Skip to main content
TimBo4
TimBo4
Level 2
June 26, 2025
Solved

AEP segments and Adobe Analytics filters

I am building batch audiences in AEP and the counts I am generating do not come close to the counts in Adobe Analytics that use the exact same evars/analytics. Why?
This post is no longer active and is closed to new replies. Need help? Start a new post to ask your question.
KevinFosterDXAccepted solution
Adobe Employee
June 28, 2025

My experience has been that at some point you need to sample.

 

Take any random ten profiles from the system where you see more than the other.

 

Then look to see what subset of the ten is in the other system, and for those that are not, investigate why they didn't make it into the same list. Usually once I under just two of those then I'm able to figure out what's going wrong.

View in thread

4 replies

Adobe Employee
ankitk1Adobe Employee
Adobe Employee
June 26, 2025

Hi Tim, 

your adobe analytics count may have multiple ECID, however in AEP you get the count of profiles which may be combination of ECID and multiple other factors hence you may not see the count is matching. Hope it helps. 

Translate
    TimBo4
    TimBo4Author
    Level 2
    June 26, 2025

    That makes sense if the counts were reasonably close but in my case, there's a 180% difference between these two counts.  

    Translate
    KumarRishii
    KumarRishii
    Level 5
    June 26, 2025

    @timbo4 Discrepancies between AEP batch audience counts and Adobe Analytics data can occur due to differences in data sources, audience evaluation logic, and identity resolution. AEP processes profile-level data with periodic updates, while Analytics operates at the hit/session level and offers near-real-time reporting.

    Translate
      Adobe Employee
      KevinFosterDXAdobe EmployeeAccepted solution
      Adobe Employee
      June 28, 2025

      My experience has been that at some point you need to sample.

       

      Take any random ten profiles from the system where you see more than the other.

       

      Then look to see what subset of the ten is in the other system, and for those that are not, investigate why they didn't make it into the same list. Usually once I under just two of those then I'm able to figure out what's going wrong.

      Translate
        Terms & ConditionsAccessibility statement

        Loading footer...