Data deletion in AEP

[DebrajRo3]

Hi, We are planning to delete users/profiles from the AEP sandbox that have not been active for a year to remain under the licensing quota. There seem to be two ways we can delete them: 1. Datahygiene tool and 2. Delete APIs.

Can we also use QS PSQL to delete profiles too?

Also, I have a few questions.

1. Does the API have a daily limit on data deletion? How is it different from other purge activities like data hygiene/PSQL?
2. How does the delete happen? Does it only delete profile datasets, or does it also delete related records from the engagement datasets?

What are your recommendations on setting up such a purge process for profiles/contact history in AEP?

Thanks

Debraj

[supratim320]

Hi @DebrajRo3 ,

There are two ways of data deletion:

1. Data hygiene: deletion of data based on "business driven decision" (e.g.- ecid causing incorrect profile stitching, your use case etc)
2. Data privacy: data deletion but recommended when individual user asks to delete his own data (consider GDPR)

Now back to your question:

• Q: Can we also use QS PSQL to delete profiles too?
  • A: Let the system handle it, it needs to delete the id from multiple places (unified profile store, identity, events etc) - so recommended to use "data hygiene"
• Q: Does the API have a daily limit on data deletion? How is it different from other purge activities like data hygiene/PSQL?
  • A: I do not see such limitation in the document (resource) - P.S. I have not used this service extensively - so I will pass other to comment on it
• Q: How does the delete happen? Does it only delete profile datasets, or does it also delete related records from the engagement datasets?
  • A: As I mentioned above, the deletion happens across all the storage places (UPS, events, related datasets/ datalake, identity service etc)
• As for your use case, I would recommend you to use "data hygiene" API (resource) - so you make a query, get the profiles' identifier, then use the API to create the work order. You will find some guidelines here

Let me know if this helps!

[ganeshPy1]

Hello , 

The Data Hygiene tool is likely the most effective method for your scenario, as it provides a robust, systematic way to delete profiles and associated data across AEP components, aligning with your goal of managing licensing quotas.

Deleting via the Profile API only removes data from the Profile Store, not the Data Lake or Identity Graph, which could leave residual data.

QS with PSQL is not a direct method for deleting profiles but can be used as a preparatory step to identify inactive profiles. Combine it with the Data Hygiene tool or Delete APIs for actual deletion.

Thanks

[Asheesh_Pandey]

@DebrajRo3  In your case, it's a good idea to set a TTL (time-to-live) for pseudonymous profiles. This helps ensure profiles are automatically cleaned up over time and keeps you within your license limits.

Check out Advanced Data Lifecycle Management in platfrom for more details. Also, take a look at the options shared by @supratim320  earlier in this thread - they’re relevant and helpful.

[Supriya_Savre]

Hi @DebrajRo3,

Did the shared solution help you out? If yes, kindly consider marking the answer as ‘correct’.

If you’re still facing any challenges, please feel free to continue the conversation here. Or if you've discovered a solution yourself, we would appreciate it if you could share it with the community.