Major Incident Manager Checklist: First 15 Minutes of a P1
What a major incident manager does in the first 15 minutes of a P1, minute by minute, from confirming impact to the first stakeholder update.
The first 15 minutes of a P1 set the tone for the next three hours. After 15 years of running P1 bridges, the pattern I see most is not a slow fix — it is a slow start: nobody declared it, nobody owned it, and the first stakeholder update went out late.
Here is the checklist I use. Print it, adapt it, and keep it next to the on-call runbook.
Minutes 0–3: Confirm and declare
- Confirm the impact. Is a customer or business service actually affected, or is it one noisy alert? Check one independent signal: a second monitor, a support ticket, a quick test.
- Classify. Customer pain that needs more than on-call to fix is a P1. Customer pain on-call can handle is a P2. See P1 vs P2 classification.
- Declare it. Create the incident record and say it out loud in the channel: "This is a P1. I'm the incident manager."
- Note the time. The declaration time starts the comms clock.
When in doubt, declare. Downgrading a P1 costs a few minutes. Discovering an hour later that it was a P1 costs much more.
Minutes 3–6: Open the bridge and get the right people
- Open one bridge. One call, one incident channel. Post the link in the channel and the incident record.
- Page the owning team's on-call. Not everyone — the team that owns the failing service.
- Pull in dependencies only when there's a reason. Network, database, cloud provider, supplier.
- Start the timeline. Every key event with a time: alert, declaration, bridge opened, who joined.
Minutes 6–9: Assign roles
- Incident manager (you): runs the bridge, makes decisions, keeps time. You don't debug.
- Technical lead: directs the investigation.
- Comms owner: writes and sends stakeholder updates on time.
- Scribe: keeps the timeline (on a small bridge, the comms owner can do this).
Say the roles out loud. Unspoken roles are how two people send different updates and nobody records the timeline.
Minutes 9–13: Frame the problem
- What is broken, for whom, since when? One sentence everyone on the bridge agrees with.
- What changed? Deployments, configuration, certificates, suppliers, traffic. Recent change is the first lead to check.
- Is there a quick mitigation? Rollback, failover, feature flag, traffic shift. Restoring service comes before finding the root cause.
- Is there a workaround users can follow now? It goes in the first update.
Minutes 13–15: Send the first update
- Use the structured format: Status, Impact, What we're doing, Workaround, Next update.
- Give the next update as an absolute time: "Next update by 14:30 UTC."
- Send it to everyone at once. One update, every stakeholder list.
- Send it on time even if you know little. "Investigating" with a clear impact and next update time is a good first update.
The P1 stakeholder update template has worked examples for every stage.
What to avoid in the first 15 minutes
- Debugging yourself instead of running the bridge
- Letting several side conversations run on the bridge
- Inviting everyone "just in case"
- Guessing a cause or a fix time in the first update
- Waiting for certainty before declaring
After minute 15
Set a rhythm: a short bridge check-in before each update deadline, a decision on mitigation at each check-in, and a new update on time. Keep the timeline going — it becomes the backbone of the post-incident review.
The checklist on one page
- 0–3: Confirm impact, classify, declare, note the time
- 3–6: One bridge, page the owning on-call, start the timeline
- 6–9: Name the incident manager, technical lead, comms owner, scribe
- 9–13: What's broken, what changed, quick mitigation, workaround
- 13–15: First structured update, absolute next-update time, to everyone
How SLAShield helps
In SLAShield you can declare a P1 from Slack, and the incident record, timeline and comms countdown start straight away. with NOVA on, the first five-field update is sent automatically (or the comms owner sends it by hand without NOVA), and SLAShield warns the comms owner before each deadline.
See it in practice
Want to see a P1 update go out against a comms deadline? Try the instant demo — no sign-up needed. Related reading: P1 stakeholder update template, Incident communication SLAs, Running a P1 bridge across languages.