Two teams blame each other for an integration failure. How do we resolve it?
Blame replacing diagnosisCreate a shared factual timeline and test the interface together. Separate restoration from root-cause accountability so a blame dispute does not prolong the failure.
What to do now
- Agree the observed failure
- Collect logs and interface evidence
- Run a joint reproducible test
- Assign corrective actions from the findings
When to escalate
Escalate when
The dispute prevents restoration or exceeds team-level authority.
Who decides
The issue owner or delegated decision authority
Example
Application and infrastructure teams reproduce the same timeout using a shared test and identify the failing boundary.
Mistake to avoid
Selecting a responsible team before examining evidence.
Prevent a repeat
Track unresolved decision age and verify the outcome before treating an issue as closed.
Suggested ApprovedFlow setup
- Tasks
- Documents
- Approvals
Record the issue and next action on an owned task, attach evidence, and route the unresolved decision to the proper authority. See the platform
Related cases
- When should I escalate a project issue?Unclear escalation boundary
- An urgent issue has no clear owner. What do we do now?Unowned urgent issue
- Two teams each wait for the other to finish. How do we break the loop?Circular dependency
- Our integration works in isolation but fails end to end. Why?Local completion without system verification