The On-Call Rotation Schedule Builder turns a list of engineers into a fair, predictable on-call schedule plus the supporting policy a real rotation needs. Ad-hoc on-call assignment breeds resentment and gaps; a clean round-robin with a documented escalation path and severity definitions keeps incidents covered and the team sane.
How it works
The builder assigns shifts by round-robin: starting from your chosen date, each shift of N days goes to the next person in the list, wrapping back to the first when the list is exhausted. Because every member appears in the same order at the same interval, the load is distributed equally and the schedule is predictable weeks ahead. Alongside the table, the tool emits an escalation path (primary → secondary → lead), a handover checklist for the start of each shift, incident severity definitions (SEV1–SEV3), and a contact list.
What the generated document includes
The output is not just a calendar table — it covers the four things a responder needs in the first minutes of an incident:
- The shift schedule — who is primary and secondary for each shift window, with start and end dates
- Escalation path — ordered contact chain so alerts never go unanswered: primary on-call → secondary on-call → team lead or engineering manager
- Handover checklist — the steps the outgoing on-call should complete before the shift ends: confirm alert routing is set up, brief the incoming engineer on any open incidents, share relevant runbook links
- Severity definitions — a shared vocabulary so everyone agrees what warrants a 3am page versus what can wait for business hours
Incident severity definitions (typical)
| Level | Description | Response expectation |
|---|---|---|
| SEV1 | Complete outage, data loss, security breach | Wake-up page; all-hands immediately |
| SEV2 | Major degradation, significant user impact | Page primary + secondary; respond within 15 min |
| SEV3 | Minor degradation, workaround available | Notify during business hours; no after-hours page |
| SEV4 | Cosmetic issues, low-impact bugs | Log and schedule for normal sprint work |
Edit these to match your team’s actual SLAs — a startup with a single product may escalate differently than an enterprise team with tiered support contracts.
Worked example
A 4-person team — Alice, Bob, Carol, Dave — on 7-day weekly shifts starting Monday:
| Week | Primary | Secondary |
|---|---|---|
| Week 1 (Mon Aug 4) | Alice | Bob |
| Week 2 (Mon Aug 11) | Bob | Carol |
| Week 3 (Mon Aug 18) | Carol | Dave |
| Week 4 (Mon Aug 25) | Dave | Alice |
| Week 5 (Mon Sep 1) | Alice | Bob |
Each engineer is primary once every four weeks. With a 7-day shift, every person also functions as secondary for the week immediately following their primary — providing natural continuity since the secondary already knows that week’s service context.
Tips for a healthy on-call culture
- Publish the schedule at least 4 weeks ahead. Engineers need to plan travel, appointments, and personal commitments around on-call weeks.
- Always have a secondary. A single primary is one missed page from an unhandled incident. The escalation path exists for exactly that case.
- Tune severities to reduce alert fatigue. If everything pages as SEV1, nothing feels urgent and people start ignoring pages — the most dangerous on-call anti-pattern.
- Review the handover checklist each shift. A well-run handover reduces the probability that the incoming engineer inherits a mystery they have to solve cold at 2am.