Missing records between systems: recover webhook and API failures
The same notification can arrive more than once or arrive late. A record of each event's progress helps staff decide whether to retry it, recover a missing record or investigate.

The workflow illustrates a possible design and is not a client case study.
- Record event
- Process and retry safely
- Reconcile regularly
After a timeout, check whether the record was saved
A transfer can fail even when both systems are running. A timeout is particularly awkward: the receiver may have saved the record but failed to reply, so sending it again can create a duplicate. Record receipt, processing and the confirmed result separately. That gives the recovery process enough information to retry without simply guessing, or leaving a missing record unresolved.
Check retry and recovery features
Check the connector’s logs and retry or replay functions. Stripe’s webhook documentation describes repeated notifications and events arriving out of order. Each provider has its own behaviour, so the integration must not assume that every event arrives once and in sequence.
Record each event’s progress
Choose one important process. Verify incoming events, save their IDs and queue the work. Show failed items with the reason, and check whether the action has already taken effect before retrying. Give staff a recovery procedure for cases that need intervention, then compare records with the source system regularly to find omissions.
Does replay repeat the charge?
Replay an event and check that it cannot charge or ship twice. Then try out-of-order notifications, rate limits, outages and timeouts, including work that was only partly completed. Confirm that missing records can be found and recovered, and that alerts make it clear which failures need someone to act.
Bring a failed event’s history
Bring event timelines and failure logs with personal details removed, the API documentation and the result expected for each case.