On-Call → Attrition Link

Bad on-call drives departures.

Overview

Bad on-call experience drives engineer departures, often months after the actual incident. The link is hard to see in any single resignation but obvious in the aggregate: teams with sustained high page volume, off-hours wakes, and missing comp time lose senior engineers faster than peers. Treating on-call quality as a retention metric (with explicit tracking, exit interviews, and quarterly review) makes the link visible early enough to act on.

• Bad on-call drives departures. Per-engineer on-call experience predicts retention; the link is invisible in single resignations and obvious in the aggregate.
• Per-engineer exit interview. Per-departure the on-call factor explicitly probed; "did on-call contribute to your decision" gets honest answers if the question is asked.
• Per-quarter retention review. Per-quarter the on-call retention link reviewed against attrition data; pattern surfaces early enough to act on.
• Burnout indicators plus per-team attrition tracking. Per-engineer burnout signals (sleep loss, page volume, off-hours load); per-team attrition rate tracked alongside on-call metrics.

The approach

The practical approach is to ask the on-call factor explicitly in every exit interview, run a quarterly retention review that correlates attrition with on-call load per team, track per-engineer burnout indicators (page volume, off-hours wakes, comp time consumption), and document the retention policy in the team handbook so the link gets actioned rather than noted. The data is only useful if the team is empowered to act on it.

• Per-engineer exit interview. Per-departure the on-call factor probed explicitly; without the question, the answer never surfaces.
• Per-quarter retention review. Per-quarter attrition rate against on-call load per team; pattern surfaces before it becomes a crisis.
• Burnout indicators. Per-engineer page volume, off-hours wakes, comp-time consumption tracked; the data anchors the early-intervention conversation.
• Per-team attrition tracking plus documented policy. Per-team attrition rate visible alongside on-call metrics; per-team retention policy committed to handbook for operational review.

Why this compounds

Attrition-link discipline compounds across years. Each tracked retention preserves institutional knowledge; each correlated metric makes the link harder to ignore at the leadership level; the on-call programme stops being treated as separate from retention strategy. The opposite, where on-call and retention are tracked separately, lets the link stay invisible until senior engineers are already gone.

• Retention. Right tracking preserves teams; intervention happens before the resignation, not after.
• Operational culture. Tracking signals that on-call matters at the company level; the team treats sustainability as part of retention strategy.
• Operational improvement. Tracked link drives investment in alert quality, comp policy, and rotation design; the data anchors the conversation.
• Institutional knowledge. Each departure teaches what the team did wrong; the lesson informs the next quarter’s on-call work.

Attrition-link discipline is an organizational discipline that pays off across years. Nova AI Ops integrates with on-call telemetry, surfaces retention patterns, and supports the team’s sustainability discipline.