Hi all,
I have two datasets in AEP (not enabled for profile). Both datasets have same identity values (CRM ID) but they are being passed in separate schema fields (but both datsets use same identity namespace)
Dataset 1 :- CRM ID is sent in the schema field :- _tenant.crmid
Dataset 2 :- CRM ID is sent in the identityMap :- identityMap.id has the CRM ID value
While using both datasets in CJA connection, it looks like the stitching is happening.
But I am unable to create a derived field to combine both identity fields into a single "CRM ID" derived field dimension. When trying to create if else condition, I cannot use identityMap.id
Is there any solution to get a single dimension instead of using _tenant.crmid and identityMap.id as separate dimensions?
Solved! Go to Solution.
Topics help categorize Community content and increase your ability to discover relevant content.
Views
Replies
Total Likes
You can’t do this with a Derived Field and that’s expected behavior.
In CJA, identityMap.* fields are not exposed as normal dimensions, so they cannot be referenced in Derived Fields or CASE/IF logic. They are used only for identity stitching, not for reporting transformations.
That’s why:
Stitching works
But identityMap.id is unavailable in Derived Fields
Option 1 (Best / Recommended): Normalize upstream
Populate the CRM ID into a regular XDM field (e.g. _tenant.crmid) in both datasets
Keep the same value in identityMap for stitching
→ CJA can then use one clean CRM ID dimension everywhere
Option 2 (Acceptable workaround): Use only one reporting field
Pick _tenant.crmid as the reporting dimension
Let identityMap exist purely for stitching
→ This is how most CJA implementations are designed
Option 3 (Not recommended): Dataset duplication
Copy identityMap CRM ID into a custom field via data prep
→ Adds complexity and maintenance overhead
Views
Replies
Total Likes
You can’t do this with a Derived Field and that’s expected behavior.
In CJA, identityMap.* fields are not exposed as normal dimensions, so they cannot be referenced in Derived Fields or CASE/IF logic. They are used only for identity stitching, not for reporting transformations.
That’s why:
Stitching works
But identityMap.id is unavailable in Derived Fields
Option 1 (Best / Recommended): Normalize upstream
Populate the CRM ID into a regular XDM field (e.g. _tenant.crmid) in both datasets
Keep the same value in identityMap for stitching
→ CJA can then use one clean CRM ID dimension everywhere
Option 2 (Acceptable workaround): Use only one reporting field
Pick _tenant.crmid as the reporting dimension
Let identityMap exist purely for stitching
→ This is how most CJA implementations are designed
Option 3 (Not recommended): Dataset duplication
Copy identityMap CRM ID into a custom field via data prep
→ Adds complexity and maintenance overhead
Views
Replies
Total Likes
Hi @SamuelPaulPeter , Thank you so much. This is very helpful!