Candidate experience survey questions, from application to rejection
Six question sets you can copy, one per stage of hiring plus one for the candidates you turned down, with the wording rules behind them and a downloadable sheet for each. Part of the wider candidate experience platform.
Hari SContent MarketerBuilds narratives on employee engagement, and on the hiring experience that comes before it.What makes a candidate survey question worth asking
Candidate experience questions are not employee questions with the nouns swapped. A candidate has no obligation to answer you: no manager is nudging them, no engagement target depends on it, and they may have just been rejected. That single fact decides everything about how these questions should be written.
Five rules do most of the work, and every set further down follows them.
One idea per question
"How clear and timely was our communication?" is two questions wearing one number. When it scores badly you cannot tell which half broke. Split it, or drop the half you will not act on.
Answerable by someone you rejected
Most people who answer will not have got the job. A question that only a hire can answer quietly turns your survey into a survey of your best case. Every question below works either way.
Tied to one stage, not to the whole process
Asking someone to rate three weeks of hiring in a single number produces an average of a good interview and a bad wait, which points at nothing. Ask about the step just taken, and a weak score arrives with an address attached.
The same wording, every quarter
Rewording resets your history. A movement in a question you have just rewritten is not a movement, it is a different question. Improve the wording once, then leave it alone.
One response format, plus one open question
Mixing 0 to 10 scales, agreement statements and star ratings inside one survey makes the results impossible to compare and slower to answer. Keep one scale, and spend your last question on free text, which is where the reason for every other number turns up.
The practical ceiling is five questions. Not because five is a magic number, but because candidate response rates fall off a cliff shortly after it, and a longer survey answered by fewer people is worse data than a short one answered by most.
Candidate experience survey templates, stage by stage
Six sets of candidate experience survey questions, one per point in hiring where a candidate has something specific to tell you. Copy them as they are, or cut a question you will not act on. The whole set downloads as one sheet, grouped by stage, with the response type against every question.
Do not run all six at once. Start with the two stages where you already suspect a problem, and add the rest when you have acted on something.
Every question on one sheet
All 27 questions across 6 stages, response type against each, on one sheet.
Application
Send on submission, or within 48 hours of it
- Q1How easy was it to find what you needed to know before applying?0 to 10
- Q2How reasonable was the length of the application?0 to 10
- Q3How clearly did the job description describe the actual role?0 to 10
- Q4What, if anything, nearly stopped you finishing the application?Open
This is the only set most applicants will ever see, because most never reach an interview. Q4 is the one that pays for the set: it is the only place a broken form or a hidden salary range will show up in words.
Screening call
Send the day after the first conversation
- Q1How clearly did the recruiter explain the role and the process?0 to 10
- Q2How clear were you on the next step and when it would happen?0 to 10
- Q3How well did the conversation respect your time?0 to 10
- Q4What would have made that first conversation more useful?Open
Q2 is the diagnostic. When it scores below the rest of the set, the problem is almost never the recruiter's manner, it is that nobody told them what the next step actually was.
Interview
Send within 48 hours of the last interview in the loop
- Q1How prepared were the interviewers you met?0 to 10
- Q2How fair a chance did the questions give you to show your ability?0 to 10
- Q3How clear were you on how and when the decision would be made?0 to 10
- Q4How closely did what you heard match the job advert?0 to 10
- Q5What is the one thing you would change about the interview?Open
Q1 is the strongest single predictor of a poor fairness score in candidate feedback, and it is the question most worth cutting by interviewer rather than reading as an average.
Offer and decision
Send once the outcome is communicated, to everyone who reached this stage
- Q1How closely did the decision arrive to when you were told it would?0 to 10
- Q2How clear were the reasons behind the decision?0 to 10
- Q3How respectfully were you treated through the final stage?0 to 10
- Q4How likely are you to recommend applying here to someone you know?0 to 10
- Q5What is the main reason for that score?Open
Q4 is the candidate NPS anchor. It belongs at the end of this set rather than at the start, because a recommendation score given before the reasons are asked for is a score with nothing attached to it.
Offer to first month
Send at the end of the first month in the job
- Q1How well did contact between accepting and starting keep you confident in the decision?0 to 10
- Q2How ready was everything you needed on your first day?0 to 10
- Q3How closely does the job match how it was described during hiring?0 to 10
- Q4What would have made your first month easier?Open
Q3 is the one question on this page that measures hiring rather than the hiring process. A low score here is a promise made in an interview that the job did not keep, and it predicts early attrition better than anything asked before day one.
The sixth set is the one most teams never send, and it is the one worth arguing for.
Candidates you turned down
Send with, or shortly after, the rejection itself
- Q1How likely are you to recommend applying here to someone you know?0 to 10
- Q2How clear was the reason you were not selected?0 to 10
- Q3How closely did the outcome arrive to the timing you were given?0 to 10
- Q4How likely are you to apply here again?0 to 10
- Q5What is the one thing we could have done better?Open
Q4 is the most useful number on this page and almost nobody collects it. A candidate who would apply again is a candidate your process did not lose, whatever the outcome was.
Almost everyone who meets your hiring process is rejected by it, so this set covers more people than the other five combined. It is also the only place you will find out whether a rejection was survivable. The candidate experience best practices cover what to do with the answer.
