Anonymize customer data in Loomi BigQuery

When you fulfill a data removal request (GDPR/CCPA), use anonymization rather than deletion. Event deletions in Marketing aren't mirrored to Loomi BigQuery, but anonymization is. Learn how deleted events behave in Loomi BigQuery.

Deletion vs. anonymization

Since you can't rely only on mirroring data deletion from app to Loomi BigQuery, in case you're fulfilling data removal request (GDPR/CCPA), you'll use anonymization functions in the app. Anonymization removes personal data also from Loomi BigQuery, see Anonymizing customers.

Anonymizing customers

When you anonymize a customer in Marketing, the data is also anonymized in BigQuery for that customer's current profile ID, including historical IDs from any prior profile merges.

Customer attributes marked as private are anonymized once a day. All customers and their attributes are replaced during a daily update from Marketing. If customer attributes are already anonymized, that anonymized (empty) value will be copied to Loomi BigQuery customer tables as well.

Event properties marked as private are anonymized during processing event increments, at least once a day, including events tracked under historical IDs from merged profiles (see Merging customers).

Technical steps of events anonymization:

1. During anonymization through the app or API, an anonymization event is tracked for the customer.
2. Anonymization event is loaded into the BigQuery during a periodic data-update along with all the other events.
3. All the new anonymization events are queried.
4. Anonymization is done by an Update query for all the anonymization events queried in step 3 which contained a customer_ID and any historical IDs linked to it via customers_id_history. This means all private event properties tracked under any of those IDs will be set to NULL.
5. ltds_anonymization_timestamp is tracked for all the anonymized events with a timestamp of the moment when the anonymization went through in the BigQuery.