The exit interview is the wrong conversation, held too late

The exit interview is one of the most trusted instruments in people analytics, and one of the least useful. Not because people lie on their way out — though the incentives are not neutral — but because it asks the right questions at precisely the wrong moment.

Signal decays

The decision to leave is the end of a process, not its beginning. By the time it surfaces in an interview, the turning point — the project that quietly broke trust, the promotion that went elsewhere, the manager change that removed the reason to stay — is months in the past. What you capture at exit is a rationalised summary, not the mechanism.

An exit interview measures the aftermath of a decision the organization could have read while it was still being made.

Where the conversation should happen

The useful signal lives in the ordinary middle of an engagement, not at its edges:

• In the shift from proposing ideas to simply executing them.
• In the one-to-one that gets shorter, then perfunctory, then rescheduled.
• In the language that moves from "our roadmap" to "the roadmap."

A different instrument

Reading engagement as it happens — qualitatively, in context — is not about predicting resignations. It is about noticing the mechanism early enough that leadership still has moves available. By the exit interview, the only move left is goodbye.