Why trait population is higher than the onboarded data




Issue Description:

The number of records which are onboarded from inbound file is less than the onboarded trait population.

Sample Use case:

Below is newly created onboarded traits and count as at 11th  April.

SID               Name                                                                      7 Day Uniques

6852718       30D Transition Small Medium Subscribed              21

6852719       30D Transition Small Medium Not Subscribed       20

6852720       30D Transition Micro Soho Subscribed                  357

6852733       30D Transition Micro Soho Not Subscribed             616

However, the counts of the traits in the onboarded files are as below, (file name - ftp_dpm_67646_1491865209.sync.1.gz)

AC_list_Tran_SM_Sub_30D="Y"     13

AC_list_Tran_SM_NSub_30D="Y"     19

AC_list_Tran_MISO_Sub_30D="Y"     274

AC_list_Tran_MISO_NSub_30D="Y"     681

How can AAM trait counts be higher than what’s on the file?


In the onboarded file we have customer IDs and corresponding this CID AAM maintains a UUID. For some records there were more than 1 UUID associations and hence we are seeing the mismatch in the number when we are comparing data from file to UI.

The process can be defined as:

  1. The customer onboarded data keyed off partner's CRM ID.
  2. At the data ingestion time, the inbound job did a lookup for the partner's CRM ID and converted each of them to 0 to many AAM ids
  3. Each individual CRM ID is mapped to a large number of AAM devices
  4. The trait population count is done in the AAM ID key space, not in the CRM Id space and hence the numbers are mismatching.

If you still have any question, send an email to amsupport@adobe.com and we will perform further investigation.