Guide · Observability
How to Set Up PagerDuty for SaaS
PagerDuty for SaaS: incident alerts, on-call rotation, escalation. Set up in minutes for small teams.
Paging and escalation policies operationalize Incident Readiness—alerts must reach a human. Primary control: Incident Readiness
Slack at 2am is not on-call
Alerts that only land in a channel die when nobody is watching. One team saw the database die at 2am and read the Slack ping at 9am—seven hours dark. PagerDuty (or similar) pages a phone, escalates if ignored, and rotates who carries the pager.
Minimum wiring that actually rings
Create a service (e.g. Production). Connect UptimeRobot, CloudWatch, or Datadog via the native integration or webhook. Define escalation: primary → backup after N minutes. Add people and a schedule (even a weekly manual rotation counts). Trigger a test page and confirm the phone buzzes before you trust it.
Incidents need a path, not just a tool
APRF Incident Readiness and Observability expect detection to reach a human. Pair PagerDuty with an uptime check and a one-page response plan so the page isn't "now what?"
Next: Incident Readiness
Open the related pillar specification for mandatory checks, artifacts, and pass conditions. Self-attest is optional.
Related
Frequently asked questions
- How do I set up PagerDuty for my SaaS?
- Create a PagerDuty account and service. Connect your monitoring tool (UptimeRobot, CloudWatch, Datadog) via integration. Set up escalation policy and on-call schedule. Test.
- What monitoring tools work with PagerDuty?
- UptimeRobot, Pingdom, CloudWatch, Datadog, New Relic, Sentry, and many others. Most have a PagerDuty integration or webhook.
- Do I need PagerDuty for a small team?
- Yes. Even a solo developer benefits from PagerDuty—alerts go to your phone, not just Slack. Escalation is less critical for one person, but you still get reliable alerts.