The short answer
Use shorter sprints with smaller done criteria; expect preemption; write handoffs that survive a page; schedule recovery after incidents instead of stacking heroics—focus debt is real.
How this differs from meeting-heavy guides
Meetings are often predictable fences. Pages are stochastic: they arrive without agenda. Calendar tactics help meetings; queue discipline and smaller WIP help on-call.
Contract with reality
If your rotation expects sub-thirty-minute response, do not schedule ninety-minute compile-heavy blocks without coverage. Negotiate coverage or scope down—otherwise you are planning fiction.
Micro-blocks and preemption
Choose tasks that checkpoint cleanly: docs, small fixes, test stabilization, triage. Avoid “start a large merge” micro-sprints unless you can pause safely mid-rebase—some cannot.
Recovery between pages
Adrenaline and incident context switching are expensive. After a serious page, take the notes, then allow a short reset before pretending normal coding returns instantly—see context switching recovery.
Audio defaults under stress
Under fatigue, lyrics often hurt. Prefer instrumental or silence during incident follow-ups; keep one stream policy—see one audio stream.
Handoff quality
Good on-call is good note-taking: timeline, blast radius, commands run, graphs links, and explicit “still unknown.” Future-you is the primary reader.
Practical takeaway
On-call sprint shapes are smaller, more interruptible, and more honest about recovery. That is not weakness—it is scheduling that matches the job.
Frequently asked questions
Is this the same as the on-call research article?
Related. The research page explains focus debt and incidents. This guide is operational: how to shape sprints around paging reality.