The exit interview that produces information, not pleasantries
Echipa HR 365 · reviewed 2026-09-04 · 5 min read
An exit interview produces information only if somebody other than the direct manager runs it, if the questions are about facts rather than feelings, and if the answers are read not individually but as a pattern, across several departures.
Who runs it
Not the direct manager. It is the most common arrangement and the one that guarantees polite answers: the person is leaving, but needs a reference, might come back, and does not want an awkward conversation in their final week. In a small company the most frequent real reason for leaving is the relationship with the direct manager — exactly the person asking.
HR runs it, or a manager from another department if HR is not distant enough. In very small companies, where everybody is close, a written form sent after the last day produces more honest answers than any face-to-face conversation.
When, and why not on the last day
| Moment | What you get | What you lose |
|---|---|---|
| At resignation | The freshest and least filtered reason | The person is still in conflict; they may exaggerate |
| A week before leaving | A calmer perspective, concrete details | They have already started detaching |
| On the last day | Almost nothing — it is a farewell day, not an analysis | Everything |
| 4–6 weeks after leaving, in writing | The most honest answers, with nothing to lose | The response rate drops by about half |
The nine questions
All of them are about facts and moments, not about states. “Did you feel appreciated?” gets “yes, on the whole”. “When did you last feel your work mattered to somebody?” gets a story.
- When did you first seriously think about leaving? (The date matters more than the reason — it shows what happened then.)
- What happened around that time?
- What would have had to change for you to stay? (Not “what would have convinced you” — that invites salary.)
- Who here did you tell about it, and what did you say?
- What about your work did you enjoy and have not found elsewhere?
- What did you learn here, concretely?
- What did we never ask that we should have?
- What should the person taking your place know, in their first week?
- Would you recommend a job here to a friend? For which role yes, and for which not?
The fourth question is the most valuable on the list and almost nobody asks it. If the person flagged the problem to somebody and nothing happened, you have two pieces of information at once: what the problem was, and where the signal stopped. The second is usually more important.
What you do with the answers
You do not read them individually, as verdicts. An exit interview is a single perspective, from a charged moment. The information appears in the pattern: the same theme in three departures from the same department, within six months, is a fact. The same theme in one departure is an opinion.
The routine that turns answers into something
- HR — records the answers against fixed themes, the same every time — otherwise they cannot be compared across departures
- HR — passes nothing attributable to the direct manager; only the theme, and only when it appears a second time
- HR — every six months, reads all the interviews from the period together and looks for repetitions
- HR and leadership — discuss the repeated themes, not the individual departures
What you do not ask and what you do not promise
- Where you are going and what they are paying you. It is not your business, and the question turns the conversation into an interrogation.
- What you think of colleague X. An exit interview is not the place for assessments of other people.
- Do not promise that “things will change” in an attempt to retain them. A counter-offer at the exit interview rarely works and usually postpones the departure by a few months.
- Do not ask for confidentiality about the conversation. Somebody leaving has no reason to honour it, and the request itself suggests you have something to hide.
What you gain beyond the information
A well-run exit interview has a second, less discussed effect: it changes how the person talks about the company after they leave. In a local market, where everybody in the same field knows each other, that matters more than any employer branding campaign.
The third effect is the most practical: some people who leave come back, a year or two later. Somebody who left properly and was listened to returns more easily than somebody who left through an awkward final day — and they come back already trained.
Where to start
Change who runs the interview first. It is the one modification that changes the quality of the answers with no other effort, and it is free.
Then move the timing: a short conversation a week before for the operational part, and a form a month after leaving for the reasons. The second part will give you answers you have not had until now.
Pass nothing attributable onwards, not even once. The instrument is lost permanently the first time an answer reaches the person it was about.
Exit forms with answers collected against the same themes
The questions stay identical from one departure to the next and the answers are read in aggregate, so the pattern shows without reconstructing anything from memory.
You can create an account in a few minutes and use every module for 7 days, no card required.