Buyer guide · deletion and rights requests
How privacy software handles deletion requests
The handoff matters: a request may be matched, sent to connected systems, held for a response or left for staff to complete.
OneTrust: California DROP matching
DROP requests are API-pulled, matched, converted to DSARs or marked not found; broker systems need matching-format hashed records or indexes.
Scope: this is the California DROP flow in the vendor's documentation; it does not establish a general completion rate or compatibility with your brokers.
DataGrail: deletion paths and exceptions
At Pending Delete, selected APIs start deletion; some integrations anonymize PII instead of deleting the whole record. Errors retry the next day and can block request progress. A skipped integration is not run again for the rest of that request, and skipping cannot be undone. Direct-contact processors have 14 days to respond, then settings determine whether the integration skips or blocks. Direct-contact and whole-record deletion integrations do not support review before deletion.
The documentation describes workflow states and integration types. Account settings, integration prevalence, completion, latency and accuracy have not been independently measured.
Terms used here
- Data-subject request (DSR)
A data-subject request (DSR) is a vendor workflow record for a person's privacy request; these examples focus on deletion.
OneTrust DROP documentation ↗- Whole-record deletion
DataGrail says whole-record deletion APIs cannot extract data first during a deletion request.
DataGrail's integration types documentation ↗- Anonymization
DataGrail describes some integrations as anonymizing PII rather than deleting the whole record.
DataGrail deletion documentation ↗- Skipped integration
DataGrail says a skipped integration does not run again in the current request, and the organization remains responsible for that work.
DataGrail deletion documentation ↗
These definitions explain terms in vendor pages; they do not replace legal or contract definitions.
Questions for a demo
- Which of our systems use whole-record deletion, and can staff review data before deletion?
- When a processor does not respond, which setting skips or blocks the request?
- How are skipped integrations surfaced, assigned and completed outside the request?
- For broker matching, what exact hash format and index must we maintain?
- What request records remain after fulfillment, and who deletes them?