diff --git a/modules/ROOT/pages/import.adoc b/modules/ROOT/pages/import.adoc
index f13dc0fe6..1f8ca87fe 100644
--- a/modules/ROOT/pages/import.adoc
+++ b/modules/ROOT/pages/import.adoc
@@ -1462,8 +1462,9 @@ Now use the previously defined ID spaces when connecting the actors to movies.
 [[import-tool-multiple-ids]]
 == Using multiple node IDs
 
-A node header can also contain multiple `ID` columns, where the relationship data references the composite value of all those columns.
-This also implies using `string` as `id-type`.
+A node header can contain multiple `ID` columns.
+The relationship data must then use a matching number of `START_ID` / `END_ID` columns as references to the composite value of those ID columns.
+This implies using `string` as `id-type`.
 
 For each `ID` column, you can specify to store its values as different node properties.
 However, the composite value cannot be stored as a node property.
@@ -1498,13 +1499,13 @@ Now use both IDs when defining the relationship:
 .relationships_header.csv
 [source, csv]
 ----
-:START_ID,:TYPE,:END_ID
+:START_ID,:START_ID,:TYPE,:END_ID,:END_ID
 ----
 
 .relationships.csv
 [source, csv]
 ----
-aa11,WORKS_WITH,bb22
+aa,11,WORKS_WITH,bb,22
 ----
 ====
 
@@ -1532,13 +1533,13 @@ Now use the defined ID space when connecting John with Paul, and use both IDs in
 .relationships_header.csv
 [source, csv]
 ----
-:START_ID(MyGroup),:TYPE,:END_ID(MyGroup)
+:START_ID(MyGroup),:START_ID(MyGroup),:TYPE,:END_ID(MyGroup),:END_ID(MyGroup)
 ----
 
 .relationships.csv
 [source, csv]
 ----
-aa11,WORKS_WITH,bb22
+aa,11,WORKS_WITH,bb,22
 ----
 ====