Set up on-call schedules and escalation policies

On-call schedules decide who is on call; escalation policies decide who is paged, and when, while an incident stays unacknowledged. Both are part of the Team plan, together with PagerDuty integration.

Step 1: make sure people can be reached

Escalation pages people through their personal destinations. On the On-call page, the box How on-call people are reached warns you if you have none. Choose Notify me personally at your email address, or add other personal channels (SMS, Telegram, push apps) under Integrations by ticking Personal destination — only notify me when I am on call. Each schedule card shows whether every participant has at least one verified personal destination.

Step 2: create a rotation

Choose New schedule and fill in:

• Name, for example "Primary on-call".
• Time zone: handoffs happen at local time in this zone and stay correct across daylight-saving changes.
• Rotation starts on and Handoff time, for example Monday at 09:00.
• Shift length (days), for example 7 for a weekly rotation.
• Participants, in rotation order: only enabled owners, admins and operators can be on call. The first person takes the first shift.

Each schedule shows a calendar and a shift list for the next 28 days. Use Download calendar (ICS) to add the rotation to your calendar app.

Step 3: build an escalation policy

Choose New policy. Each step has a Delay (minutes), the schedules whose current on-call person is paged, and optional shared destinations to notify as well. Steps run in order. For example:

1. Step 1, delay 0: page whoever is on call in "Primary on-call".
2. Step 2, delay 10: page "Secondary on-call" and post to Slack #incidents.
3. Step 3, delay 20: notify the engineering manager's destination.

Set Repeat the whole policy from "Do not repeat" up to 5 more times. Tick Default policy for monitors without their own policy to apply it workspace-wide.

When escalation stops

Escalation stops as soon as someone acknowledges the incident or the monitor recovers. The incident page shows the running policy and the next step under Escalation.

Test it

Choose Simulate on a policy to see who would be paged and when, without sending anything.

PagerDuty

If your team already uses PagerDuty, add a PagerDuty destination in Integrations with your Events API v2 integration key, on the US or EU endpoint. Incidents are triggered and resolved in PagerDuty and de-duplicated.