Experience Event expirations
In Adobe Experience Platform, you can configure expiration times for all Experience Events that are ingested into a dataset enabled for Real-Time Customer Profile. This lets you automatically remove data from the Profile store that is no longer valid or useful for your use cases.
Experience Event expirations cannot be configured through the Experience Platform UI or APIs. Instead, you must contact support in order to enable Experience Event expirations on your required datasets.
IMPORTANTAutomated expiration process
After Experience Event expirations have been enabled on a Profile-enabled dataset, Experience Platform automatically applies the expiration values for each captured event in a two-step process:
- All new data that is ingested into the dataset has the expiration value applied at ingestion time based on the event timestamp.
- All existing data in the dataset has the expiration value retroactively applied as a one-time backfill system job. Once the expiration value has been placed on the dataset, events that are older than the expiration value will be immediately dropped as soon as the system job runs. All other events will be dropped off as soon as they reach their expiration values from the event timestamp. When all Experience Events have been removed, if the profile no longer has any profile attributes, the profile will no longer exist.
WARNINGFor example, if you applied an expiration value of 30 days on May 15th, the following steps would occur:
- All new events get an expiration value of 30 days applied as they are ingested.
- All existing events that have a timestamp older than April 15th are immediately deleted with the system job.
- All existing events that have a timestamp that is newer than April 15th have an expiration value 30 days after their event timestamp. So, if an event has a timestamp of April 18th, it is deleted 30 days after that timestamp’s date, which would be May 18th.
Effects on segmentation
You must ensure that the lookback windows for your audiences are within the expiration boundaries of their dependent datasets in order to keep results accurate. For example, if you apply an expiration value of 30 days and have an audience that tries to view data from up to 45 days ago, the resulting audience will likely be inaccurate.
You should therefore keep the same Experience Event expiration value for all datasets, if possible, to avoid the impact of different expiration values across different datasets in your segmentation logic.
Frequently asked questions
The following section lists frequently asked questions regarding Experience Event data expiration:
How does Experience Event data expiry differ from Pseudonymous Profile data expiry?
Experience Event data expiry and Pseudonymous Profile data expiry are complementary features.
Granularity
Experience Event data expiration works on a dataset level. As a result, each dataset can have a different data expiry setting.
Pseudonymous Profile data expiration works on a sandbox level. As a result, the data expiration will affect all profiles in the sandbox.
Identity types
Experience Event data expiration removes events only based on the event record’s timestamp. The identity namespaces included are ignored for expiration purposes.
Pseudonymous Profile data expiration only considers profiles that have identity graphs which contain identity namespaces that were selected by the customer, such as ECID, AAID, or other types of cookies. If the profile contains any additional identity namespace that was not in the customer’s selected list, the profile will not be deleted.