Exit Interview Surveys: How to Get Honest Answers from People Who've Already Left
Most exit interviews are theatre. Run an anonymous exit interview survey that surfaces real reasons. 12 questions, 30/60/90 cadence, and how to share themes back.
Hushwork Team
You ask why they left. They say "great opportunity, no hard feelings." You write it up. Three months later another person quits for the same actual reason, and you still don't know what it is.
Exit interviews are one of the most under-used signals in HR. Done well, they tell you exactly which managers, projects, and policies are pushing people out. Done as named in-person conversations, they tell you almost nothing.
Anonymous exit interview surveys fix the trust gap. People who are leaving are the most honest population in your company, but only if they believe their answer won't follow them to their next job through a reference or a backchannel. Anonymity is the unlock.
Why named exit interviews are theatre
Three things break a named exit interview:
- Reference risk. The person leaving still needs you for references. They will not say "my manager was the problem" if you might say it back to a future employer.
- Industry overlap. In most sectors, people loop back into your network. They don't burn bridges over a 45-minute conversation.
- Recency bias. The last week before someone leaves is full of farewell lunches and goodbye chat. The actual reason for leaving usually surfaced months ago and is now half-forgotten.
The result is a sanitised version of the truth. "Better opportunity" gets logged. The real reason ("I haven't had a useful 1:1 in nine months") doesn't.
The 12 questions that surface real reasons
Send a short anonymous survey. Eight to ten minutes. No login required. Hush AI summarises the free-text answers across recent exits into themes so you stop reading every survey individually and start seeing patterns.
Triggers (3 questions)
- What was the moment you started looking for another job?
- What would have made you stay?
- Was there one specific event or person that pushed you to leave?
Job content (3 questions)
- Were you doing the work you were hired to do?
- Did your role grow in the time you were here?
- What part of your work felt like a waste of time?
People (3 questions)
- How would you describe your manager's impact on your decision to leave?
- Did you feel respected by your team?
- Were there decisions about your work that were made without your input?
Company (3 questions)
- What did this company get right that you'd carry forward?
- What did this company get wrong that you'd warn a friend about?
- What's one change that would meaningfully improve this place?
The 30/60/90 sequence
Most exit surveys go out the day someone leaves. That's the worst time to send it. They're packed, emotional, and focused on transitions. Stagger instead:
- Day 1 (last day): a short pulse, 3 questions, focused on triggers. "What pushed you to leave?"
- Day 30: the full 12-question survey. They've started the new role and have perspective.
- Day 90: one last open-ended prompt. "Looking back, what's the one thing you'd want this company to know?"
You'll get the highest-signal answers from day 30 and day 90. People answer day 1 too quickly; the deeper reasons take a few weeks of contrast with the new job to surface.
What to do with the answers
Hush AI clusters open-text answers across recent exits into themes, with quotes per theme. Read it monthly.
Look for:
- A specific manager mentioned in 3+ exits
- A specific project or system mentioned in 3+ exits
- A timing pattern (people leaving 6 months after a reorg, 3 months after a manager change)
- The same word or phrase appearing in unrelated answers ("micromanagement", "no growth", "broken promotion process")
When you see a pattern, that's where you act. One person's complaint is data. Three people's identical complaint is a problem you can fix.
Closing the loop without breaking anonymity
You cannot reply to the person who left. But you can use what they said:
- Quarterly, share an anonymous summary of exit themes with leadership.
- When you make a change based on exit data, tell the team. "We heard X across recent exits. Here's what we're doing." Future leavers see the action and trust the next survey more.
- Never quote a specific exit response, even months later. Themes only.
A simple template
Hush AI will draft this in 30 seconds. Or use the 12 questions above as-is.
Send via:
- Day 1: short link in the offboarding email
- Day 30: a follow-up email to their personal address (with their permission, captured at offboarding)
- Day 90: one last prompt, same channel
Track response rates. If they drop, the survey is too long or the trust isn't there. Cut a question or check whether anonymity is actually believed.
The pattern that pays off
Companies that run honest anonymous exit surveys learn something specific within six months: usually one or two managers, one or two policies, or one or two role designs that systematically push people out. Fixing those costs less than the next round of recruiting. The survey pays for itself ten times over the first time it surfaces something nobody was talking about.
Get started
Sign in to Hushwork and ask Hush AI for an exit interview template. Edit for your context, send the link from the offboarding email, and start reading themes monthly. Free for any team size.
Related reading:
