Use this when an incident is big enough that one person shouldn’t both debug and run the response. The incident commander guide is the longer version of the same role.
Open
- Say, in one channel, that this is an incident.
- State the severity you actually use, and the user impact you know about. Say what you don’t know.
- Name the incident commander, the person on operations, and the person writing the timeline. Say if a role is empty.
- Put the next check-in time in the channel.
While it’s open
- Keep decisions and conclusions in that channel. Paste back anything useful that happened elsewhere.
- Prefer one mitigation. Say which alternatives you are not doing.
- Prefer a mitigation you can reverse.
- Don’t announce a cause you can’t support. Describe what you see and what you’re changing.
- Send status on the cadence you promised: what changed, what users still can’t do, when the next update is.
Hand off
- Current user impact, and when it started.
- What is in flight, and who is doing it.
- Leading idea, and what you’ve ruled out.
- The next decision, and when you’ll make it if nothing changes.
- Who holds each role after the handoff. The new commander can restate it.
Close
- User impact has stopped, or you’ve explicitly accepted what’s left. Say which.
- Write the end time in the timeline.
- Name who will draft the postmortem, and when the review is. The postmortem template is the blank doc.