时机 + +```mermaid +flowchart TD + Start[触发汇总计算] --> Type{触发类型} + + Type -->|定时任务| Schedule[定时任务触发
每小时执行一次] + Type -->|实时事件| Event[关键业务事件触发] + Type -->|手动刷新| Manual[管理员手动触发] + + Schedule --> SelectAccounts[选择需要更新的Account
条件：距上次计算超过1小时] + Event --> SelectAccount[选择特定Account
如：商机赢单触发] + Manual --> SelectManual[选择指定Account] + + SelectAccounts --> Calculate[执行汇总计算] + SelectAccount --> Calculate + SelectManual --> Calculate + + Calculate --> Query1[查询Contact数量] + Calculate --> Query2[查询Opportunity统计] + Calculate --> Query3[查询Lead数量] + Calculate --> Query4[查询Touchpoint统计] + Calculate --> Query5[计算健康度评分] + + Query1 --> Aggregate[聚合所有统计数据] + Query2 --> Aggregate + Query3 --> Aggregate + Query4 --> Aggregate + Query5 --> Aggregate + + Aggregate --> Upsert[写入AccountSummary表
INSERT或UPDATE] + Upsert --> UpdateTime[更新calculated_at时间戳] + UpdateTime --> Cache[更新Redis缓存] + Cache --> End[完成] +``` + +#### 计算SQL示例 + +```sql +-- 完整的AccountSummary计算和更新 +INSERT INTO AccountSummary ( + summary_id, + account_id, + total_contacts, + total_opportunities, + total_leads, + total_revenue, + won_opportunities, + lost_opportunities, + open_opportunities, + win_rate, + total_touchpoints, + health_score, + last_activity_at, + last_purchase_at, + days_since_last_contact, + calculated_at +) +SELECT + CONCAT('ACCS_', account_id) as summary_id, + a.account_id, + + -- 联系人统计 + COALESCE(c.contact_count, 0) as total_contacts, + + -- 商机统计 + COALESCE(o.opportunity_count, 0) as total_opportunities, + + -- 线索统计 + COALESCE(l.lead_count, 0) as total_leads, + + -- 收入统计 + COALESCE(o.total_revenue, 0) as total_revenue, + COALESCE(o.won_count, 0) as won_opportunities, + COALESCE(o.lost_count, 0) as lost_opportunities, + COALESCE(o.open_count, 0) as open_opportunities, + + -- 赢单率 + CASE + WHEN COALESCE(o.won_count, 0) + COALESCE(o.lost_count, 0) > 0 + THEN ROUND(COALESCE(o.won_count, 0) * 100.0 / (COALESCE(o.won_count, 0) + COALESCE(o.lost_count, 0)), 2) + ELSE 0 + END as win_rate, + + -- 触点统计 + COALESCE(t.touchpoint_count, 0) as total_touchpoints, + + -- 健康度评分（简化版） + CASE + WHEN DATEDIFF(NOW(), COALESCE(t.last_touchpoint, a.created_at)) <= 7 THEN 90 + WHEN DATEDIFF(NOW(), COALESCE(t.last_touchpoint, a.created_at)) <= 30 THEN 70 + WHEN DATEDIFF(NOW(), COALESCE(t.last_touchpoint, a.created_at)) <= 90 THEN 50 + ELSE 30 + END as health_score, + + -- 时间统计 + COALESCE(t.last_touchpoint, a.created_at) as last_activity_at, + o.last_purchase as last_purchase_at, + DATEDIFF(NOW(), COALESCE(t.last_touchpoint, a.created_at)) as days_since_last_contact, + + NOW() as calculated_at + +FROM Account a + +-- 联系人统计 +LEFT JOIN ( + SELECT + primary_account_id, + COUNT(*) as contact_count + FROM Contact + WHERE contact_status = 'ACTIVE' + GROUP BY primary_account_id +) c ON a.account_id = c.primary_account_id + +-- 商机统计 +LEFT JOIN ( + SELECT + account_id, + COUNT(*) as opportunity_count, + SUM(CASE WHEN is_won = TRUE THEN amount ELSE 0 END) as total_revenue, + SUM(CASE WHEN is_won = TRUE THEN 1 ELSE 0 END) as won_count, + SUM(CASE WHEN is_lost = TRUE THEN 1 ELSE 0 END) as lost_count, + SUM(CASE WHEN is_won = FALSE AND is_lost = FALSE THEN 1 ELSE 0 END) as open_count, + MAX(CASE WHEN is_won = TRUE THEN actual_close_date END) as last_purchase + FROM Opportunity + GROUP BY account_id +) o ON a.account_id = o.account_id + +-- 线索统计 +LEFT JOIN ( + SELECT + converted_account_id, + COUNT(*) as lead_count + FROM Lead + WHERE converted_account_id IS NOT NULL + GROUP BY converted_account_id +) l ON a.account_id = l.converted_account_id + +-- 触点统计 +LEFT JOIN ( + SELECT + account_id, + COUNT(*) as touchpoint_count, + MAX(touchpoint_time) as last_touchpoint + FROM Touchpoint + GROUP BY account_id +) t ON a.account_id = t.account_id + +WHERE a.account_id = 'ACC_001' -- 可以批量处理多个账户 + +ON DUPLICATE KEY UPDATE + total_contacts = VALUES(total_contacts), + total_opportunities = VALUES(total_opportunities), + total_leads = VALUES(total_leads), + total_revenue = VALUES(total_revenue), + won_opportunities = VALUES(won_opportunities), + lost_opportunities = VALUES(lost_opportunities), + open_opportunities = VALUES(open_opportunities), + win_rate = VALUES(win_rate), + total_touchpoints = VALUES(total_touchpoints), + health_score = VALUES(health_score), + last_activity_at = VALUES(last_activity_at), + last_purchase_at = VALUES(last_purchase_at), + days_since_last_contact = VALUES(days_since_last_contact), + calculated_at = VALUES(calculated_at), + updated_at = NOW(); +``` + +--- + +### 4. 查询使用场景 + +#### 场景1：查询Account基础信息（高频，快速） + +```sql +-- 只需要基础信息，不需要汇总数据 +-- 查询速度快，不受汇总数据影响 +SELECT + account_id, + account_name, + account_type, + account_status, + industry_id, + province, + city, + owner_user_id +FROM Account +WHERE account_id = 'ACC_001'; +``` + +#### 场景2：查询Account完整信息（包含汇总） + +```sql +-- 需要基础信息 + 汇总数据 +-- 使用LEFT JOIN，即使汇总数据不存在也能查到基础信息 +SELECT + a.account_id, + a.account_name, + a.account_type, + a.account_status, + a.industry_id, + a.province, + a.city, + a.owner_user_id, + -- 汇总数据 + COALESCE(s.total_contacts, 0) as total_contacts, + COALESCE(s.total_opportunities, 0) as total_opportunities, + COALESCE(s.total_revenue, 0) as total_revenue, + COALESCE(s.health_score, 0) as health_score, + COALESCE(s.win_rate, 0) as win_rate, + s.last_activity_at, + s.calculated_at +FROM Account a +LEFT JOIN AccountSummary s ON a.account_id = s.account_id +WHERE a.account_id = 'ACC_001'; +``` + +#### 场景3：客户列表查询（按汇总指标排序和筛选） + +```sql +-- 查询健康度评分低于60分的客户，按收入排序 +SELECT + a.account_id, + a.account_name, + a.account_status, + a.owner_user_id, + s.health_score, + s.total_revenue, + s.win_rate, + s.last_activity_at, + s.days_since_last_contact +FROM Account a +INNER JOIN AccountSummary s ON a.account_id = s.account_id +WHERE a.account_status = 'ACTIVE' + AND s.health_score < 60 +ORDER BY s.total_revenue DESC +LIMIT 50; +``` + +#### 场景4：客户360度视图（完整数据） + +```sql +-- 查询某个客户的完整360度视图 +-- 包括基础信息、汇总数据、多渠道身份 +SELECT + -- 基础信息 + a.account_id, + a.account_name, + a.account_type, + a.account_status, + a.account_level, + a.industry_id, + a.province, + a.city, + a.owner_user_id, + a.lifecycle_stage, + + -- 汇总统计 + s.total_contacts, + s.total_opportunities, + s.total_revenue, + s.lifetime_value, + s.won_opportunities, + s.win_rate, + s.health_score, + s.last_activity_at, + s.days_since_last_contact, + s.calculated_at, + + -- 多渠道身份 + GROUP_CONCAT( + CONCAT(ch.channel_name, ':', ci.channel_account_id) + ORDER BY ci.first_seen_at + SEPARATOR '; ' + ) as channel_identities + +FROM Account a +LEFT JOIN AccountSummary s ON a.account_id = s.account_id +LEFT JOIN AccountChannelIdentity ci ON a.account_id = ci.account_id +LEFT JOIN Channel ch ON ci.channel_id = ch.channel_id +WHERE a.account_id = 'ACC_001' +GROUP BY a.account_id; +``` + +--- + +### 5. 缓存策略 + +#### Redis缓存结构 + +```python +# 缓存键设计 +# 基础信息缓存（较长TTL，1小时） +account:basic:{account_id} -> JSON(Account基础字段) + +# 汇总数据缓存（较短TTL，5分钟） +account:summary:{account_id} -> JSON(AccountSummary所有字段) + +# 完整数据缓存（综合TTL，15分钟） +account:full:{account_id} -> JSON(Account + AccountSummary) + +# 示例数据 +account:basic:ACC_001 = { + "account_id": "ACC_001", + "account_name": "阿里巴巴网络技术有限公司", + "account_type": "CUSTOMER", + "account_status": "ACTIVE", + "industry_id": "IND_001", + "province": "浙江省", + "city": "杭州市", + "owner_user_id": "USER_001" +} + +account:summary:ACC_001 = { + "summary_id": "ACCS_001", + "account_id": "ACC_001", + "total_contacts": 25, + "total_opportunities": 8, + "total_revenue": 5000000.00, + "won_opportunities": 5, + "win_rate": 62.5, + "health_score": 85, + "last_activity_at": "2023-11-05 14:30:00", + "calculated_at": "2023-11-05 16:00:00" +} +``` + +--- + +## 实现示例 + +### 完整代码示例（Python + SQLAlchemy） + +```python +from sqlalchemy import Column, String, Integer, Decimal, DateTime, Boolean, JSON, ForeignKey +from sqlalchemy.orm import relationship +from datetime import datetime + +# Account基础表模型 +class Account(Base): + __tablename__ = 'account' + + account_id = Column(String(64), primary_key=True) + account_name = Column(String(200), nullable=False) + unified_social_credit_code = Column(String(18), unique=True) + account_type = Column(String(50), nullable=False) + account_status = Column(String(50), nullable=False) + account_level = Column(String(50)) + industry_id = Column(String(64)) + province = Column(String(50)) + city = Column(String(50)) + owner_user_id = Column(String(64)) + created_at = Column(DateTime, default=datetime.now) + updated_at = Column(DateTime, default=datetime.now, onupdate=datetime.now) + lifecycle_stage = Column(String(50)) + + # 关系 + summary = relationship("AccountSummary", uselist=False, back_populates="account") + channel_identities = relationship("AccountChannelIdentity", back_populates="account") + contacts = relationship("Contact", back_populates="account") + opportunities = relationship("Opportunity", back_populates="account") + +# AccountSummary汇总表模型 +class AccountSummary(Base): + __tablename__ = 'account_summary' + + summary_id = Column(String(64), primary_key=True) + account_id = Column(String(64), ForeignKey('account.account_id'), unique=True, nullable=False) + + # 统计字段 + total_contacts = Column(Integer, default=0) + total_opportunities = Column(Integer, default=0) + total_leads = Column(Integer, default=0) + total_revenue = Column(Decimal(18, 2), default=0) + lifetime_value = Column(Decimal(18, 2), default=0) + won_opportunities = Column(Integer, default=0) + lost_opportunities = Column(Integer, default=0) + win_rate = Column(Decimal(5, 2), default=0) + total_touchpoints = Column(Integer, default=0) + health_score = Column(Integer, default=0) + + # 时间字段 + last_activity_at = Column(DateTime) + last_purchase_at = Column(DateTime) + days_since_last_contact = Column(Integer, default=0) + + # 元数据 + calculated_at = Column(DateTime, nullable=False) + updated_at = Column(DateTime, default=datetime.now, onupdate=datetime.now) + + # 关系 + account = relationship("Account", back_populates="summary") + +# AccountChannelIdentity渠道身份表模型 +class AccountChannelIdentity(Base): + __tablename__ = 'account_channel_identity' + + identity_id = Column(String(64), primary_key=True) + account_id = Column(String(64), ForeignKey('account.account_id'), nullable=False) + channel_id = Column(String(64), nullable=False) + channel_account_id = Column(String(200), nullable=False) + identity_type = Column(String(50)) + is_verified = Column(Boolean, default=False) + first_seen_at = Column(DateTime, nullable=False) + last_seen_at = Column(DateTime, nullable=False) + additional_info = Column(JSON) + + # 关系 + account = relationship("Account", back_populates="channel_identities") + + +# 业务服务类 +class AccountService: + def __init__(self, db_session, redis_client): + self.db = db_session + self.redis = redis_client + + def get_account_full(self, account_id: str) -> dict: + """获取Account完整信息（包含汇总数据）""" + # 1. 尝试从缓存获取 + cache_key = f"account:full:{account_id}" + cached = self.redis.get(cache_key) + if cached: + return json.loads(cached) + + # 2. 从数据库查询 + account = self.db.query(Account).filter_by(account_id=account_id).first() + if not account: + return None + + # 3. 构建返回数据 + result = { + # 基础信息 + "account_id": account.account_id, + "account_name": account.account_name, + "account_type": account.account_type, + "account_status": account.account_status, + "industry_id": account.industry_id, + "province": account.province, + "city": account.city, + "owner_user_id": account.owner_user_id, + + # 汇总数据 + "summary": { + "total_contacts": account.summary.total_contacts if account.summary else 0, + "total_opportunities": account.summary.total_opportunities if account.summary else 0, + "total_revenue": float(account.summary.total_revenue) if account.summary else 0, + "health_score": account.summary.health_score if account.summary else 0, + "win_rate": float(account.summary.win_rate) if account.summary else 0, + "last_activity_at": account.summary.last_activity_at.isoformat() if account.summary and account.summary.last_activity_at else None, + "calculated_at": account.summary.calculated_at.isoformat() if account.summary and account.summary.calculated_at else None, + }, + + # 多渠道身份 + "channel_identities": [ + { + "channel_id": ci.channel_id, + "channel_account_id": ci.channel_account_id, + "is_verified": ci.is_verified, + "first_seen_at": ci.first_seen_at.isoformat() + } + for ci in account.channel_identities + ] + } + + # 4. 写入缓存（15分钟） + self.redis.setex(cache_key, 900, json.dumps(result)) + + return result + + def calculate_account_summary(self, account_id: str): + """计算并更新Account汇总数据""" + from sqlalchemy import func + + # 1. 统计联系人数量 + total_contacts = self.db.query(func.count(Contact.contact_id))\ + .filter(Contact.primary_account_id == account_id)\ + .filter(Contact.contact_status == 'ACTIVE')\ + .scalar() or 0 + + # 2. 统计商机数据 + opp_stats = self.db.query( + func.count(Opportunity.opportunity_id).label('total'), + func.sum(case((Opportunity.is_won == True, Opportunity.amount), else_=0)).label('revenue'), + func.sum(case((Opportunity.is_won == True, 1), else_=0)).label('won'), + func.sum(case((Opportunity.is_lost == True, 1), else_=0)).label('lost'), + ).filter(Opportunity.account_id == account_id).first() + + total_opportunities = opp_stats.total or 0 + total_revenue = opp_stats.revenue or 0 + won_opportunities = opp_stats.won or 0 + lost_opportunities = opp_stats.lost or 0 + + # 3. 计算赢单率 + win_rate = 0 + if won_opportunities + lost_opportunities > 0: + win_rate = round(won_opportunities * 100.0 / (won_opportunities + lost_opportunities), 2) + + # 4. 统计触点数量 + total_touchpoints = self.db.query(func.count(Touchpoint.touchpoint_id))\ + .filter(Touchpoint.account_id == account_id)\ + .scalar() or 0 + + # 5. 计算健康度评分（简化版） + last_touchpoint = self.db.query(func.max(Touchpoint.touchpoint_time))\ + .filter(Touchpoint.account_id == account_id)\ + .scalar() + + if last_touchpoint: + days_since_last = (datetime.now() - last_touchpoint).days + if days_since_last <= 7: + health_score = 90 + elif days_since_last <= 30: + health_score = 70 + elif days_since_last <= 90: + health_score = 50 + else: + health_score = 30 + else: + health_score = 0 + + # 6. 更新或插入AccountSummary + summary = self.db.query(AccountSummary).filter_by(account_id=account_id).first() + + if summary: + # 更新 + summary.total_contacts = total_contacts + summary.total_opportunities = total_opportunities + summary.total_revenue = total_revenue + summary.won_opportunities = won_opportunities + summary.lost_opportunities = lost_opportunities + summary.win_rate = win_rate + summary.total_touchpoints = total_touchpoints + summary.health_score = health_score + summary.last_activity_at = last_touchpoint + summary.calculated_at = datetime.now() + else: + # 插入 + summary = AccountSummary( + summary_id=f"ACCS_{account_id}", + account_id=account_id, + total_contacts=total_contacts, + total_opportunities=total_opportunities, + total_revenue=total_revenue, + won_opportunities=won_opportunities, + lost_opportunities=lost_opportunities, + win_rate=win_rate, + total_touchpoints=total_touchpoints, + health_score=health_score, + last_activity_at=last_touchpoint, + calculated_at=datetime.now() + ) + self.db.add(summary) + + self.db.commit() + + # 7. 清除缓存 + self.redis.delete(f"account:full:{account_id}") + self.redis.delete(f"account:summary:{account_id}") +``` + +--- + +## 总结建议 + +### 强烈推荐：基础字段和汇总字段分离 + +**核心理由：** + +1. **职责分离原则** + - Account表：企业基础属性（相对稳定，来自企业自身） + - AccountSummary表：业务统计数据（频繁变化，从其他表计算得出） + - 符合单一职责原则 + +2. **性能优化** + - 基础信息查询不受汇总计算影响 + - 可以针对不同表设置不同的缓存策略 + - 汇总计算可以异步执行，不阻塞业务流程 + +3. **扩展性强** + - 新增汇总维度只需修改AccountSummary表 + - 可以支持多版本汇总（历史汇总表） + - 便于进行A/B测试不同的汇总算法 + +4. **数据一致性** + - 通过`calculated_at`字段明确知道数据新鲜度 + - 可以通过定时任务保证最终一致性 + - 支持重新计算修正数据 + +5. **多渠道支持** + - AccountChannelIdentity独立表管理渠道身份 + - 清晰的身份映射关系 + - 便于渠道数据追溯和审计 + +**实施建议：** + +1. **初期阶段** + - 先实现Account和AccountSummary两张表 + - 使用定时任务（每小时）计算汇总数据 + - 在应用层做数据聚合展示 + +2. **优化阶段** + - 引入Redis缓存，缓存热点Account数据 + - 实现关键事件的实时触发更新 + - 优化汇总计算SQL性能 + +3. **进阶阶段** + - 实现汇总计算的增量更新 + - 引入消息队列异步处理 + - 实现历史汇总数据归档 From 81988900d8847ea2ceb958c20763c8af85c7f256 Mon Sep 17 00:00:00 2001 From: joker <13585811473@163.com> Date: Wed, 5 Nov 2025 23:20:20 +0800 Subject: [PATCH 07/10] Update account-entity-design-analysis.md --- docs/account-entity-design-analysis.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/account-entity-design-analysis.md b/docs/account-entity-design-analysis.md index ccbb33fe8..ef4f23b85 100644 --- a/docs/account-entity-design-analysis.md +++ b/docs/account-entity-design-analysis.md @@ -140,7 +140,7 @@ erDiagram AccountSummary { varchar summary_id PK "主键" - varchar account_id FK UK "账户ID-唯一" + varchar account_id FK "账户ID-唯一" int total_contacts "联系人总数" int total_opportunities "商机总数" int total_leads "线索总数" From 2864343c12ef88f89eb9593df2c10a8bd5ba6a94 Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 15:55:41 +0000 Subject: [PATCH 08/10] feat: Add B2B CDP production architecture documentation Co-authored-by: 13585811473 <13585811473@163.com> --- docs/b2b-cdp-production-architecture.md | 1142 +++++++++++++++++++++++ 1 file changed, 1142 insertions(+) create mode 100644 docs/b2b-cdp-production-architecture.md diff --git a/docs/b2b-cdp-production-architecture.md b/docs/b2b-cdp-production-architecture.md new file mode 100644 index 000000000..0241e4a21 --- /dev/null +++ b/docs/b2b-cdp-production-architecture.md @@ -0,0 +1,1142 @@ +# B2B CDP 生产级系统架构设计 + +## 目录 +- [系统整体架构](#系统整体架构) +- [核心实体设计](#核心实体设计) +- [关键时序流程](#关键时序流程) +- [数据流转架构](#数据流转架构) +- [技术实现方案](#技术实现方案) + +--- + +## 系统整体架构 + +### 整体架构分层图 + +```mermaid +graph TB + subgraph DataSource["数据源层 Data Source"] + CRM[CRM系统