The questions are the easy part
CultureMonkey triggers each set at the stage it belongs to, holds the reporting threshold, and cuts the results by stage, recruiter and hiring manager, so a weak score arrives with a name and a step attached.
The survey email decides how many of these get answered
A good question set sent badly returns nothing. Candidates have no company inbox, no workspace account and no reason to check a portal, so the invitation email carries more weight here than it ever does with employees. There is no template worth copying for it, but there are five things it has to do.
- Say how long it takes, and be right. "Four questions, about a minute" is worth more than any subject line. If it is really nine questions, you have spent the trust you will need next quarter.
- Send it from a person, not from a mailbox nobody reads. A reply to a no-reply email address is a candidate who tried to tell you something and could not.
- Say what happens to the answer. Not that it matters to you, but what it is used for and who sees it. Confidentiality in reporting, stated plainly, is more persuasive than a promise of anonymity you cannot keep.
- Do not send it in the same message as the rejection. Give it a few hours at least. The answers are more useful and the request reads less like a demand.
- Ask in the language they applied in. A survey in the wrong language is a non-response counted as indifference. multilingual candidate surveys covers how that works across markets.
One follow-up is worth sending. Two is not, and the second one costs you more goodwill than the responses are worth.
Questions to leave out
These appear on most candidate feedback forms and earn their place on few.
"How did you hear about us?"
A sourcing question wearing a candidate experience costume. It is useful data and it belongs in the application form, where it costs nothing, not in a five-question survey where it costs you a real question.
Anything about the decision itself
"Do you feel the right decision was made?" invites a rejected candidate to relitigate an outcome you are not going to change. It produces heat rather than signal, and it makes the rest of the set feel like a formality.
Demographic questions, unless you have a specific plan for them
Collected without a stated purpose, they lower response rates and raise reasonable questions about why you are asking a rejected candidate for personal detail. If you are running a fairness analysis, say so in the invitation. If you are not, leave them out.
"Any other comments?" as the only open question
An unfocused prompt at the end of a set returns thanks, complaints and nothing in between. Point the open question at the stage, as every set above does, and the same people write something you can use.
Reading the answers without fooling yourself
Good candidate experience survey questions still produce a slide nobody acts on if they are read badly. Three habits separate the two:
- Read the open text before the scores. Scores tell you something moved. Only the free text tells you what happened, and reading it second means you arrive with a theory to confirm.
- Cut by stage, recruiter and hiring manager before drawing a conclusion. An overall score that has not moved often hides one stage improving while another slips. The candidate experience metrics covers what to track and at what grain.
- Compare against your own history, not an industry figure. Published candidate benchmarks vary too much by source, sector and question wording to be worth measuring against. candidate NPS sets out why.
If you would rather fix the obvious things before asking anyone, the candidate experience checklist covers what a survey would have told you anyway. For how the surveys themselves get sent and scored, the candidate experience survey is the place to start, and the candidate experience software compares the tools that run them.
Candidate survey questions, answered
What questions should be on a candidate experience survey?
Ask about the stage the candidate has just been through rather than the process as a whole, in four or five questions. A workable set covers how clear the information was, how well the timing matched what they were promised, how prepared and fair the people they met were, and one open question pointed at that same stage. Add the recommendation anchor, how likely they are to recommend applying here, once per candidate rather than once per survey. Every question should be answerable by someone you rejected, because most of the people answering will have been.
Is there a candidate experience survey template we can copy?
Yes. This page carries six, one per hiring stage plus a set written for the candidates you turned down, and each downloads as a one-page sheet with the response type against every question. They are editorial templates you are free to lift and reword, not a product feature. Copy the two that match the stages you are most worried about rather than running all six at once.
How many questions should a candidate survey have?
Five or fewer. Candidates have no obligation to answer, which is exactly what makes their response rate an honest signal, and length is the fastest way to lose it. Four scale questions and one open question is enough to produce a score you can act on. If a question is on the survey because it would be interesting rather than because you will do something about the answer, cut it.
When should each survey be sent?
Within about 48 hours of the stage it asks about, while the candidate can still remember specifics rather than an overall impression. The exception is the onboarding set, which belongs at the end of the first month, and the rejection set, which is better sent a few hours after the rejection itself rather than attached to it.
Should we survey candidates we rejected?
Yes, and it is the set most teams skip. Almost everyone who meets your hiring process is rejected by it, so a survey sent only to hires measures your best case and misses the group that shapes your reputation in the market. The most useful single question in the whole set lives here: how likely they are to apply again.
What is the difference between a candidate experience survey, a questionnaire and a feedback form?
In practice the three terms are used interchangeably, and a candidate experience questionnaire and a candidate experience feedback form usually mean the same thing this page calls a survey. The useful distinction is scope. A feedback form is usually one open-ended box collected at the end of the process, which produces comments nobody can compare over time. A survey asks the same short set of questions at the same stage every time, so the answers form a trend rather than an anecdote.
Can we ask these questions anonymously?
Candidate feedback usually cannot be anonymous the way employee feedback is, because it has to stay attached to a stage and a role to be worth reading. What can be promised honestly is confidentiality in reporting, where scores are held back until a cohort is large enough that no individual can be identified. Say that plainly in the invitation rather than promising anonymity you cannot deliver.