The most detailed free FDE + DevOps library: 140+ lessons, 70+ labs and 80 long-form articles, in English and Turkish. Start learning →

DevOps Professional · Module 7: Incident leadership and handover

Ops Ownership and Sustainable Rotation

On-call is a design problem, not a character test. Ownership covers decisions plus pages plus follow-through; rotations respect sleep, load and backup; compensation is explicit; and health is reviewed with the people who carry the pager.

10 min reading

Objectives

  • Explain what on-call ownership covers: decisions, pages, follow-through
  • Design rotations that humans survive: length, load, backup, compensation
  • Compensate on-call explicitly: time, money, rest, no heroics
  • Review rotation health with the holders, not about them

Why this matters

A team of four carries a noisy pager with one-week rotations, no backup and no compensation beyond thanks. Within a year two engineers leave citing burnout, and the survivors stop answering quickly, which management reads as attitude. The failure was roster design: load without relief, heroics without payback, health never reviewed. On-call done this way selects against the careful and keeps the exhausted. The design decides retention before any single incident does.

Concepts

Ownership has three parts. Decision rights during incidents (the on-call decides mitigation, no escalation for permission). The pages themselves (routed by severity with the M15 budget discipline: pages for user pain, tickets otherwise). Follow-through (postmortem actions owned, not orphaned when the rotation turns). Partial ownership (pages without authority, or authority without follow-through time) is how good engineers burn out while the system stays fragile.

Rotation design respects biology. Length fits the load (a week of quiet differs from a week of fire); backup exists and is real (secondary who actually takes overflow, not a name on a slide); handoff includes the open issues with context, not just the pager. Load is measured (pages per rotation, night interruptions) and capped; sustained overload hires or de-Scopes before it injures. Rest after rough rotations is scheduled, not requested as a favor.

Compensation is explicit and discussed.Time back, pay differentials, rest days, promotion credit for operational excellence: the mix varies, the explicitness does not. Uncompensated on-call taxes the conscientious most and teaches that operations is where careers stall. Review health with holders quarterly: load numbers, sleep impact, process gripes, with changes afterward. A review without changes is surveillance; with changes, it is stewardship.

Worked example

A fictional rotation redesign starts from measured load (night pages per month, MTTR per severity) and lands: shorter rotations in noisy quarters, a real secondary, post-incident rest days, and explicit compensation. The before-after numbers are quoted; holder feedback after one cycle adjusts the design again. The roster is a living document with owners, not a spreadsheet with victims.

Common wrong move

Treating on-call as an initiation rite the juniors must survive alone. It concentrates load on the least experienced, hides systemic noise behind individual endurance, and teaches newcomers that suffering is skill. Design for humans, rotate the load, compensate openly.

Quick check

An optional 4-question self-check. Answers never leave your device, are not stored, and never count toward any assessment.

Lesson feedback

No published feedback yet.

Log in and complete the lesson to leave feedback.

Exercise

Redesign a fictional rotation from measured load numbers: length, backup, rest and compensation explicit, with a quarterly review holding the changes.

Pass criteria

The record shows the load numbers, the redesigned roster with all four elements, and the review mechanism with holder feedback quoted.

Sources

Log in to track progressFree account: stores only your lesson progress and quiz results.