Exit survey questions

What this survey is for

An exit survey collects what someone is willing to say on the way out. Used well, it identifies patterns you can fix: a manager whose team keeps leaving, a grade where pay has fallen behind, a shift pattern nobody can sustain. Used badly it collects polite non-answers about pursuing new opportunities.

The design problem is candour. People protect references, future contact and former colleagues. So ask specific, factual questions rather than inviting a verdict, separate the trigger from the underlying cause, and make clear that answers will not reach a reference. A written survey plus an optional conversation with someone outside the leaver's reporting line gets further than either alone.

The core question set

The decision

• What prompted you to start looking? Select all that apply: pay or benefits, progression, workload, manager, team, commute or location, working pattern, job content, organisational change, personal or family reasons, something else.
• Which of those mattered most?
• How long had you been considering leaving before you resigned? Under a month, one to three months, three to six months, over six months.
• Was there a point where something could have changed your mind? If so, what?

The experience

• My work was recognised accurately.
• I was paid fairly for the work I did.
• I had a realistic route to develop here.
• My workload was sustainable.
• My manager supported me.
• I could raise concerns and be taken seriously.

The new role

• What does your new role offer that this one did not? Optional, free text.
• Would you consider working here again? Yes, no, maybe.
• Would you recommend us as a place to work? Yes, no, maybe.

Open question

• What one change would most help the person who replaces you? Optional, free text.

Response scale and scoring

Use the five-point agreement scale for the experience block so it can be read against your engagement results. Keep the decision block as structured choices, because free text alone produces unquantifiable sympathy.

Report quarterly, not per leaver, and always with the leaver count. Three resignations in one team is a pattern; one is an event. Cut by manager, grade, tenure band and location, and watch the regretted leaver population separately: losing people you wanted to keep is the number that should drive action. Read "how long had you been considering leaving" closely, because a long average means your engagement survey had time to catch the problem and did not.

Adjust it for your sector

Leaving looks different where work is seasonal, licensed, rota-based or project-bound. Adjust the choice list rather than inventing new scales.

Professional services

Add "utilisation expectations" and "route to partner or principal" to the trigger list. Ask whether the leaver is moving to a client or a competitor, which matters for handover, and run the survey after the notice period has settled in professional services firms.

Construction and property

Add "project ending" and "travel to site" to the list, and ask whether they are moving to another contractor or going self-employed. Separate genuine resignations from end-of-project moves, which otherwise inflate turnover.

Technology

Add "technical direction", "on-call burden" and "remote working policy". Ask whether compensation structure rather than salary drove the move, since equity and levelling often matter more than base pay.

Banking

Add "regulatory or control workload" and "bonus outcome". Keep the survey entirely separate from any conduct process, and route anything disclosing misconduct through the proper channel rather than the survey report.

Healthcare

Add "staffing levels", "rota pattern" and "moving to bank or agency work". Ask whether they are leaving the profession or only the employer, because the two have completely different implications.

Energy

Add "rotation pattern" and "site location". Ask whether the move is to a contractor or operator, and time the survey after demobilisation so the answer is not collected in the last week of a rotation.

Media and agencies

Add "pitch workload" and "creative direction", and ask whether they are going client-side, to another agency or freelance. Read results by account team, since one punishing account can produce a cluster of leavers.

Public sector

Add "grade and progression", "restructure" and "moving to another public body". Note whether the leaver is retiring, since retirement patterns otherwise distort turnover reporting by grade band.

Education

Add "workload in term time", "behaviour and support" and "leaving teaching". Anchor timing to the resignation deadlines rather than the calendar month, and report by department and phase.

Universities

Add "fixed-term contract ending", "research funding" and "promotion criteria". Separate the end of a funded contract from a resignation, and report academic, research and professional services staff separately.

Hospitality

Keep it to five questions on a phone, because the response window is short and turnover is high. Add "rota predictability", "tips and service charge" and "seasonal end". Report by site and tenure band, with under-three-months treated as an onboarding signal.

Retail

Add "hours offered", "shift pattern" and "store transfer refused". Keep it short and mobile. Report by store, and track whether leavers are moving to another retailer or out of retail entirely.

Manufacturing

Add "shift pattern", "overtime availability" and "physical demands of the role". Offer paper and a low reading level, and report by shift as well as by line, since one shift's supervision often explains a cluster.

Anonymity, consent and small teams

Exit surveys are confidential, never anonymous: you know who is leaving. Say that plainly. Tell leavers who reads the answers, confirm the content will not appear in a reference, and keep it out of the personnel file. Where volumes are low, publish only aggregated themes, because a quarterly report from two leavers identifies both.

Set a retention period for the responses and apply it. If a leaver discloses discrimination, harassment or a safety failure, acknowledge it, explain the route it will follow and follow it. A survey is not a grievance process, and treating a serious disclosure as data is the fastest way to lose the trust of everyone still employed.

From results to action

Pair the exit results with the onboarding survey once a quarter. Together they show whether you are losing people because of how they arrived or because of what happened later, and they usually implicate different owners.

Where a manager appears repeatedly, act on the pattern rather than the individual complaint: coaching, a change in span of control or a closer look at the team's own upward feedback. Where pay or progression dominates, the fix is a commercial decision, so report it as one rather than as an engagement problem. Record the agreed changes where the work is tracked, then check the next quarter's leavers for the same reasons.

Frequently asked questions

Do exit surveys actually produce honest answers?

Partly. People understate manager and culture problems and overstate pay, because pay is the socially safe answer. You reduce that by asking structured questions about specific conditions, by separating what triggered the search from what made them look, and by having the conversation with someone outside the reporting line. Treat the results as a directional signal corroborated by your engagement data.

Should we survey or interview leavers?

Both, in that order. The written survey gives comparable data across every leaver; the optional conversation adds the detail that explains it. Make the conversation voluntary, never conducted by the leaver's own manager, and ask for permission before sharing anything specific from it.

What about people who are dismissed or made redundant?

Do not send this set. The questions assume a voluntary choice, and sending them after a dismissal reads as careless. For redundancy, ask separately about how the process was handled, which is a consultation quality question rather than a retention one, and keep it clear of the individual's circumstances.

About New Dynamics

New Dynamics is a Bristol-based performance management business. Most of what an exit survey reveals should have surfaced in a one-to-one months earlier, which is the problem our software exists to fix. Book a demo if you want those conversations recorded and followed up.