Exit interviews: how to run them so they actually tell you something

Key takeaways

• Most exit interviews are useless because they're run by the wrong person, at the wrong time, with questions designed to be comfortable rather than honest.
• You'll get more truth if the conversation is confidential, run by someone neutral, and framed as help rather than a tick-box on the way out.
• The point of an exit interview isn't the individual conversation. It's the pattern across many of them, which is where the real signal lives.
• An exit interview is a post-mortem, not an early warning system. Pair it with something that catches people while they can still be kept.
• Data you collect and never act on is worse than no data. It just documents, in writing, the problems you chose to ignore.

An HR manager once showed me a folder of exit interview forms going back three years, genuinely proud of how thorough it was. I asked what had changed as a result of any of them. Long pause. "We keep them on file," she said. Which is a polite way of saying nothing had changed at all.

That folder is the exit interview problem in one image. Plenty of effort collecting, almost no effort using. So let's talk about how to run one that's worth the time, and what to do with what it tells you.

Why most exit interviews are useless

The honest starting point is that a lot of exit interviews are theatre. They happen because it's process, not because anyone expects to learn something and act on it. Three things usually go wrong.

They're run by the wrong person. When someone's direct manager, or the HR business partner tied to that team, runs the exit interview, the leaver isn't going to be fully honest, especially if the manager is part of why they're leaving. People manage their answers to protect a reference and avoid an awkward final fortnight.

The questions are built to be comfortable. "What did you enjoy about your role?" gets you a pleasant answer and nothing you can use. Comfortable questions produce comfortable, useless data.

They come far too late. By the exit interview, the decision is made, the next role is usually signed, and the person has already emotionally left. What you get is a tidy retrospective, not the messy real-time truth of the eighteen months that led here. I've written about that timing problem more fully in why employees leave.

There's also a self-selection trap. The people most willing to give you a full, honest account are often the ones who felt safest, which usually means they weren't the ones most let down. The sharpest feedback frequently walks out the door in silence.

How to run one that gets the truth

You can get real signal out of an exit interview, but only if you design it to overcome the reasons people hold back.

Have someone neutral run it. Not the leaver's manager. Someone from HR with no stake in that team, or an external party, gets more honest answers, because there's no relationship to protect and no reference on the line.

Make it genuinely confidential, and say so. Be clear about how the information will be used and that it won't be attributed to them personally. In a small team especially, people need to trust that before they'll say anything real.

Ask open, specific questions. Not "did you enjoy it," but "what would have made you stay," "when did you first start thinking about leaving," and "what's something we could fix that we probably don't realise is a problem." Ask about the manager relationship, workload, fairness, and whether they could see a future here.

Consider timing and format. Some people open up more in a written, confidential survey than face to face in their final week. Some will tell you more in a call a month after they've left, once the awkwardness has passed and the reference anxiety is gone. There's no single right format, so don't be rigid about it.

Listen more than you defend. The moment you start explaining why their complaint isn't fair, the conversation is over. You're there to learn, not to win.

What to actually do with the data

Here's the part that separates a useful exit interview program from a folder nobody reads. The value isn't in any single conversation. It's in the pattern across many of them.

One person saying their manager was unsupportive is a data point. Five people from the same team saying it over six months is a finding, and it's one you can act on. So the real work is aggregating: coding the themes, tracking them over time, and watching for clusters by team, by manager, and by tenure.

Then, and this is the step the folder-keepers skip, you act on the pattern and you close the loop. Take the clearest, most repeated theme and change something. You won't always be able to tell the leavers, they've gone, but you can and should tell the people still there when a change came from what departing colleagues told you. That's what turns exit data into a trust signal instead of a filing exercise.

Where exit interviews fall short (and what fills the gap)

Even a well-run exit interview has a hard limit built into it: it only ever talks to people who've already decided to leave. It's a post-mortem. It can confirm a pattern and it can occasionally surprise you, but it can't catch someone in time to keep them, because by definition you're only asking once they're gone.

That's why an exit interview program should never be your main retention tool. It should sit alongside something that asks the same honest questions while there's still time to act on the answers. That's exactly what a stay interview is: the exit interview's questions, asked of the people you want to keep, while you can still do something. Run stay interviews for the early warning, use exit interviews to confirm and refine what they're telling you, and you've got both halves. The development and listening side of that is what our Transformation work is built to support.

A short example

A retail business ran exit interviews diligently and learned nothing from them for years, because the store managers ran their own team's exits and the questions were the standard comfortable set. On paper, everyone left for "a better opportunity."

We changed two things. HR started running the exits centrally, confidentially, with sharper questions, and they began coding the answers by store rather than filing them individually. Within a couple of quarters a pattern was impossible to miss: two stores accounted for most of the regretted departures, and the same theme, a manager who played favourites, came up again and again.

That was never visible in the individual forms. It only appeared once someone looked across them. They addressed the two managers, told the remaining staff that exit feedback had driven the change, and the regretted turnover in those stores dropped. Same tool. Completely different result, because they finally used the data instead of storing it. For the fuller strategy this sits inside, the complete guide on employee retention puts it in context.

Frequently asked questions

Who should run an exit interview?

Someone neutral, with no stake in the leaver's team. The direct manager is the worst choice, because people won't be honest with the person they may need a reference from, especially if that manager is part of why they're leaving. A neutral HR contact or an external party gets far more truthful answers.

What questions should an exit interview ask?

Open, specific ones that invite honesty rather than politeness. "What would have made you stay," "when did you first start thinking about leaving," and "what's a problem we probably don't realise we have" tell you more than "what did you enjoy." Cover the manager relationship, workload, fairness and whether they could see a future here.

Are exit interviews even worth doing?

Yes, but only if you aggregate and act on them. A single exit interview is a weak signal, and one you can never act on for that person. The pattern across many of them, coded and tracked over time, is genuinely valuable. If you collect them and never look for the pattern, you're documenting problems, not solving them.

What's the difference between an exit interview and a stay interview?

Timing, and therefore usefulness. An exit interview happens once someone's already leaving, so it's a post-mortem you can't act on for them. A stay interview asks the same honest questions of people you want to keep, while there's still time to change something. Use stay interviews for early warning and exit interviews to confirm the patterns.

If you pulled together every exit interview from the last two years and read them as one document, would a pattern jump out at you, and would you be comfortable with how little you'd done about it?