The TrenchesIncident handoff desk

Hand over the incident while it’s still moving.

Give the next responder enough evidence to act and enough context to spot when the situation has changed.

Open the incident desk

Start with the current impact and when you checked it

Name who is affected, what they can’t do, and when you last saw it. A recovery graph and a user report can both be true at the same time. Give each observation its timestamp and source so the incoming responder can check the difference.

Keep a short, sourced timeline

Record what someone observed, what someone changed, and what remains a theory. Keep contradictions visible until someone has checked them. If a team reversed a change, give the action its own entry and describe the result you observed separately.

Example. At 22:17 UTC the dispatch queue returned to baseline. At 22:19 UTC support reproduced stale status in an existing rider session. Check whether a returning rider still sees the stale update after refreshing. Record the route and timestamps together.

Give the next action an owner and a stop condition

Name the responder, the next check, the result they should see if it works, and what they should do otherwise. Keep the brief short enough to read aloud. Send the handoff through your team’s approved channel and get explicit acknowledgement from the incoming responder.

Transfer a snapshot

The Trenches can copy a text brief or download a handoff file. The recipient can open that file in the desk, read its snapshot, and copy an acknowledgement to send back. Send only incident information that your team permits in that channel.

Further reading from Google SRE. Managing Incidents and Incident Response.