Salesforce/SAP] + WEWORK[企业微信
API/Webhook] + WECHAT[微信公众号
表单提交] + INTERNAL[内部系统
ERP/OA] + end + + subgraph Ingestion["数据接入层 Ingestion Gateway"] + API[API Gateway
统一接入] + ADAPTER1[CRM适配器] + ADAPTER2[企微适配器] + ADAPTER3[公众号适配器] + end + + subgraph MQ["消息队列层 Message Queue"] + KAFKA[Kafka Cluster
3 Broker] + T1[data-ingestion] + T2[identity-resolution] + T3[summary-calculation] + T4[notification] + end + + subgraph Processing["数据处理层 Processing Engine"] + CLEANSING[数据清洗服务] + IDENTITY[身份识别引擎] + SCORING[评分引擎] + TAGGING[标签引擎] + SUMMARY[汇总计算服务] + end + + subgraph Storage["数据存储层 Storage"] + PG[(PostgreSQL集群
主从分库分表)] + CH[(ClickHouse集群
行为数据)] + REDIS[(Redis Cluster
缓存)] + ES[(Elasticsearch
搜索)] + end + + subgraph Application["应用服务层 Application"] + LEAD_SVC[Lead服务] + CONTACT_SVC[Contact服务] + ACCOUNT_SVC[Account服务] + OPP_SVC[Opportunity服务] + QUERY_SVC[查询服务] + end + + subgraph Client["客户端 Client"] + WEB[Web Portal] + MOBILE[Mobile App] + OPENAPI[Open API] + end + + CRM --> API + WEWORK --> API + WECHAT --> API + INTERNAL --> API + + API --> ADAPTER1 + API --> ADAPTER2 + API --> ADAPTER3 + + ADAPTER1 --> KAFKA + ADAPTER2 --> KAFKA + ADAPTER3 --> KAFKA + + KAFKA --> T1 + KAFKA --> T2 + KAFKA --> T3 + KAFKA --> T4 + + T1 --> CLEANSING + CLEANSING --> IDENTITY + T2 --> IDENTITY + IDENTITY --> LEAD_SVC + IDENTITY --> CONTACT_SVC + + T3 --> SUMMARY + T4 --> SUMMARY + + LEAD_SVC --> PG + CONTACT_SVC --> PG + ACCOUNT_SVC --> PG + OPP_SVC --> PG + + PG --> REDIS + PG --> ES + CH --> REDIS + + QUERY_SVC --> REDIS + QUERY_SVC --> PG + QUERY_SVC --> CH + QUERY_SVC --> ES + + WEB --> Application + MOBILE --> Application + OPENAPI --> Application +``` + +--- + +### 技术架构详图 + +```mermaid +graph TB + subgraph HA["高可用架构 High Availability"] + direction TB + LB[负载均衡
Nginx/ALB] + + subgraph AppCluster["应用集群"] + APP1[应用节点1] + APP2[应用节点2] + APP3[应用节点3] + end + + subgraph DBCluster["数据库集群"] + PG_MASTER[PostgreSQL主库] + PG_SLAVE1[PostgreSQL从库1] + PG_SLAVE2[PostgreSQL从库2] + end + + subgraph CacheCluster["缓存集群"] + REDIS_M1[Redis Master1] + REDIS_S1[Redis Slave1] + REDIS_M2[Redis Master2] + REDIS_S2[Redis Slave2] + end + + subgraph CHCluster["ClickHouse集群"] + CH1[CH节点1
Shard1 Replica1] + CH2[CH节点2
Shard1 Replica2] + CH3[CH节点3
Shard2 Replica1] + end + + LB --> APP1 + LB --> APP2 + LB --> APP3 + + APP1 --> PG_MASTER + APP2 --> PG_SLAVE1 + APP3 --> PG_SLAVE2 + + PG_MASTER --> PG_SLAVE1 + PG_MASTER --> PG_SLAVE2 + + APP1 --> REDIS_M1 + APP2 --> REDIS_M2 + + REDIS_M1 --> REDIS_S1 + REDIS_M2 --> REDIS_S2 + end +``` + +--- + +## 核心实体设计 + +### Account实体ER图（完整版） + +```mermaid +erDiagram + Account ||--|| AccountSummary : aggregates + Account ||--o{ AccountChannelIdentity : has + Account ||--o{ Contact : owns + Account ||--o{ Opportunity : has + + Account { + varchar account_id PK "分片键" + varchar account_name "企业名称" + varchar unified_social_credit_code UK "信用代码" + varchar account_type "类型" + varchar account_status "状态" + int shard_id "分片ID" + datetime created_at "创建时间" + datetime updated_at "更新时间" + } + + AccountSummary { + varchar summary_id PK + varchar account_id FK UK "关联账户" + int total_contacts "联系人数" + int total_opportunities "商机数" + decimal total_revenue "累计收入" + int health_score "健康度" + datetime calculated_at "计算时间" + } + + AccountChannelIdentity { + varchar identity_id PK + varchar account_id FK "关联账户" + varchar channel_id FK "渠道" + varchar channel_account_id "渠道ID" + boolean is_verified "已验证" + datetime first_seen_at "首次发现" + datetime last_seen_at "最后活跃" + } + + Contact { + varchar contact_id PK "分片键" + varchar contact_name "姓名" + varchar mobile_phone UK "手机号" + varchar email UK "邮箱" + varchar primary_account_id FK "主账户" + int shard_id "分片ID" + datetime created_at "创建时间" + } + + Opportunity { + varchar opportunity_id PK + varchar account_id FK "关联账户" + decimal amount "金额" + varchar stage "阶段" + boolean is_won "已赢单" + datetime created_at "创建时间" + } +``` + +--- + +### Contact实体ER图（完整版） + +```mermaid +erDiagram + Contact ||--|| ContactSummary : aggregates + Contact ||--o{ ContactChannelIdentity : has + Contact ||--o{ AccountContactRelation : belongs_to + Contact ||--o{ Touchpoint : receives + + Contact { + varchar contact_id PK "分片键" + varchar contact_name "姓名" + varchar mobile_phone UK "手机号" + varchar email UK "邮箱" + varchar wechat_id "微信ID" + varchar primary_account_id FK "主账户" + int shard_id "分片ID-hash(mobile_phone)" + datetime created_at "创建时间" + } + + ContactSummary { + varchar summary_id PK + varchar contact_id FK UK "关联联系人" + int total_touchpoints "触点数" + int email_opens "邮件打开" + int engagement_score "参与度" + datetime last_activity_at "最后活跃" + datetime calculated_at "计算时间" + } + + ContactChannelIdentity { + varchar identity_id PK + varchar contact_id FK "关联联系人" + varchar channel_id FK "渠道" + varchar channel_user_id "渠道用户ID" + varchar identity_type "身份类型" + boolean is_verified "已验证" + datetime first_seen_at "首次发现" + } + + AccountContactRelation { + varchar relation_id PK + varchar account_id FK "企业" + varchar contact_id FK "联系人" + varchar role_in_account "角色" + varchar decision_level "决策级别" + boolean is_primary_contact "主联系人" + } + + Touchpoint { + varchar touchpoint_id PK + varchar contact_id FK "联系人" + varchar channel_id FK "渠道" + datetime touchpoint_time "触点时间" + varchar touchpoint_type "类型" + } +``` + +--- + +### Lead实体ER图（完整版） + +```mermaid +erDiagram + Lead ||--|| LeadSummary : aggregates + Lead ||--o{ LeadChannelIdentity : has + Lead ||--o| Contact : converts_to + Lead ||--o| Opportunity : converts_to + + Lead { + varchar lead_id PK "分片键" + varchar lead_name "姓名" + varchar company_name "公司" + varchar mobile_phone "手机号" + varchar email "邮箱" + varchar channel_id FK "来源渠道" + varchar campaign_id FK "来源活动" + varchar lead_status "状态" + int lead_score "评分" + varchar lead_grade "等级" + int shard_id "分片ID" + datetime created_at "创建时间" + datetime converted_at "转化时间" + varchar converted_contact_id FK "转化联系人" + } + + LeadSummary { + varchar summary_id PK + varchar lead_id FK UK "关联线索" + int total_touchpoints "触点数" + int total_events "事件数" + int form_submissions "表单提交" + int days_in_pipeline "管道天数" + datetime last_activity_at "最后活跃" + datetime calculated_at "计算时间" + } + + LeadChannelIdentity { + varchar identity_id PK + varchar lead_id FK "关联线索" + varchar channel_id FK "渠道" + varchar channel_user_id "渠道用户ID" + datetime captured_at "捕获时间" + varchar utm_source "UTM来源" + varchar utm_campaign "UTM活动" + } +``` + +--- + +## 关键时序流程 + +### 1. 多渠道数据采集与身份识别流程 + +```mermaid +sequenceDiagram + participant Source as 数据源
CRM/企微/公众号 + participant Gateway as API Gateway
数据接入网关 + participant Kafka as Kafka
data-ingestion + participant Cleansing as 数据清洗服务 + participant Kafka2 as Kafka
identity-resolution + participant Identity as 身份识别引擎 + participant PG as PostgreSQL
主库 + participant Redis as Redis
缓存 + participant ES as Elasticsearch
搜索引擎 + participant Kafka3 as Kafka
summary-calculation + participant Summary as 汇总计算服务 + + Source->>Gateway: 1.推送数据
Webhook/API + Note over Gateway: 数据格式转换
统一数据模型 + Gateway->>Kafka: 2.写入消息队列
Topic: data-ingestion + Note over Kafka: 数据缓冲
峰值削峰 + + Kafka->>Cleansing: 3.消费消息 + Note over Cleansing: 数据清洗
- 格式规范化
- 必填字段验证
- 数据去重 + + Cleansing->>Kafka2: 4.发送到身份识别队列
Topic: identity-resolution + + Kafka2->>Identity: 5.消费身份识别消息 + + Identity->>PG: 6.查询是否存在
按手机号/邮箱/企业名称 + PG-->>Identity: 7.返回查询结果 + + alt 找到唯一匹配 + Identity->>PG: 8a.更新现有实体
合并数据 + Identity->>PG: 8b.新增ChannelIdentity
记录渠道身份 + Note over Identity: 更新模式
保留历史数据 + else 未找到匹配 + Identity->>PG: 8c.创建新实体
Lead/Contact/Account + Identity->>PG: 8d.创建ChannelIdentity
初始渠道身份 + Note over Identity: 创建模式
生成全局唯一ID + else 找到多个匹配（疑似重复） + Identity->>PG: 8e.写入待审核队列
人工确认 + Note over Identity: 风控模式
防止误合并 + end + + Identity->>Redis: 9.更新缓存
