5 DevOps Checklists That Save Small Teams From 3am Disasters

When production goes down at 3am, you don't need theory. You need a checklist.

After years of incident response, these are the 5 checklists I keep next to my keyboard:

1. The First 30 Minutes Checklist

• [ ] Confirm: Is this real? (Check monitoring, not just alerts)
• [ ] Say out loud: "I think we have an incident"
• [ ] Assign severity: SEV1 (business stopped), SEV2 (degraded), SEV3 (minor)
• [ ] Page the right people (not everyone)
• [ ] Freeze all deployments
• [ ] Announce next update time (15 min from now)

2. Communication Checklist

• [ ] Create a dedicated incident channel
• [ ] Post initial message: what's broken, what we know, what we're doing
• [ ] Update every 15 minutes (even if "still investigating")
• [ ] Notify stakeholders (not just engineering)
• [ ] Document timeline as you go

3. Recovery Checklist

• [ ] Identify root cause (not symptom)
• [ ] Test fix in staging first
• [ ] Deploy fix
• [ ] Verify recovery (monitoring green, not just "no errors")
• [ ] Monitor for 30 minutes before declaring resolved

4. Post-Incident Checklist

• [ ] Schedule blameless post-mortem within 48 hours
• [ ] Write timeline of events
• [ ] Identify contributing factors (not just root cause)
• [ ] Create action items with owners and deadlines
• [ ] Share learnings with the team

5. Prevention Checklist

• [ ] Add monitoring for the failure mode
• [ ] Add alerting thresholds
• [ ] Update runbook with new procedure
• [ ] Schedule disaster recovery drill
• [ ] Review dependent systems

These checklists are extracted from the Ops Field Cards - 12 printable incident checklists for small teams.

Free resource: The First 30 Minutes - free incident quick-start checklist.

Full bundle: Hive80 Ops Mega Bundle - all 5 kits, one download.

Use code LAUNCH50 for 50% off anything in the store.

Don't wait for the 3am call to wish you had checklists. Get them now.