Practice note
Turn postmortem findings into follow-up work
A way to write action items with a clear owner and a checkable outcome.
Start from a finding¶ ↑
Pick a concrete moment in the timeline when a different signal, decision, or control could have changed the outcome. State the gap in terms of what responders could see or do at the time. For example, a rollback procedure might have existed, while the responders lacked a way to tell whether it had reached every region.
Define what completion looks like¶ ↑
Write an action with a named owner, a target date, and a verification step. For the rollback example, the verification step could be a staging exercise in which the responder confirms the deployed version in every region from a single dashboard.
Group related improvements when they share the same outcome. Small changes with a clear verification step can be scheduled and reviewed alongside broader work.
Bring the result back to the postmortem¶ ↑
Link the follow-up ticket from the postmortem and record the outcome when it lands. If the team defers an action, capture the decision and the condition that would bring it back into scope. Future responders can then see which improvements were actually in place.