写入最新数据 + Identity->>ES: 10.更新搜索索引
异步同步 + + Identity->>Kafka3: 11.触发汇总计算
Topic: summary-calculation + Note over Identity: 异步触发
不阻塞主流程 + + Kafka3->>Summary: 12.消费汇总消息 + Summary->>PG: 13.查询关联数据
聚合统计 + Summary->>PG: 14.更新Summary表
写入汇总结果 + Summary->>Redis: 15.更新汇总缓存 + + Gateway-->>Source: 16.返回响应
同步快速响应 + Note over Gateway: 响应时间 < 100ms
不等待后续处理 +``` + +**关键设计说明：** + +1. **同步响应快速返回**：API Gateway在写入Kafka后立即返回，响应时间 < 100ms +2. **异步处理解耦**：身份识别、汇总计算都是异步处理，互不影响 +3. **消息队列削峰**：Kafka缓冲高峰流量，保护下游服务 +4. **多级缓存**：Redis缓存热点数据，减少数据库压力 +5. **最终一致性**：汇总数据允许30秒内延迟，保证高可用 + +--- + +### 2. Lead转化为Contact/Opportunity流程 + +```mermaid +sequenceDiagram + participant Sales as 销售人员 + participant App as 应用服务 + participant LeadSvc as Lead服务 + participant ContactSvc as Contact服务 + participant AccountSvc as Account服务 + participant OppSvc as Opportunity服务 + participant PG as PostgreSQL + participant Kafka as Kafka + participant Summary as 汇总计算服务 + participant Notification as 通知服务 + + Sales->>App: 1.确认Lead转化
lead_id + 转化信息 + App->>LeadSvc: 2.调用Lead转化接口 + + LeadSvc->>PG: 3.查询Lead详情
验证状态 + PG-->>LeadSvc: 4.返回Lead数据 + + alt Lead已转化 + LeadSvc-->>App: 5a.返回错误
Lead已转化 + else Lead可转化 + LeadSvc->>PG: 5b.开启事务
BEGIN TRANSACTION + + LeadSvc->>ContactSvc: 6.查询是否存在Contact
按手机号/邮箱 + ContactSvc->>PG: 7.查询Contact表 + PG-->>ContactSvc: 8.返回查询结果 + + alt Contact不存在 + ContactSvc->>PG: 9a.创建新Contact
INSERT Contact + ContactSvc->>PG: 9b.创建ContactChannelIdentity + Note over ContactSvc: 新建Contact
继承Lead的渠道信息 + else Contact已存在 + ContactSvc->>PG: 9c.更新Contact
UPDATE Contact + ContactSvc->>PG: 9d.补充ContactChannelIdentity + Note over ContactSvc: 更新Contact
合并Lead数据 + end + + ContactSvc->>AccountSvc: 10.关联或创建Account + AccountSvc->>PG: 11.查询Account
按企业名称/信用代码 + PG-->>AccountSvc: 12.返回Account + + alt Account不存在 + AccountSvc->>PG: 13a.创建Account + AccountSvc->>PG: 13b.创建AccountChannelIdentity + else Account存在 + AccountSvc->>PG: 13c.关联Contact到Account
AccountContactRelation + end + + LeadSvc->>OppSvc: 14.创建Opportunity
可选步骤 + OppSvc->>PG: 15.创建Opportunity记录 + + LeadSvc->>PG: 16.更新Lead状态
lead_status=CONVERTED + LeadSvc->>PG: 17.记录转化关系
converted_contact_id等 + + LeadSvc->>PG: 18.提交事务
COMMIT + + LeadSvc->>Kafka: 19.发送转化事件
Topic: summary-calculation + Note over Kafka: 触发汇总重算
- LeadSummary
- ContactSummary
- AccountSummary + + Kafka->>Summary: 20.消费转化事件 + Summary->>PG: 21.重新计算汇总数据
批量更新Summary表 + + LeadSvc->>Kafka: 22.发送通知事件
Topic: notification + Kafka->>Notification: 23.消费通知事件 + Notification->>Sales: 24.发送通知
邮件/站内信/企微 + + LeadSvc-->>App: 25.返回转化结果
contact_id, account_id + App-->>Sales: 26.展示转化成功 + end +``` + +**关键设计说明：** + +1. **事务保证一致性**：Lead转化涉及多表操作，使用数据库事务保证原子性 +2. **幂等性设计**：检查Lead状态，防止重复转化 +3. **异步汇总计算**：转化完成后，异步触发汇总重算，不阻塞主流程 +4. **通知解耦**：通过消息队列发送通知，失败可重试 + +--- + +### 3. 实时汇总数据计算流程 + +```mermaid +sequenceDiagram + participant Trigger as 触发源
业务事件/定时任务 + participant Kafka as Kafka
summary-calculation + participant Summary as 汇总计算服务 + participant PG as PostgreSQL + participant Redis as Redis + participant Lock as 分布式锁
Redis + + Trigger->>Kafka: 1.发送汇总任务
account_id/contact_id + Note over Trigger: 触发场景：
- 新增Contact
- 商机赢单
- 新增Touchpoint
- 定时全量刷新 + + Kafka->>Summary: 2.消费汇总消息
并发消费 + + Summary->>Lock: 3.尝试获取分布式锁
key: summary:account:ACC001 + Note over Lock: 防止并发重复计算
锁超时时间: 30秒 + + alt 获取锁失败 + Lock-->>Summary: 4a.锁已被占用 + Summary->>Summary: 4b.跳过本次计算
等待下次触发 + else 获取锁成功 + Lock-->>Summary: 4c.锁获取成功 + + Summary->>PG: 5.查询Account基础信息 + PG-->>Summary: 6.返回Account数据 + + par 并行查询统计数据 + Summary->>PG: 7a.统计Contact数量
SELECT COUNT FROM Contact + Summary->>PG: 7b.统计Opportunity数据
SELECT SUM/COUNT FROM Opportunity + Summary->>PG: 7c.统计Lead数量
SELECT COUNT FROM Lead + Summary->>PG: 7d.统计Touchpoint数据
SELECT COUNT FROM Touchpoint + Summary->>PG: 7e.查询最后活跃时间
SELECT MAX FROM Touchpoint + end + + PG-->>Summary: 8.返回所有统计结果 + + Summary->>Summary: 9.计算汇总指标
- total_contacts
- total_revenue
- win_rate
- health_score等 + Note over Summary: 计算逻辑：
win_rate = won / (won + lost)
health_score = f(last_activity) + + Summary->>PG: 10.写入AccountSummary表
INSERT ON DUPLICATE KEY UPDATE + Note over Summary: Upsert操作
记录calculated_at时间戳 + + Summary->>Redis: 11.更新缓存
SET account:summary:ACC001 + Note over Redis: 缓存TTL: 300秒
下次查询直接从缓存读取 + + Summary->>Lock: 12.释放分布式锁 + Lock-->>Summary: 13.锁释放成功 + + Summary->>Kafka: 14.发送计算完成事件
可选通知下游 + end +``` + +**关键设计说明：** + +1. **分布式锁防重**：使用Redis分布式锁，防止同一Account并发重复计算 +2. **并行查询优化**：统计数据可以并行查询，提升性能 +3. **Upsert操作**：使用`INSERT ON DUPLICATE KEY UPDATE`，支持增量更新 +4. **缓存策略**：计算完成后立即更新缓存，保证数据新鲜度 +5. **异步非阻塞**：整个计算过程异步进行，不影响主业务 + +--- + +### 4. 客户360度查询流程（高性能） + +```mermaid +sequenceDiagram + participant User as 用户 + participant App as 应用服务 + participant Cache as Redis
L1缓存 + participant Query as 查询服务 + participant PG as PostgreSQL
主数据 + participant CH as ClickHouse
行为数据 + participant ES as Elasticsearch
搜索引擎 + + User->>App: 1.查询客户360度视图
account_id: ACC001 + App->>Cache: 2.查询L1缓存
GET account:full:ACC001 + + alt 缓存命中 + Cache-->>App: 3a.返回完整数据
响应时间 < 10ms + App-->>User: 4a.返回结果 + Note over App: 缓存命中率 > 90%
热点数据实时响应 + else 缓存未命中 + Cache-->>App: 3b.缓存MISS + + par 并行查询多数据源 + App->>Query: 4b.查询基础信息 + Query->>PG: 5a.查询Account基础表 + Query->>PG: 5b.查询AccountSummary汇总表 + Query->>PG: 5c.查询Contact列表
LIMIT 10 + Query->>PG: 5d.查询Opportunity列表
LIMIT 10 + + App->>Query: 4c.查询行为数据 + Query->>CH: 6a.查询最近30天Event
GROUP BY event_type + Query->>CH: 6b.查询最近Touchpoint
ORDER BY time DESC LIMIT 20 + + App->>Query: 4d.查询多渠道身份 + Query->>PG: 7.查询AccountChannelIdentity + end + + PG-->>Query: 8.返回主数据 + CH-->>Query: 9.返回行为数据 + + Query->>Query: 10.数据聚合
组装360度视图 + Note over Query: 组装数据结构：
- 基础信息
- 汇总数据
- 关联实体
- 行为统计
- 多渠道身份 + + Query-->>App: 11.返回完整数据 + + App->>Cache: 12.写入缓存
SET account:full:ACC001
TTL: 300秒 + Note over Cache: 缓存策略：
- 热点数据长期缓存
- 冷数据短期缓存
- 更新时主动失效 + + App-->>User: 13.返回结果
响应时间 < 200ms + end +``` + +**关键设计说明：** + +1. **多级缓存**：Redis L1缓存 + 应用内存缓存（可选），缓存命中率 > 90% +2. **并行查询**：同时查询PostgreSQL、ClickHouse、ES，减少总响应时间 +3. **查询优化**： + - Contact/Opportunity只查前10条，避免大结果集 + - ClickHouse查询带时间范围，利用分区裁剪 +4. **缓存失效策略**： + - 写操作主动失效缓存（Cache Aside模式） + - 设置合理的TTL（5分钟） +5. **降级策略**： + - Redis宕机 → 直接查数据库 + - ClickHouse超时 → 降级返回部分数据 + +--- + +### 5. Event事件采集流程（高吞吐） + +```mermaid +sequenceDiagram + participant User as 用户行为 + participant SDK as 埋点SDK
Web/Mobile + participant Gateway as API Gateway + participant Kafka as Kafka
event-stream + participant Flink as Flink
实时计算 + participant CH as ClickHouse
行为数据库 + participant Redis as Redis
实时统计 + + User->>SDK: 1.触发行为事件
页面访问/按钮点击 + SDK->>SDK: 2.批量打包
100条/批或1秒/批 + Note over SDK: 客户端批量上报
减少网络请求 + + SDK->>Gateway: 3.批量上报事件
HTTP POST + Note over Gateway: 接收批量事件
QPS: 10000 + + Gateway->>Gateway: 4.快速验证
签名/限流/格式 + + Gateway->>Kafka: 5.写入Kafka
Topic: event-stream
Partition: hash(user_id) + Note over Kafka: 分区策略：
按user_id哈希
保证同一用户有序 + + Gateway-->>SDK: 6.快速响应
响应时间 < 50ms + + Kafka->>Flink: 7.Flink消费事件流
并行度: 16 + + Flink->>Flink: 8.实时计算
- Session会话识别
- 漏斗分析
- 实时统计 + Note over Flink: 窗口计算：
5秒tumbling window + + Flink->>Redis: 9.写入实时指标
PV/UV/转化率等 + Note over Redis: 实时大屏展示
秒级更新 + + Flink->>CH: 10.批量写入ClickHouse
10000条/批 + Note over CH: 批量插入优化
减少写入次数 + + par 异步关联分析 + Flink->>Flink: 11a.用户画像更新 + Flink->>Flink: 11b.行为标签计算 + Flink->>Kafka: 11c.触发下游任务 + end +``` + +**关键设计说明：** + +1. **客户端批量上报**：SDK端批量打包，减少HTTP请求数 +2. **API快速响应**：写入Kafka后立即返回，响应时间 < 50ms +3. **Kafka分区策略**：按user_id哈希分区，保证同一用户事件有序 +4. **Flink实时计算**： + - Session会话识别 + - 滑动窗口统计 + - 实时漏斗分析 +5. **批量写入ClickHouse**：累积10000条或10秒批量插入，优化写入性能 + +--- + +## 数据流转架构 + +### 数据同步与流转总览 + +```mermaid +graph LR + subgraph Sources["数据源"] + CRM[CRM] + WEWORK[企微] + WECHAT[公众号] + end + + subgraph Ingestion["数据接入"] + API[API Gateway] + end + + subgraph MQ["消息队列"] + K1[data-ingestion] + K2[identity-resolution] + K3[summary-calculation] + end + + subgraph Processing["数据处理"] + CLEAN[数据清洗] + IDENTITY[身份识别] + SUMMARY[汇总计算] + end + + subgraph Storage["数据存储"] + PG[(PostgreSQL)] + CH[(ClickHouse)] + REDIS[(Redis)] + ES[(Elasticsearch)] + end + + subgraph Sync["数据同步"] + CANAL[Canal CDC] + SYNC[数据同步] + end + + CRM --> API + WEWORK --> API + WECHAT --> API + + API --> K1 + K1 --> CLEAN + CLEAN --> K2 + K2 --> IDENTITY + IDENTITY --> PG + IDENTITY --> K3 + K3 --> SUMMARY + SUMMARY --> PG + + PG --> CANAL + CANAL --> ES + CANAL --> REDIS + + PG --> SYNC + SYNC --> CH +``` + +--- + +### PostgreSQL分库分表策略 + +```mermaid +graph TB + subgraph App["应用层"] + APP[应用服务] + end + + subgraph Proxy["分库分表中间件 ShardingSphere"] + SHARDING[分片路由
