Free tools for HR leaders

Exit interview questions

14 questions across 6 themes, each with the scale to use. Filter, copy, print. Free, no sign-up.

Pick a theme, then a format. Every question carries the scale to use with it, because a good question on the wrong scale produces data you cannot act on. Copy what you need, then cut the survey down until people will actually finish it.

How these were written, and the five steps to a survey worth running ↓

Theme:

Format:

14 showing

Reason for leaving

What made you start looking?

Open text. Keep it optional.Open text

What could we have done to keep you, and when?

Open text. Keep it optional.Open text

I had raised the issues behind my decision before resigning.

Five-point agreement scale, strongly disagree to strongly agree.Rating scale

Manager

My manager supported my development.

Five-point agreement scale, strongly disagree to strongly agree.Rating scale

What was the working relationship with your manager like?

Open text. Keep it optional.Open text

Growth

I could see a realistic next step for myself here.

Five-point agreement scale, strongly disagree to strongly agree.Rating scale

Where did your career stall, if it did?

Open text. Keep it optional.Open text

Culture and team

The culture here matched what I was told during hiring.

Five-point agreement scale, strongly disagree to strongly agree.Rating scale

What would you warn a friend about before they took a job here?

Open text. Keep it optional.Open text

Pay and recognition

I was paid fairly for the work I did.

Five-point agreement scale, strongly disagree to strongly agree.Rating scale

How much of your decision was about pay?

Open text. Keep it optional.Open text

Would you return

I would consider working here again.

Five-point agreement scale, strongly disagree to strongly agree.Rating scale

I would recommend this organisation to someone looking for work.

Five-point agreement scale, strongly disagree to strongly agree.Rating scale

What should we not change?

Open text. Keep it optional.Open text

No questions match those two filters together. Clear one of them.

What questions should you ask in an exit interview?

What made them start looking, what could have kept them and when, and whether they raised it before resigning. That last one tells you whether your escalation paths work.

How these were written

Nothing here is calculated. Each question is written against two tests: can somebody answer it honestly, and can you act on the answer. Each one carries the scale it should be asked on, because the scale is part of the question and mixing scales inside a survey makes results much harder to compare. Filter by theme to cut whole sections rather than trimming words, and by format to balance rating questions against open text.

Assumptions

These are a starting point, not a survey. Cut hard. Response rate falls with every question you add.
Scales are part of the question. Five-point agreement for most, frequency for behaviour. Pick one convention and hold it.
Anonymity has to be explained, not asserted. People assume the worst when you are vague, and it shows up in the answers.
Written for Australian workplaces and Australian English. Compono product terms are reproduced exactly as they appear in the platform.
Do not ask more often than you can act. Asking without visible follow-through does more damage than not asking at all.

Sources

Compono, The culture, engagement and performance model

What to do about it

Most survey problems are not question problems. They are what-happens-next problems, and people can tell the difference by the second round.

1

Decide what you will do with the answers before you ask

If no decision is waiting on the result, do not run it. An unused survey teaches people not to answer the next one.

2

Cut it until it hurts

Remove whole themes rather than trimming words. Eight sharp questions beat forty thorough ones nobody finishes.

3

Say how anonymity actually works

Who sees what, at what group size, and what happens to free text. Vagueness reads as surveillance.

4

Report back before you fix anything

Tell people what you heard, including the uncomfortable parts, within two weeks. This is the single biggest driver of next year's response rate.

5

Where a tool helps

Pick one thing and visibly change it

One change people can see beats five they cannot. The point of a survey is the change, not the score.

They already decided.
Ask when you lost them.

Compono Engage surfaces the signals that precede a resignation, so more of this conversation happens while the person still works here and something can still be done about it.

Step 5 is where most survey programs stall: the score arrives and nobody agrees what it means. That is an interpretation problem, and interpretation is what the platform is for.

See how it works

Compono Engage

8%+

reduction in turnover

Common questions

Do exit interviews actually work?

Only if somebody outside the reporting line asks, and only if the answers reach the person who can act. Otherwise you are collecting polite fiction.

Who should conduct the exit interview?

Not the departing person's manager. If the manager is part of the reason, that conversation cannot happen with them in the room.

When should you do an exit interview?

Late in the notice period, and consider a follow-up a few months after they leave. People are noticeably more candid once they are properly gone.

These questions are a starting point, not a survey. Cut hard: response rates fall with every extra question, and a short survey people finish beats a thorough one they abandon. Say what you will do with the answers before you ask, and be honest about how anonymity actually works, because people assume the worst when you are vague. Last reviewed July 2026.

Next step

Employee engagement

What engagement actually measures, and how it differs from satisfaction.

Read the definition →Related tool

Cost of employee turnover calculator

Put a number on what the survey surfaces.

Run it next →Buying

Requirements builder

Score the platforms you are looking at against the two risks, must-haves first.

Build the scorecard →Talk to us

Bring us your results

A conversation about what your survey is telling you, not a demo script.