Template · SaaS incident response

SaaS Incident Response SOP Template

A standard operating procedure for responding to production incidents in a SaaS product — from the first alert to the customer-facing post-mortem and the follow-up fixes.

6 steps · 60m
SaaS incident response
SaaS Incident Response SOP
When production breaks, the right person has authority to act within 5 minutes, customers hear from us before they file a ticket, and we leave the incident with a written timeline and follow-up tasks.

Procedure

  1. Detect: alert fans out to the on-call

    Alert fires from the monitoring system, fans out to the on-call primary, secondary and a Slack channel named #inc-<date>-<short-id>. The on-call declares an incident within 5 minutes.

  2. Declare severity + convening

    Severity uses the standard matrix (S0 customer-impact + revenue, S1 degraded service, S2 internal-only). S0/S1 convenes an Incident Commander, Comms Lead and Tech Lead in a dedicated Zoom.

  3. Mitigate first, investigate second

    The goal of the first hour is to stop the customer pain: rollback, toggle a feature flag, shed load, or block the offending deploy. The root cause comes later — restoring service wins.

  4. Communicate every 30 minutes (external)

    Status page, in-app banner and the customer account managers all update on a 30-minute cadence. Comms Lead owns the wording; engineers own the facts.

  5. Resolve + write the customer-facing RCA

    When mitigation holds, the Incident Commander declares resolved. The Comms Lead posts a public post-mortem within 24 hours: timeline, customer impact, short-term and long-term fixes.

  6. Blameless post-mortem + action items

    Within 5 working days, hold a blameless review with everyone on the channel. Document timeline, decisions, contributing factors and 3 action items — each with an owner and a date.

Quality check

A reviewer signs off only when every item below is true.

  • A severity is declared within 5 minutes of the alert
  • Customer-facing comms went out within 15 minutes for S0/S1
  • Status updates went out every 30 minutes while the incident was open
  • A public post-mortem is on the status page within 24 hours
  • Every action item has a named owner and a target date
From template to your SOP library

Upload your existing manual or policy document to customize this template instantly

Drop a TXT or PDF and Procify AI rewrites the procedure in your voice, with your tool-stack and your internal steps — ready to share on Slack or export to PDF.