按account_id/contact_id哈希] + end + + subgraph DB0["数据库0 shard_0"] + PG0_M[(PostgreSQL
Master)] + PG0_S1[(PostgreSQL
Slave1)] + PG0_S2[(PostgreSQL
Slave2)] + end + + subgraph DB1["数据库1 shard_1"] + PG1_M[(PostgreSQL
Master)] + PG1_S1[(PostgreSQL
Slave1)] + end + + subgraph DB15["数据库15 shard_15"] + PG15_M[(PostgreSQL
Master)] + PG15_S1[(PostgreSQL
Slave1)] + end + + APP --> SHARDING + + SHARDING -->|写操作| PG0_M + SHARDING -->|写操作| PG1_M + SHARDING -->|写操作| PG15_M + + SHARDING -->|读操作| PG0_S1 + SHARDING -->|读操作| PG0_S2 + SHARDING -->|读操作| PG1_S1 + SHARDING -->|读操作| PG15_S1 + + PG0_M -.主从复制.-> PG0_S1 + PG0_M -.主从复制.-> PG0_S2 + PG1_M -.主从复制.-> PG1_S1 + PG15_M -.主从复制.-> PG15_S1 +``` + +**分片策略：** + +```sql +-- Account表分片键：account_id +-- 分片规则：hash(account_id) % 16 +-- 分片数：16个库 + +-- Contact表分片键：contact_id（由mobile_phone生成） +-- 分片规则：hash(mobile_phone) % 16 +-- 保证同一手机号的Contact在同一分片 + +-- 关联查询优化： +-- AccountContactRelation表按account_id分片 +-- 保证Account与其Contact在同一分片，避免跨库JOIN +``` + +--- + +## 技术实现方案 + +### 1. 身份识别引擎实现 + +**身份匹配规则（优先级从高到低）：** + +```python +class IdentityResolutionEngine: + """身份识别引擎""" + + def resolve_account(self, data: dict) -> str: + """Account身份识别""" + # 1. 优先级1：统一社会信用代码精确匹配 + if data.get('unified_social_credit_code'): + account = self.find_by_credit_code(data['unified_social_credit_code']) + if account: + return account.account_id + + # 2. 优先级2：企业全称精确匹配 + if data.get('account_name'): + account = self.find_by_exact_name(data['account_name']) + if account: + return account.account_id + + # 3. 优先级3：企业简称+城市模糊匹配 + if data.get('account_name') and data.get('city'): + candidates = self.fuzzy_match_by_name_and_city( + data['account_name'], + data['city'] + ) + if len(candidates) == 1: + return candidates[0].account_id + elif len(candidates) > 1: + # 多个匹配，进入人工审核 + self.send_to_manual_review(data, candidates) + return None + + # 4. 未找到匹配，创建新Account + return self.create_new_account(data) + + def resolve_contact(self, data: dict) -> str: + """Contact身份识别""" + # 1. 优先级1：手机号精确匹配 + if data.get('mobile_phone'): + contact = self.find_by_phone(data['mobile_phone']) + if contact: + return contact.contact_id + + # 2. 优先级2：邮箱精确匹配 + if data.get('email'): + contact = self.find_by_email(data['email']) + if contact: + return contact.contact_id + + # 3. 优先级3：企业微信UserID + if data.get('wework_user_id'): + contact = self.find_by_wework_id(data['wework_user_id']) + if contact: + return contact.contact_id + + # 4. 未找到匹配，创建新Contact + return self.create_new_contact(data) +``` + +--- + +### 2. 汇总计算优化方案 + +**增量计算 vs 全量计算：** + +```python +class SummaryCalculationService: + """汇总计算服务""" + + def calculate_account_summary(self, account_id: str, mode: str = 'incremental'): + """计算Account汇总数据""" + if mode == 'incremental': + # 增量计算：只计算变化的部分 + return self._incremental_calculate(account_id) + else: + # 全量计算：重新统计所有数据 + return self._full_calculate(account_id) + + def _incremental_calculate(self, account_id: str): + """增量计算（推荐）""" + # 1. 获取上次计算结果 + last_summary = self.get_last_summary(account_id) + + # 2. 只统计增量数据（自上次计算后的新增数据） + last_calc_time = last_summary.calculated_at + + # 3. 统计增量 + new_contacts = self.count_new_contacts(account_id, since=last_calc_time) + new_opportunities = self.count_new_opportunities(account_id, since=last_calc_time) + new_revenue = self.sum_new_revenue(account_id, since=last_calc_time) + + # 4. 累加到原有数据 + new_summary = AccountSummary( + account_id=account_id, + total_contacts=last_summary.total_contacts + new_contacts, + total_opportunities=last_summary.total_opportunities + new_opportunities, + total_revenue=last_summary.total_revenue + new_revenue, + # ... 其他字段 + calculated_at=datetime.now() + ) + + return new_summary + + def _full_calculate(self, account_id: str): + """全量计算（定时任务使用）""" + # 重新统计所有数据 + return self._calculate_from_scratch(account_id) +``` + +--- + +### 3. 缓存策略实现 + +**多级缓存架构：** + +```python +class CacheManager: + """缓存管理器""" + + def get_account_full(self, account_id: str) -> dict: + """获取Account完整数据（多级缓存）""" + # L1: 本地内存缓存（可选，使用LRU） + # data = self.local_cache.get(f'account:{account_id}') + # if data: + # return data + + # L2: Redis缓存 + cache_key = f'account:full:{account_id}' + cached = self.redis.get(cache_key) + if cached: + return json.loads(cached) + + # L3: 数据库查询 + account = self.db.query_account_with_summary(account_id) + + # 写入缓存 + self.redis.setex( + cache_key, + 300, # TTL: 5分钟 + json.dumps(account) + ) + + return account + + def invalidate_account_cache(self, account_id: str): + """使Account缓存失效""" + # 删除所有相关缓存 + self.redis.delete(f'account:full:{account_id}') + self.redis.delete(f'account:basic:{account_id}') + self.redis.delete(f'account:summary:{account_id}') +``` + +**缓存更新策略（Cache Aside模式）：** + +```python +# 写操作：先更新数据库，再删除缓存 +def update_account(account_id: str, data: dict): + # 1. 更新数据库 + db.update(account_id, data) + + # 2. 删除缓存（让下次读取时重新加载） + cache.delete(f'account:full:{account_id}') + +# 读操作：先查缓存，缓存miss则查数据库并写缓存 +def get_account(account_id: str): + # 1. 查缓存 + data = cache.get(f'account:full:{account_id}') + if data: + return data + + # 2. 查数据库 + data = db.query(account_id) + + # 3. 写缓存 + cache.set(f'account:full:{account_id}', data, ttl=300) + + return data +``` + +--- + +### 4. 消息队列Topic设计 + +```yaml +Kafka Topic设计: + +1. data-ingestion (数据采集) + - Partition: 16个分区 + - Replication: 3副本 + - Retention: 7天 + - 用途: 接收所有渠道的原始数据 + +2. identity-resolution (身份识别) + - Partition: 16个分区 + - Replication: 3副本 + - Retention: 3天 + - 用途: 身份识别任务队列 + +3. summary-calculation (汇总计算) + - Partition: 16个分区 + - Replication: 3副本 + - Retention: 3天 + - 用途: 汇总计算任务队列 + +4. notification (通知事件) + - Partition: 8个分区 + - Replication: 3副本 + - Retention: 7天 + - 用途: 通知任务（邮件、短信、企微） + +5. event-stream (事件流) + - Partition: 32个分区 + - Replication: 3副本 + - Retention: 30天 + - 用途: 用户行为事件流（高吞吐） +``` + +--- + +### 5. 监控与告警 + +**核心监控指标：** + +```yaml +业务指标: + - 数据采集QPS + - 身份识别成功率 + - 汇总计算延迟 + - 查询P99响应时间 + - 缓存命中率 + +技术指标: + - Kafka消息积压 + - PostgreSQL慢查询 + - Redis内存使用率 + - ClickHouse查询延迟 + - 服务健康检查 + +告警规则: + - Kafka消息积压 > 100万 → P1告警 + - 数据采集失败率 > 1% → P2告警 + - 汇总计算延迟 > 5分钟 → P2告警 + - 查询P99 > 1秒 → P3告警 + - 服务不可用 → P0告警 +``` + +--- + +## 总结 + +### 核心技术特点 + +1. **高可用架构** + - PostgreSQL主从复制 + 读写分离 + - Redis Cluster 3主3从 + - ClickHouse 3副本 + - Kafka 3 broker集群 + +2. **高性能设计** + - PostgreSQL分库分表（16分片） + - 多级缓存（Redis + 本地） + - 异步处理（Kafka解耦） + - 批量写入优化 + +3. **实时性保障** + - 数据采集响应 < 100ms + - 身份识别延迟 < 5秒 + - 汇总计算延迟 < 30秒 + - 查询P99 < 200ms + +4. **数据一致性** + - 事务保证原子性 + - 消息队列保证可靠性 + - 分布式锁防重复 + - 最终一致性设计 + +5. **扩展性** + - 水平扩展（加节点） + - 垂直扩展（加资源） + - 服务解耦（微服务） + - 存储分离（冷热分离） + +### 容量评估 + +``` +数据规模： +- Account: 1000万 +- Contact: 5000万 +- Lead: 2000万/年 +- Event: 10亿/年 +- Touchpoint: 5000万/年 + +存储容量： +- PostgreSQL: 2TB (SSD) +- ClickHouse: 10TB (SSD) +- Redis: 256GB (内存) + +服务器配置： +- 应用服务器: 8核16G * 5台 +- PostgreSQL: 16核64G * 16台（主） + 32台（从） +- ClickHouse: 32核128G * 3台 +- Redis: 16核64G * 6台 +- Kafka: 16核32G * 3台 +``` + +### 后续优化方向 + +1. **短期优化（3个月内）** + - 完善监控告警 + - 优化慢查询 + - 调整缓存策略 + - 压测验证性能 + +2. **中期优化（6个月内）** + - 引入本地缓存（Caffeine） + - 实现数据归档 + - 优化分库分表策略 + - 引入ES辅助查询 + +3. **长期优化（1年内）** + - 冷热数据分离 + - 引入数据湖 + - AI/ML模型集成 + - 实时OLAP分析 From 1495bd425816dc142926b1b9cf46dd56149ae41f Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 16:05:45 +0000 Subject: [PATCH 09/10] feat: Add B2B CDP core architecture design document Co-authored-by: 13585811473 <13585811473@163.com> --- docs/b2b-cdp-core-architecture.md | 1043 +++++++++++++++++++++++++++++ 1 file changed, 1043 insertions(+) create mode 100644 docs/b2b-cdp-core-architecture.md diff --git a/docs/b2b-cdp-core-architecture.md b/docs/b2b-cdp-core-architecture.md new file mode 100644 index 000000000..98a17bf62 --- /dev/null +++ b/docs/b2b-cdp-core-architecture.md @@ -0,0 +1,1043 @@ +# B2B CDP 核心架构设计方案 + +## 目录 +- [系统架构总览](#系统架构总览) +- [核心实体设计](#核心实体设计) +- [关键业务流程](#关键业务流程) +- [技术实现方案](#技术实现方案) + +--- + +## 系统架构总览 + +### 整体架构图 + +```mermaid +graph TB + subgraph 数据源层 + CRM[CRM系统] + WEWORK[企业微信] + WECHAT[微信公众号] + INTERNAL[内部系统] + end + + subgraph 数据接入层 + GATEWAY[API网关] + ADAPTER1[CRM适配器] + ADAPTER2[企微适配器] + ADAPTER3[公众号适配器] + end + + subgraph 消息队列 + KAFKA[Kafka消息队列] + end + + subgraph 数据处理层 + CLEAN[数据清洗服务] + MAPPING[身份映射服务] + MERGE[数据合并服务] + TAG[标签引擎] + SEGMENT[分群引擎] + end + + subgraph 数据存储层 + PG[(PostgreSQL)] + REDIS[(Redis缓存)] + ES[(Elasticsearch)] + end + + subgraph 应用层 + API[业务API] + QUERY[查询服务] + end + + CRM --> ADAPTER1 + WEWORK --> ADAPTER2 + WECHAT --> ADAPTER3 + INTERNAL --> GATEWAY + + ADAPTER1 --> GATEWAY + ADAPTER2 --> GATEWAY + ADAPTER3 --> GATEWAY + + GATEWAY --> KAFKA + KAFKA --> CLEAN + CLEAN --> MAPPING + MAPPING --> MERGE + MERGE --> PG + + PG --> TAG + PG --> SEGMENT + TAG --> PG + SEGMENT --> PG + + PG --> REDIS + PG --> ES + + API --> QUERY + QUERY --> REDIS + QUERY --> PG + QUERY --> ES +``` + +### 核心流程说明 + +**六大核心流程：** +1. **数据采集**：多渠道数据统一接入 +2. **数据清洗**：格式标准化、数据验证 +3. **身份映射**：识别不同渠道的同一主体 +4. **全渠道整合**：合并为统一的实体 +5. **标签打标**：根据规则和行为打标签 +6. **客户圈人**：基于条件创建客户分群 + +--- + +## 核心实体设计 + +### 1. Account账户实体模型 + +```mermaid +erDiagram + Account { + varchar account_id PK + varchar account_name + varchar unified_credit_code UK + varchar account_status + varchar province + varchar city + datetime created_at + datetime updated_at + } + + AccountSummary { + varchar summary_id PK + varchar account_id FK + int total_contacts + int total_opportunities + decimal total_revenue + int health_score + datetime calculated_at + } + + AccountChannelIdentity { + varchar identity_id PK + varchar account_id FK + varchar channel_id FK + varchar channel_account_id + boolean is_verified + datetime first_seen_at + } + + Account ||--|| AccountSummary : 汇总 + Account ||--o{ AccountChannelIdentity : 多渠道身份 +``` + +**表关系说明：** +- Account（1） : AccountSummary（1）- 一对一汇总关系 +- Account（1） : AccountChannelIdentity（N）- 一个企业在多个渠道有多个身份 + +--- + +### 2. Contact联系人实体模型 + +```mermaid +erDiagram + Contact { + varchar contact_id PK + varchar contact_name + varchar mobile_phone UK + varchar email UK + varchar primary_account_id FK + varchar contact_status + datetime created_at + } + + ContactSummary { + varchar summary_id PK + varchar contact_id FK + int total_interactions + int engagement_score + datetime last_activity_at + datetime calculated_at + } + + ContactChannelIdentity { + varchar identity_id PK + varchar contact_id FK + varchar channel_id FK + varchar channel_user_id + boolean is_verified + datetime first_seen_at + } + + AccountContactRelation { + varchar relation_id PK + varchar account_id FK + varchar contact_id FK + varchar role_in_account + boolean is_primary + } + + Contact ||--|| ContactSummary : 汇总 + Contact ||--o{ ContactChannelIdentity : 多渠道身份 + Contact ||--o{ AccountContactRelation : 企业关系 +``` + +**表关系说明：** +- Contact（1） : ContactSummary（1）- 一对一汇总关系 +- Contact（1） : ContactChannelIdentity（N）- 一个联系人在多个渠道有多个身份 +- Contact（N） : Account（M）- 通过AccountContactRelation关联 + +--- + +### 3. Lead线索实体模型 + +```mermaid +erDiagram + Lead { + varchar lead_id PK + varchar lead_name + varchar company_name + varchar mobile_phone + varchar email + varchar channel_id FK + varchar lead_status + int lead_score + datetime created_at + datetime converted_at + varchar converted_contact_id FK + } + + LeadSummary { + varchar summary_id PK + varchar lead_id FK + int form_submissions + int days_in_pipeline + datetime last_activity_at + datetime calculated_at + } + + LeadChannelIdentity { + varchar identity_id PK + varchar lead_id FK + varchar channel_id FK + varchar channel_user_id + datetime captured_at + } + + Lead ||--|| LeadSummary : 汇总 + Lead ||--o{ LeadChannelIdentity : 多渠道身份 +``` + +**表关系说明：** +- Lead（1） : LeadSummary（1）- 一对一汇总关系 +- Lead（1） : LeadChannelIdentity（N）- 一个线索可能来自多个渠道 +- Lead在转化时通过converted_contact_id关联到Contact + +--- + +### 4. 标签系统实体模型 + +```mermaid +erDiagram + Tag { + varchar tag_id PK + varchar tag_name UK + varchar tag_category + varchar tag_type + varchar description + datetime created_at + } + + TagRelation { + varchar relation_id PK + varchar tag_id FK + varchar entity_type + varchar entity_id FK + datetime tagged_at + boolean is_auto + } + + TagRule { + varchar rule_id PK + varchar tag_id FK + varchar rule_type + text rule_definition + boolean is_active + datetime created_at + } + + Tag ||--o{ TagRelation : 应用于 + Tag ||--o{ TagRule : 规则 +``` + +**表关系说明：** +- Tag（1） : TagRelation（N）- 一个标签可以打给多个实体 +- Tag（1） : TagRule（N）- 一个标签可以有多个自动打标规则 +- TagRelation通过entity_type和entity_id关联到Account/Contact/Lead + +--- + +### 5. 分群系统实体模型 + +```mermaid +erDiagram + Segment { + varchar segment_id PK + varchar segment_name + varchar target_entity_type + text segment_rules + int member_count + boolean is_dynamic + datetime last_calculated_at + datetime created_at + } + + SegmentMember { + varchar member_id PK + varchar segment_id FK + varchar entity_type + varchar entity_id FK + datetime joined_at + boolean is_active + } + + Segment ||--o{ SegmentMember : 包含 +``` + +**表关系说明：** +- Segment（1） : SegmentMember（N）- 一个分群包含多个成员 +- SegmentMember通过entity_type和entity_id关联到Account/Contact/Lead + +--- + +### 6. 渠道实体模型 + +```mermaid +erDiagram + Channel { + varchar channel_id PK + varchar channel_name + varchar channel_type + varchar channel_status + datetime created_at + } + + Campaign { + varchar campaign_id PK + varchar campaign_name + varchar channel_id FK + date start_date + date end_date + datetime created_at + } + + Channel ||--o{ Campaign : 发起 +``` + +**表关系说明：** +- Channel（1） : Campaign（N）- 一个渠道可以发起多个营销活动 + +--- + +## 关键业务流程 + +### 1. 数据采集与清洗流程 + +```mermaid +sequenceDiagram + participant 数据源 as 数据源
CRM/企微/公众号 + participant 适配器 as 渠道适配器 + participant 网关 as API网关 + participant Kafka as Kafka队列 + participant 清洗 as 数据清洗服务 + participant 数据库 as PostgreSQL + + 数据源->>适配器: 1.推送原始数据 + Note over 适配器: 数据格式转换
统一字段映射 + + 适配器->>网关: 2.发送标准格式数据 + Note over 网关: 数据验证
签名校验
限流控制 + + 网关->>Kafka: 3.写入消息队列
快速响应 + 网关-->>数据源: 4.返回接收成功 + + Kafka->>清洗: 5.消费消息 + Note over 清洗: 数据清洗处理 + + 清洗->>清洗: 6.格式标准化 + Note over 清洗: - 手机号格式统一
- 企业名称清洗
- 地址标准化
- 必填字段验证 + + 清洗->>清洗: 7.数据去重 + Note over 清洗: - 同批次去重
- 24小时内去重 + + 清洗->>清洗: 8.数据验证 + Note over 清洗: - 手机号格式
- 邮箱格式
- 身份证格式
- 企业信用代码 + + alt 验证通过 + 清洗->>数据库: 9a.写入原始数据表
保留完整记录 + 清洗->>Kafka: 9b.发送到身份映射队列 + Note over 清洗: 继续下一步处理 + else 验证失败 + 清洗->>数据库: 9c.写入异常数据表
人工处理 + Note over 清洗: 记录失败原因 + end +``` + +**关键点：** +- 适配器层：各渠道数据统一格式 +- 网关层：快速响应，写入队列后立即返回 +- 清洗层：格式标准化、去重、验证 +- 异步处理：不阻塞数据采集主流程 + +--- + +### 2. 身份映射与全渠道整合流程 + +```mermaid +sequenceDiagram + participant Kafka as Kafka队列 + participant 映射 as 身份映射服务 + participant 数据库 as PostgreSQL + participant Redis as Redis缓存 + participant 合并 as 数据合并服务 + + Kafka->>映射: 1.消费清洗后数据 + + 映射->>映射: 2.提取身份标识符 + Note over 映射: - 手机号
- 邮箱
- 企业信用代码
- 企业名称 + + 映射->>Redis: 3.查询缓存
是否已有映射 + + alt 缓存命中 + Redis-->>映射: 4a.返回已有实体ID + 映射->>合并: 4b.直接进入合并流程 + else 缓存未命中 + Redis-->>映射: 4c.缓存MISS + + 映射->>数据库: 5.按优先级查询匹配 + Note over 数据库: Contact查询优先级：
1. 手机号精确匹配
2. 邮箱精确匹配
3. 企微UserID匹配 + + 数据库-->>映射: 6.返回查询结果 + + alt 找到唯一匹配 + 映射->>映射: 7a.确认为同一实体 + 映射->>Redis: 7b.写入映射缓存 + 映射->>合并: 7c.进入合并流程 + else 找到多个匹配 + 映射->>数据库: 7d.写入待审核队列 + 映射->>映射: 7e.触发人工审核流程 + Note over 映射: 疑似重复
需人工确认 + else 未找到匹配 + 映射->>数据库: 7f.创建新实体 + 映射->>数据库: 7g.创建ChannelIdentity + 映射->>Redis: 7h.写入映射缓存 + Note over 映射: 新客户首次进入 + end + end + + 合并->>数据库: 8.读取现有数据 + 合并->>合并: 9.数据合并策略 + Note over 合并: - 非空字段优先保留
- 最新数据优先
- 重要字段不覆盖 + + 合并->>数据库: 10.更新实体数据 + 合并->>数据库: 11.新增ChannelIdentity记录 + Note over 数据库: 记录本次渠道来源 + + 合并->>Redis: 12.更新实体缓存 + 合并->>Kafka: 13.发送更新事件
触发标签和汇总 +``` + +**关键点：** +- 缓存优先：减少数据库查询压力 +- 匹配优先级：手机号 > 邮箱 > 企微ID +- 疑似重复：人工审核避免误合并 +- 数据合并策略：保护重要字段不被覆盖 + +--- + +### 3. 标签打标流程 + +```mermaid +sequenceDiagram + participant Kafka as Kafka队列 + participant 标签引擎 as 标签引擎 + participant 规则库 as 规则库 + participant 数据库 as PostgreSQL + participant Redis as Redis缓存 + + Kafka->>标签引擎: 1.消费实体更新事件 + Note over Kafka: 触发条件：
- 实体创建
- 实体更新
- 定时批量 + + 标签引擎->>规则库: 2.加载标签规则 + 规则库-->>标签引擎: 3.返回活跃规则列表 + + loop 遍历每个标签规则 + 标签引擎->>标签引擎: 4.评估规则条件 + Note over 标签引擎: 规则类型：
- 属性规则
- 行为规则
- 统计规则
- 时间规则 + + alt 规则条件满足 + 标签引擎->>数据库: 5a.查询是否已有该标签 + + alt 标签不存在 + 标签引擎->>数据库: 6a.新增标签关系 + Note over 数据库: INSERT TagRelation + else 标签已存在 + 标签引擎->>标签引擎: 6b.跳过（幂等性） + end + else 规则条件不满足 + 标签引擎->>数据库: 5b.查询是否已有该标签 + + alt 标签存在 + 标签引擎->>数据库: 6c.删除标签关系 + Note over 数据库: DELETE TagRelation
标签自动移除 + end + end + end + + 标签引擎->>Redis: 7.更新标签缓存 + Note over Redis: 缓存实体的所有标签
用于快速查询 + + 标签引擎->>数据库: 8.记录打标日志 + Note over 数据库: 审计追溯 +``` + +**标签规则示例：** + +``` +属性规则： +- 行业 = "互联网" → 打标签"互联网行业" +- 年营收 > 1亿 → 打标签"大型企业" +- 省份 = "浙江" → 打标签"浙江客户" + +行为规则： +- 30天内访问次数 > 10 → 打标签"高活跃" +- 下载过白皮书 → 打标签"内容营销线索" +- 参加过线下活动 → 打标签"线下活动客户" + +统计规则： +- 商机总数 > 5 → 打标签"重点跟进客户" +- 成交金额 > 100万 → 打标签"高价值客户" +- 流失天数 > 180 → 打标签"流失预警" + +时间规则： +- 注册时间 < 7天 → 打标签"新客户" +- 最后活跃 < 30天 → 打标签"活跃客户" +``` + +--- + +### 4. 客户圈人（分群）流程 + +```mermaid +sequenceDiagram + participant 用户 as 营销人员 + participant 前端 as 分群界面 + participant 分群引擎 as 分群引擎 + participant 数据库 as PostgreSQL + participant Redis as Redis缓存 + participant Kafka as Kafka队列 + + 用户->>前端: 1.创建分群
