Skip to main content
Level 2
Nov 27, 2024 10:32 AM UTC
Solved

Non-stitched profile data set records - Customer Journey Analytics

Hi there,

 

I have an AEP profile dataset with some million CRM records being connected to CJA. Most of profiles do not have events tied to the CRM ID, so we cannot view them in Analysis Workspace (as Adobe FAQ explains). However, the client needs to analyze all the customer records in CJA no matter they have had an event.

 

In order to solve this, we generated one "fake event" per customer and the solution worked. The problem of this solution is that we need to ingest some million fake records monthly to analyze CRM data with a (monthly) time window, which is not recommendable from both licensing and data cleansing perspective.

 

What would be the best fit approach to circumvent this limitation? 

 

Thanks in advance,

FarazHusain
FarazHusainAccepted solution
Adobe Employee
Nov 27, 2024 11:11 AM UTC

Hey @jlasso71 Events basically corresponds to hits in CJA. CJA is primarily built to work on event datasets, so if you only have a profile dataset, I would also recommend ingesting fake events such that these records are visible in CJA reports. 

 

As far as I know, CJA is currently not designed to work solely on profile data sets. Maybe this could be raised as a feature request, but at the moment, you have to prepare the event data sets against the CRM ID; that's the only way, as per my knowledge.

5 replies

FarazHusain
Adobe Employee
FarazHusainAdobe EmployeeAccepted solution
Adobe Employee
Nov 27, 2024 11:11 AM UTC

Hey @jlasso71 Events basically corresponds to hits in CJA. CJA is primarily built to work on event datasets, so if you only have a profile dataset, I would also recommend ingesting fake events such that these records are visible in CJA reports. 

 

As far as I know, CJA is currently not designed to work solely on profile data sets. Maybe this could be raised as a feature request, but at the moment, you have to prepare the event data sets against the CRM ID; that's the only way, as per my knowledge.

Translate
Feb 17, 2025 9:52 AM UTC

Hi @jlasso71, That's very interesting. Could you share how you did that? We want to do the same but haven't found information on how to proceed.

Thank you!

Translate
jlasso71Author
Level 2
Feb 18, 2025 5:18 PM UTC

Hi @mmarti ,

 

We use a daily scheduled query in Query Service to extract unique customers (client_id) from a dataset and insert them into another dataset which is connected to CJA.

 

It was something similar to the next query:

 

INSERT INTO table_final
SELECT DISTINCT client_id,
	  now() as timestamp,
	  'CJA_event' as eventType,
	  event_id as _id
FROM table_origin
WHERE clientType = 'cliente'

 Hope it helps,

Jose

Translate
May 23, 2025 7:17 AM UTC

Hi Jose,

 

Thank you for the info provided, it's really helpful. When it comes to define event_id, do you have any criteria or it's just a random value?

 

Regards!

Translate