On-Call Scope by Service
Specialised vs generalist.
Overview
On-call scope by service decides whether on-callers cover specific services they own deeply or general infrastructure across many services. Service-scoped rotations produce fast response and deep expertise but require enough engineers per service to staff the rotation. Generalist rotations cover more ground with fewer engineers but trade depth for breadth. The right answer depends on team size and service criticality, and most mature organizations end up with a hybrid: specialised rotations for tier-1 services, generalist with escalation for the rest.
- Specialised vs generalist. Per-team scope choice; depends on engineer count, service count, and criticality distribution.
- Per-service rotation. Dedicated rotation for tier-1 services; deep expertise; needs 6-plus engineers per service to staff sustainably.
- Generalist rotation. Broad rotation across many services; works for small teams; trades depth for staffing feasibility.
- Per-service escalation plus quarterly scope review. Per-incident escalation to service specialists when generalists hit their depth limit; quarterly review catches scope drift.
The approach
The practical approach is per-service rotations for tier-1 services where engineer count supports it, generalist rotation for tier-2 and below with explicit specialist escalation paths, quarterly scope review against actual incident patterns (which scopes are absorbing the right load, which are mismatched), and documented scope policy committed to the team handbook so the model survives team changes.
- Per-service rotation. Tier-1 services get dedicated rotations; the depth pays for itself in MTTR on critical pages.
- Generalist with escalation. Tier-2 and below run generalist; specialist escalation kicks in when depth is needed.
- Per-service escalation. Per-incident specialist escalation; the generalist on-call reaches the right depth without paging the wrong people.
- Per-quarter scope review plus documented policy. Quarterly scope review against incident patterns; per-team scope policy committed for onboarding.
Why this compounds
Scope discipline compounds across services and years. Each correctly-scoped rotation produces fast response on the incidents that match its scope; each escalation path teaches the team where its depth limits actually are; the rotation shape evolves to match the service shape rather than the org chart shape.
- Incident response. Right scope produces fast response; MTTR drops because the on-call has the depth required for the incident class.
- Team expertise. Per-service rotation grows depth; engineers know the systems they own beyond the documented behaviour.
- Operational culture. Right scope preserves teams; engineers are not paged for systems they cannot reasonably debug.
- Institutional knowledge. Each scope decision teaches operational patterns; the team learns where depth pays for itself.
Scope discipline is an organizational discipline that pays off across years. Nova AI Ops integrates with on-call telemetry, surfaces scope patterns, and supports the team’s rotation discipline.