Incident Severity Levels and Escalation
You'll learn: How to classify incident severity and route urgent issues to the right people at the right speed.
The Core Idea
Not all incidents are created equal. A typo on a settings page is very different from a complete payment system outage. Severity levels (commonly P0–P4 or SEV-0–SEV-4) classify how much business impact an incident has, which determines how urgently you respond and whom you wake up at 3 AM.
Standard Severity Tiers
- P0 (Critical): Complete service outage or major revenue loss. Page everyone immediately; all hands on deck.
- P1 (High): Significant degradation affecting many users. Page primary on-call; escalate to secondary if no response in 5–10 minutes.
- P2 (Medium): Partial feature failure or isolated user impact. Notify on-call during business hours; can wait until morning if after-hours.
- P3 (Low): Minor bugs with workarounds. Create ticket; handle during normal sprint work.
- P4 (Trivial): Cosmetic issues, documentation fixes. Backlog only.
Escalation Mechanics
When to escalate: If the incident isn't resolved within the SLO response time for that severity, or if the on-call engineer needs expertise they don't have.
Who to page: Start with the primary on-call for the affected service. If they don't acknowledge in 5 minutes (P0/P1), automatically escalate to the secondary. For cross-team dependencies, loop in relevant service owners immediately.
Business impact drives urgency: A P0 for a payments service during Black Friday demands instant all-hands response. The same bug on a Sunday at 2 AM for an internal tool might be P2 — handle next business day.
Key Takeaway: Match incident severity to business impact and response urgency—clear severity tiers prevent under-reacting to critical issues and over-reacting to minor bugs, while escalation paths ensure the right experts engage when front-line responders need help.