设置筛选条件 + Note over 前端: 条件示例：
- 行业=互联网
- 省份=浙江
- 有标签"高活跃"
- 商机数>3 + + 前端->>分群引擎: 2.提交分群规则 + + 分群引擎->>分群引擎: 3.解析规则
生成SQL查询 + Note over 分群引擎: 规则解析器
转换为SQL WHERE条件 + + 分群引擎->>数据库: 4.执行分群查询 + Note over 数据库: 复杂查询示例：
SELECT account_id
FROM Account a
JOIN AccountSummary s
JOIN TagRelation t
WHERE 条件... + + 数据库-->>分群引擎: 5.返回符合条件的实体ID列表 + + 分群引擎->>分群引擎: 6.统计分群人数 + + alt 动态分群 + 分群引擎->>数据库: 7a.创建分群记录
is_dynamic=true + Note over 数据库: 不保存成员
每次实时计算 + else 静态分群 + 分群引擎->>数据库: 7b.创建分群记录
is_dynamic=false + + loop 批量插入成员 + 分群引擎->>数据库: 7c.插入SegmentMember
批量1000条/次 + end + + Note over 数据库: 保存成员快照
不随条件变化 + end + + 分群引擎->>Redis: 8.缓存分群结果 + Note over Redis: 缓存成员ID列表
提升查询性能 + + 分群引擎-->>前端: 9.返回分群结果 + 前端-->>用户: 10.展示分群统计 + Note over 用户: - 分群人数
- 分群规则
- 最后更新时间 + + opt 营销活动使用 + 用户->>前端: 11.选择分群
执行营销动作 + 前端->>Kafka: 12.发送营销任务 + Note over Kafka: - 群发邮件
- 推送消息
- 分配线索 + end +``` + +**动态分群 vs 静态分群：** + +| 对比项 | 动态分群 | 静态分群 | +|--------|---------|---------| +| 成员存储 | 不存储，实时计算 | 存储快照 | +| 成员变化 | 自动更新 | 固定不变 | +| 查询性能 | 较慢（每次计算） | 快速（直接查表） | +| 存储占用 | 小 | 大 | +| 适用场景 | 持续性营销 | 一次性活动 | +| 示例 | "活跃客户"群 | "双11参与客户"群 | + +--- + +### 5. 汇总计算流程 + +```mermaid +sequenceDiagram + participant Kafka as Kafka队列 + participant 汇总服务 as 汇总计算服务 + participant 分布式锁 as Redis分布式锁 + participant 数据库 as PostgreSQL + participant Redis as Redis缓存 + + Kafka->>汇总服务: 1.消费汇总触发事件 + Note over Kafka: 触发场景：
- 新增Contact
- 商机状态变化
- 定时全量刷新 + + 汇总服务->>分布式锁: 2.尝试获取锁
key: summary:ACC001 + + alt 获取锁失败 + 分布式锁-->>汇总服务: 3a.锁被占用 + 汇总服务->>汇总服务: 3b.跳过本次计算 + Note over 汇总服务: 防止重复计算 + else 获取锁成功 + 分布式锁-->>汇总服务: 3c.锁获取成功
超时30秒 + + par 并行查询统计数据 + 汇总服务->>数据库: 4a.统计Contact数 + 汇总服务->>数据库: 4b.统计Opportunity数据 + 汇总服务->>数据库: 4c.统计Lead数 + 汇总服务->>数据库: 4d.查询最后活跃时间 + end + + 数据库-->>汇总服务: 5.返回所有统计结果 + + 汇总服务->>汇总服务: 6.计算汇总指标 + Note over 汇总服务: 计算：
- total_contacts
- total_revenue
- win_rate
- health_score + + 汇总服务->>数据库: 7.更新Summary表
记录calculated_at + + 汇总服务->>Redis: 8.更新汇总缓存 + Note over Redis: 缓存TTL: 300秒 + + 汇总服务->>分布式锁: 9.释放锁 + end +``` + +**汇总计算优化策略：** + +``` +增量计算： +- 只计算变化部分 +- 减少全表扫描 +- 提升计算速度 + +并行查询： +- 统计数据并行执行 +- 减少总耗时 +- 提升吞吐量 + +分布式锁： +- 防止重复计算 +- 保证数据一致性 +- 避免资源浪费 + +定时 + 实时： +- 定时任务：每小时全量刷新 +- 实时触发：关键事件立即计算 +- 保证数据新鲜度 +``` + +--- + +## 技术实现方案 + +### 1. 数据库表结构设计 + +#### Account核心表 + +```sql +-- Account基础表 +CREATE TABLE account ( + account_id VARCHAR(64) PRIMARY KEY COMMENT '账户ID', + account_name VARCHAR(200) NOT NULL COMMENT '企业名称', + unified_social_credit_code VARCHAR(18) UNIQUE COMMENT '统一信用代码', + account_type VARCHAR(50) NOT NULL COMMENT '客户类型', + account_status VARCHAR(50) NOT NULL COMMENT '账户状态', + industry_id VARCHAR(64) COMMENT '行业ID', + province VARCHAR(50) COMMENT '省份', + city VARCHAR(50) COMMENT '城市', + account_source VARCHAR(100) COMMENT '来源', + owner_user_id VARCHAR(64) COMMENT '负责人', + created_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP, + updated_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP, + INDEX idx_name (account_name), + INDEX idx_status (account_status), + INDEX idx_city (province, city) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='企业账户表'; + +-- AccountSummary汇总表 +CREATE TABLE account_summary ( + summary_id VARCHAR(64) PRIMARY KEY COMMENT '汇总ID', + account_id VARCHAR(64) NOT NULL UNIQUE COMMENT '账户ID', + total_contacts INT DEFAULT 0 COMMENT '联系人总数', + total_opportunities INT DEFAULT 0 COMMENT '商机总数', + total_leads INT DEFAULT 0 COMMENT '线索总数', + total_revenue DECIMAL(18,2) DEFAULT 0 COMMENT '累计收入', + won_opportunities INT DEFAULT 0 COMMENT '赢单数', + lost_opportunities INT DEFAULT 0 COMMENT '输单数', + win_rate DECIMAL(5,2) DEFAULT 0 COMMENT '赢单率', + health_score INT DEFAULT 0 COMMENT '健康度评分', + last_activity_at DATETIME COMMENT '最后活跃时间', + calculated_at DATETIME NOT NULL COMMENT '计算时间', + updated_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP, + FOREIGN KEY (account_id) REFERENCES account(account_id), + INDEX idx_health_score (health_score DESC), + INDEX idx_revenue (total_revenue DESC) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='企业账户汇总表'; + +-- AccountChannelIdentity渠道身份映射表 +CREATE TABLE account_channel_identity ( + identity_id VARCHAR(64) PRIMARY KEY COMMENT '身份ID', + account_id VARCHAR(64) NOT NULL COMMENT '账户ID', + channel_id VARCHAR(64) NOT NULL COMMENT '渠道ID', + channel_account_id VARCHAR(200) NOT NULL COMMENT '渠道内账户ID', + identity_type VARCHAR(50) COMMENT '身份类型', + is_verified BOOLEAN DEFAULT FALSE COMMENT '是否已验证', + first_seen_at DATETIME NOT NULL COMMENT '首次发现时间', + last_seen_at DATETIME NOT NULL COMMENT '最后活跃时间', + UNIQUE KEY uk_account_channel (account_id, channel_id), + INDEX idx_channel_account (channel_id, channel_account_id), + FOREIGN KEY (account_id) REFERENCES account(account_id) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='企业渠道身份映射表'; +``` + +#### Contact核心表 + +```sql +-- Contact基础表 +CREATE TABLE contact ( + contact_id VARCHAR(64) PRIMARY KEY COMMENT '联系人ID', + contact_name VARCHAR(100) NOT NULL COMMENT '姓名', + mobile_phone VARCHAR(20) UNIQUE COMMENT '手机号', + email VARCHAR(200) UNIQUE COMMENT '邮箱', + wechat_id VARCHAR(100) COMMENT '微信ID', + job_title VARCHAR(100) COMMENT '职位', + contact_status VARCHAR(50) NOT NULL COMMENT '状态', + primary_account_id VARCHAR(64) COMMENT '主要企业ID', + contact_source VARCHAR(100) COMMENT '来源', + owner_user_id VARCHAR(64) COMMENT '负责人', + created_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP, + updated_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP, + INDEX idx_phone (mobile_phone), + INDEX idx_email (email), + INDEX idx_account (primary_account_id) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='联系人表'; + +-- ContactSummary汇总表 +CREATE TABLE contact_summary ( + summary_id VARCHAR(64) PRIMARY KEY COMMENT '汇总ID', + contact_id VARCHAR(64) NOT NULL UNIQUE COMMENT '联系人ID', + total_interactions INT DEFAULT 0 COMMENT '互动总数', + email_opens INT DEFAULT 0 COMMENT '邮件打开数', + email_clicks INT DEFAULT 0 COMMENT '邮件点击数', + engagement_score INT DEFAULT 0 COMMENT '参与度评分', + last_activity_at DATETIME COMMENT '最后活跃时间', + calculated_at DATETIME NOT NULL COMMENT '计算时间', + updated_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP, + FOREIGN KEY (contact_id) REFERENCES contact(contact_id), + INDEX idx_engagement (engagement_score DESC) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='联系人汇总表'; + +-- ContactChannelIdentity渠道身份映射表 +CREATE TABLE contact_channel_identity ( + identity_id VARCHAR(64) PRIMARY KEY COMMENT '身份ID', + contact_id VARCHAR(64) NOT NULL COMMENT '联系人ID', + channel_id VARCHAR(64) NOT NULL COMMENT '渠道ID', + channel_user_id VARCHAR(200) NOT NULL COMMENT '渠道用户ID', + identity_type VARCHAR(50) COMMENT '身份类型', + is_verified BOOLEAN DEFAULT FALSE COMMENT '是否已验证', + first_seen_at DATETIME NOT NULL COMMENT '首次发现时间', + last_seen_at DATETIME NOT NULL COMMENT '最后活跃时间', + UNIQUE KEY uk_contact_channel (contact_id, channel_id), + INDEX idx_channel_user (channel_id, channel_user_id), + FOREIGN KEY (contact_id) REFERENCES contact(contact_id) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='联系人渠道身份映射表'; +``` + +#### 标签系统表 + +```sql +-- Tag标签表 +CREATE TABLE tag ( + tag_id VARCHAR(64) PRIMARY KEY COMMENT '标签ID', + tag_name VARCHAR(100) NOT NULL UNIQUE COMMENT '标签名称', + tag_category VARCHAR(50) COMMENT '标签分类', + tag_type VARCHAR(50) COMMENT '标签类型', + description TEXT COMMENT '描述', + created_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP, + INDEX idx_category (tag_category) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='标签表'; + +-- TagRelation标签关系表 +CREATE TABLE tag_relation ( + relation_id VARCHAR(64) PRIMARY KEY COMMENT '关系ID', + tag_id VARCHAR(64) NOT NULL COMMENT '标签ID', + entity_type VARCHAR(50) NOT NULL COMMENT '实体类型', + entity_id VARCHAR(64) NOT NULL COMMENT '实体ID', + tagged_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP COMMENT '打标时间', + is_auto BOOLEAN DEFAULT FALSE COMMENT '是否自动打标', + UNIQUE KEY uk_tag_entity (tag_id, entity_type, entity_id), + INDEX idx_entity (entity_type, entity_id), + FOREIGN KEY (tag_id) REFERENCES tag(tag_id) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='标签关系表'; + +-- TagRule标签规则表 +CREATE TABLE tag_rule ( + rule_id VARCHAR(64) PRIMARY KEY COMMENT '规则ID', + tag_id VARCHAR(64) NOT NULL COMMENT '标签ID', + rule_type VARCHAR(50) NOT NULL COMMENT '规则类型', + rule_definition TEXT NOT NULL COMMENT '规则定义', + is_active BOOLEAN DEFAULT TRUE COMMENT '是否启用', + created_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP, + FOREIGN