When a business launches a marketing campaign to interest customers in some new or existing product(s), they will typically use a set of business rules to select leads for their campaign. Machine learning can be used to help increase the response rate from these leads. This solution packet shows how to use a prediction model to increase the response rate to a campaign by recommending how to contact (for example, e-mail, SMS, or cold call) as well as when to contact (day of week and time of day) each lead identified for use in a new campaign.
This template uses simulated data from the insurance industry to model the campaign response for an acquisition campaign. The model uses predictors like demographic details of the lead, historical campaign performance and product details. The model predicts the probability of a lead making a purchase from each channel, at various times of the day and days of the week, for every lead in the database. The final recommendation for targeting each user is decided based on the combination of channel, day of week and time of day with the highest probability of making a purchase.
For customers who prefer an on-premise solution, the implementation with SQL Server R Services is a great option, which takes advantage of the power of SQL Server and the innnovation of the R language. We then use PowerBI to visualize the deployed results.
A high level description of this solution is described here.
This template showcases the use of Random Forest and Gradient Boosting to model for Campaign Responses. Data Scientists can follow the steps of data creation, model development, scoring and deployment. The final deployed recommendations will then be visualized in PowerBI. See more details and explanations of all the files involved in this solution.. Also make sure to use the Typical Workflow below to get a better sense of how these scripts fit into the overall solution.
If you’re interested in creating this solution yourself, use this fully automated solution that uploads the data to your SQL Server, preprocesses data, performs feature engineering, trains and scores the models by executing a PowerShell script. This is the fastest way to deploy the entire solution.
-
Visit the Cortana Intelligence Gallery to deploy this solution on an Azure VM.
-
Or use the PowerShell Instructions to deploy to your own machine. There are a number of steps to perform here that have already been done for you if you use the above Cortana Intelligence Gallery deployment.
We have modeled the steps in the template after a real-life data science process, where the data preparation, model training and evaluation can be done by a data scientist, from the convenience of their R IDE, and the deployment is done using SQL stored procedures with embedded R code.
Follow along with this typical workflow and view the details and follow the solution from start to finish:
-
If you are using the solution deployed from the Cortana Intelligence Gallery, click here.
-
If you are deploying this solution on your own machine, click here
##Template Contents
View the contents of this solution template
NOTE: Please don't use "Download ZIP" to get this repository, as it will change the line endings in the data files. Use "git clone" to get a local copy of this repository.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.