How the problem develops
An account coordinator fixes a service address, then an integration restores an old value. Each connector treats its own copy as authoritative. Updating a field triggers another update, and the team cannot tell which change was intentional.
What to check first
- Inspect the field history and list every workflow allowed to write the disputed value.
- Check whether empty values, stale timestamps or retries are treated as genuine corrections.
A better process
- Assign an authoritative source and a permitted direction for each shared field.
- Record the origin and time of updates, and distinguish a deletion from a missing value.
- Send conflicting edits to a review queue rather than letting the last write win silently.
What a system can do
Copy approved changes in the allowed direction and alert the owner when systems disagree.
Where people stay in control
Bulk overwrites and deletions should need explicit review and a recovery plan.
Benefits and tradeoffs
The useful outcome
Corrections can remain stable and staff can understand how a value arrived.
The tradeoff
Field ownership can reveal incompatible business definitions. Those disagreements need an operational decision before integration rules.
Test before going live
Test an old event arriving late, an empty field and a simultaneous manual correction. Confirm that no stale update silently replaces the approved value.