KEY (tag_id) REFERENCES tag(tag_id) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='标签规则表'; +``` + +#### 分群系统表 + +```sql +-- Segment分群表 +CREATE TABLE segment ( + segment_id VARCHAR(64) PRIMARY KEY COMMENT '分群ID', + segment_name VARCHAR(200) NOT NULL COMMENT '分群名称', + target_entity_type VARCHAR(50) NOT NULL COMMENT '目标实体类型', + segment_rules TEXT NOT NULL COMMENT '分群规则', + member_count INT DEFAULT 0 COMMENT '成员数量', + is_dynamic BOOLEAN DEFAULT FALSE COMMENT '是否动态分群', + last_calculated_at DATETIME COMMENT '最后计算时间', + created_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP, + INDEX idx_entity_type (target_entity_type) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='分群表'; + +-- SegmentMember分群成员表（仅静态分群使用） +CREATE TABLE segment_member ( + member_id VARCHAR(64) PRIMARY KEY COMMENT '成员ID', + segment_id VARCHAR(64) NOT NULL COMMENT '分群ID', + entity_type VARCHAR(50) NOT NULL COMMENT '实体类型', + entity_id VARCHAR(64) NOT NULL COMMENT '实体ID', + joined_at DATETIME NOT NULL DEFAULT CURRENT_TIMESTAMP COMMENT '加入时间', + is_active BOOLEAN DEFAULT TRUE COMMENT '是否活跃', + UNIQUE KEY uk_segment_entity (segment_id, entity_type, entity_id), + INDEX idx_entity (entity_type, entity_id), + FOREIGN KEY (segment_id) REFERENCES segment(segment_id) +) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT='分群成员表'; +``` + +--- + +### 2. Kafka Topic设计 + +```yaml +Topic设计: + +1. data-ingestion (数据采集) + Partition: 8 + Replication: 3 + Retention: 7天 + 用途: 接收多渠道原始数据 + +2. data-cleaning (数据清洗) + Partition: 8 + Replication: 3 + Retention: 3天 + 用途: 清洗后的标准数据 + +3. identity-mapping (身份映射) + Partition: 8 + Replication: 3 + Retention: 3天 + 用途: 身份识别和映射任务 + +4. entity-merge (实体合并) + Partition: 8 + Replication: 3 + Retention: 3天 + 用途: 实体数据合并任务 + +5. tag-calculation (标签计算) + Partition: 8 + Replication: 3 + Retention: 3天 + 用途: 标签打标任务 + +6. summary-calculation (汇总计算) + Partition: 8 + Replication: 3 + Retention: 3天 + 用途: 汇总统计任务 +``` + +--- + +### 3. 缓存策略设计 + +```yaml +Redis缓存设计: + +1. 实体基础数据缓存 + Key: entity:{type}:{id} + TTL: 300秒 + 示例: entity:account:ACC001 + 用途: 缓存Account/Contact基础信息 + +2. 汇总数据缓存 + Key: summary:{type}:{id} + TTL: 300秒 + 示例: summary:account:ACC001 + 用途: 缓存Summary汇总数据 + +3. 标签缓存 + Key: tags:{type}:{id} + TTL: 600秒 + 示例: tags:account:ACC001 + Value: ["高价值客户", "互联网行业", "浙江"] + 用途: 缓存实体的所有标签 + +4. 分群成员缓存 + Key: segment:members:{segment_id} + TTL: 1800秒 + Value: Set类型，存储entity_id + 用途: 缓存分群成员列表 + +5. 身份映射缓存 + Key: identity:{channel}:{channel_user_id} + TTL: 3600秒 + 示例: identity:wechat:openid_xxx + Value: contact_id + 用途: 快速查询渠道身份对应的实体ID + +6. 分布式锁 + Key: lock:summary:{entity_id} + TTL: 30秒 + 用途: 防止汇总计算重复执行 +``` + +--- + +### 4. 核心服务接口设计 + +```yaml +数据采集服务: + POST /api/ingestion/crm + - 接收CRM数据 + POST /api/ingestion/wework + - 接收企业微信数据 + POST /api/ingestion/wechat + - 接收公众号数据 + +查询服务: + GET /api/account/{id} + - 查询Account详情 + GET /api/account/{id}/360 + - 查询Account 360度视图 + GET /api/contact/{id} + - 查询Contact详情 + GET /api/contact/search + - 搜索Contact + +标签服务: + GET /api/tag/list + - 获取标签列表 + POST /api/tag/create + - 创建标签 + POST /api/tag/apply + - 手动打标签 + GET /api/tag/entity/{type}/{id} + - 查询实体的所有标签 + +分群服务: + POST /api/segment/create + - 创建分群 + GET /api/segment/{id}/members + - 获取分群成员 + POST /api/segment/calculate + - 重新计算分群 + GET /api/segment/list + - 获取分群列表 +``` + +--- + +## 总结 + +### 核心能力 + +**1. 多渠道数据采集** +- 统一接入层，支持CRM、企微、公众号等多渠道 +- 快速响应，异步处理 +- 数据清洗和标准化 + +**2. 全渠道身份整合** +- 基于手机号、邮箱、企业信用代码等多维度匹配 +- ChannelIdentity记录每个渠道身份 +- 疑似重复人工审核 + +**3. 智能标签系统** +- 支持自动打标和手动打标 +- 规则引擎驱动 +- 标签分类管理 + +**4. 灵活分群能力** +- 动态分群：实时计算 +- 静态分群：快照固化 +- 支持复杂条件组合 + +**5. 实时汇总统计** +- 异步计算，不阻塞主流程 +- 分布式锁防重 +- 缓存优化查询性能 + +### 技术特点 + +``` +简洁清晰： +- 专注核心业务流程 +- 去除复杂行为数据 +- 架构易于理解 + +高性能： +- 消息队列异步解耦 +- 多级缓存策略 +- 数据库索引优化 + +高可用： +- PostgreSQL主从复制 +- Redis Cluster +- Kafka集群部署 + +可扩展： +- 微服务架构 +- 水平扩展能力 +- 插件化设计 +``` + +### 适用场景 + +- 企业客户数：100万-1000万 +- 联系人数：500万-5000万 +- 线索数：200万-2000万/年 +- 数据采集QPS：1000-10000 +- 查询QPS：5000-50000 From 3f0e95cdfcab75fc7b53b067aa02a8c113c90371 Mon Sep 17 00:00:00 2001 From: Cursor Agent Date: Wed, 5 Nov 2025 17:38:23 +0000 Subject: [PATCH 10/10] Add Flink architecture documentation Co-authored-by: 13585811473 <13585811473@163.com> --- docs/flink-architecture.md | 214 +++++++++++++++++++++++++++++++++++++ 1 file changed, 214 insertions(+) create mode 100644 docs/flink-architecture.md diff --git a/docs/flink-architecture.md b/docs/flink-architecture.md new file mode 100644 index 000000000..1477355a8 --- /dev/null +++ b/docs/flink-architecture.md @@ -0,0 +1,214 @@ +# Flink 系统架构图 + +## Flink 完整系统架构 + +```mermaid +graph TB + subgraph 应用层 + APP[Flink Application
DataStream API / Table API / SQL] + end + + subgraph 运行时层JobManager + JM[JobManager] + DISPATCHER[Dispatcher
接收作业提交] + RM[ResourceManager
资源管理] + JOBMASTER[JobMaster
作业协调] + CHECKPOINT[CheckpointCoordinator
检查点协调器] + end + + subgraph 运行时层TaskManager + TM1[TaskManager 1] + TM2[TaskManager 2] + TM3[TaskManager N] + + SLOT1[Task Slot 1
执行线程] + SLOT2[Task Slot 2
执行线程] + + NETWORK[Network Buffers
数据交换缓冲区] + MEMORY[Managed Memory
托管内存] + end + + subgraph 状态管理层 + STATE[State Backend] + MEMORY_STATE[Memory] + FS_STATE[FileSystem] + ROCKS_STATE[RocksDB] + end + + subgraph 部署层 + STANDALONE[Standalone
独立部署] + YARN[YARN
资源调度] + K8S[Kubernetes
容器编排] + MESOS[Mesos
资源调度] + end + + APP -->|提交作业| DISPATCHER + DISPATCHER --> JOBMASTER + JOBMASTER --> RM + JOBMASTER --> CHECKPOINT + + RM -->|申请资源| YARN + RM -->|申请资源| K8S + + RM -->|分配Slot| TM1 + RM -->|分配Slot| TM2 + RM -->|分配Slot| TM3 + + JOBMASTER -->|部署Task| SLOT1 + JOBMASTER -->|部署Task| SLOT2 + + SLOT1 --> NETWORK + SLOT2 --> NETWORK + + SLOT1 -.->|读写状态| STATE + SLOT2 -.->|读写状态| STATE + + STATE --> MEMORY_STATE + STATE --> FS_STATE + STATE --> ROCKS_STATE + + CHECKPOINT -.->|触发检查点| TM1 + CHECKPOINT -.->|触发检查点| TM2 +``` + +## Flink 核心组件说明 + +### 1. JobManager（主节点） +- **Dispatcher**: 接收客户端提交的作业 +- **ResourceManager**: 管理TaskManager资源和Slot分配 +- **JobMaster**: 每个作业一个，负责作业执行协调 +- **CheckpointCoordinator**: 协调分布式快照 + +### 2. TaskManager（工作节点） +- **Task Slot**: 任务执行槽位，隔离CPU和内存 +- **Network Buffers**: 任务间数据交换缓冲区 +- **Managed Memory**: 排序、哈希表等操作的托管内存 + +### 3. State Backend（状态后端） +- **MemoryStateBackend**: 内存存储（开发测试） +- **FsStateBackend**: 文件系统存储（HDFS/S3） +- **RocksDBStateBackend**: RocksDB存储（大状态场景） + +### 4. 部署模式 +- **Standalone**: 独立集群部署 +- **YARN**: Hadoop资源调度 +- **Kubernetes**: 容器化部署 +- **Mesos**: 资源调度框架 + +## Flink 数据流处理架构 + +```mermaid +graph TB + subgraph 数据源层 + KAFKA[Kafka] + DB[Database CDC] + FILE[File System] + end + + subgraph Flink处理层 + SOURCE[Source Operator
数据接入] + + MAP[Map Operator
数据转换] + FILTER[Filter Operator
数据过滤] + FLATMAP[FlatMap Operator
数据展开] + + KEYBY[KeyBy
数据分组] + + WINDOW[Window Operator
窗口聚合] + + PROCESS[Process Function
自定义处理] + + SINK[Sink Operator
结果输出] + end + + subgraph 状态管理 + KEYED_STATE[Keyed State
键控状态] + OPERATOR_STATE[Operator State
算子状态] + end + + subgraph 输出层 + KAFKA_OUT[Kafka] + DB_OUT[Database] + REDIS_OUT[Redis] + ES_OUT[Elasticsearch] + end + + KAFKA --> SOURCE + DB --> SOURCE + FILE --> SOURCE + + SOURCE --> MAP + MAP --> FILTER + FILTER --> FLATMAP + + FLATMAP --> KEYBY + KEYBY --> WINDOW + WINDOW --> PROCESS + + PROCESS --> SINK + + KEYBY -.->|读写| KEYED_STATE + WINDOW -.->|读写| KEYED_STATE + PROCESS -.->|读写| KEYED_STATE + + SOURCE -.->|读写| OPERATOR_STATE + SINK -.->|读写| OPERATOR_STATE + + SINK --> KAFKA_OUT + SINK --> DB_OUT + SINK --> REDIS_OUT + SINK --> ES_OUT +``` + +## Flink Checkpoint 机制 + +```mermaid +sequenceDiagram + participant JM as JobManager + participant TM1 as TaskManager 1 + participant TM2 as TaskManager 2 + participant STATE as State Backend + participant STORAGE as 持久化存储 + + JM->>JM: 1.触发Checkpoint + Note over JM: 定时触发
间隔如10秒 + + JM->>TM1: 2.发送Barrier
checkpoint-123 + JM->>TM2: 2.发送Barrier
checkpoint-123 + + TM1->>TM1: 3.对齐Barrier + Note over TM1: 等待所有输入流
Barrier到达 + + TM1->>STATE: 4.快照状态 + Note over STATE: 保存算子状态 + + STATE->>STORAGE: 5.持久化 + Note over STORAGE: HDFS/S3 + + TM1->>JM: 6.确认完成 + + TM2->>TM2: 3.对齐Barrier + TM2->>STATE: 4.快照状态 + STATE->>STORAGE: 5.持久化 + TM2->>JM: 6.确认完成 + + JM->>JM: 7.Checkpoint完成 + Note over JM: 所有TaskManager
都确认完成 +``` + +## Checkpoint 机制说明 + +### Barrier对齐 +- Source算子接收到Checkpoint触发信号后插入Barrier +- Barrier随数据流向下游传播 +- 算子等待所有输入流的Barrier都到达后才执行快照 + +### 状态快照 +- 保存Keyed State（键控状态） +- 保存Operator State（算子状态） +- 异步写入持久化存储 + +### 故障恢复 +- 从最近成功的Checkpoint恢复 +- 重放Checkpoint之后的数据 +- 保证Exactly-Once语义