An employee recognition survey that asks why, not how often
Nearly every survey on this subject resolves to one number: how much recognition people get. That is a rate with no reason attached, which is why it never moves. The twelve questions below ask what a rate cannot.
- 12questions
- About 3 minto answer
- 5blocks you can cut
- 0Gallup Q12 items
Twelve employee recognition survey questions, and what each one is for
Five blocks. Each is a failure a recognition scheme can have on its own, with its own fix, which is why they are worth keeping apart rather than averaging into a score.
Does recognition here land at all
Three items about quality rather than quantity. They open the survey because somebody who has never been recognized personally can still answer all three about the place they work.
- When someone is recognized here, it means something.The summary read, and it asks whether recognition means anything rather than whether it arrived. Its parent item scored 64 percent positive across close to 187,000 Canadian federal public servants.[4]
- In my work unit, differences in performance are recognized in a meaningful way.Printed word for word so the published benchmark still holds when you compare against it: 47 percent positive in 2024, the lowest-scoring item of all 90.[2]
- Recognition here names the specific thing that was done.Specificity, asked about the organization rather than about one manager. It is the cheapest thing on this list to change and the thing that turns recognition into an instruction rather than a compliment.
Is it fair, and could anyone explain it
The two questions this subject skips. A scheme can be busy and still be unfair, and fair and still be a mystery to the people inside it. Neither shows up in a rate.
- Recognition here reaches more than the same few people.Distribution, worded so agreement is the good answer and it reads the same way round as its neighbours. Split this one by team before you average anything.
- I could explain how it is decided who gets recognized here.Procedural transparency, asked as something a person can do rather than something they feel. If nobody can explain the rule, the scheme has a publishing problem and not a budget problem.
Where it comes from, and what form it takes
Three questions with no rating scale on them. The manager appears here as one option of six and nowhere else on this survey: a read on a named manager is a different instrument and it is linked below.
- Whose recognition means the most to you here?Read this against items 1 and 2 before deciding whether managers or the organization is the thing to fix. The national figures say those two answers come apart.[1]
- Which kind of recognition would you rather have?A preference question with no free-text follow-up, so nobody has to explain an answer they were allowed to skip. Money sits in the list as one option: a meta-analysis of 72 behavioural management studies found that money, feedback and social recognition each worked alone and worked strongest in combination.[8]
- Which kind of recognition would you rather not have?The mirror of the one above and the more useful half of the pair. A public mention is one person's reward and another person's worst afternoon, and a scheme cannot know which without asking.
Whether anyone can give it
The giving side. Recognition nobody is able to hand over is a policy rather than a practice, and these two items are the ones a manager can act on this week without a budget.
- I can recognize a colleague's work here without needing anyone's approval.Permission. A low answer here beside a low answer to item 4 describes a scheme where recognition is rationed from one desk.
- I know what to do when I want to recognize a colleague.Knowing the mechanism, which is a different failure from not being allowed. This one is fixed with an email rather than a policy change.
In their own words
Two open boxes, both optional and both last, so somebody who stops here has already answered everything you can count.
- What would make recognition here mean more to you?Phrased as a change rather than as a complaint, so the answers arrive as things somebody could actually do.
- Anything else about recognition here that these questions missed?The catch-all. Read it before the next wave and it will tell you which item to add.
What is deliberately missing. No question here asks how often you have been recognized, or whether it happened inside a set number of days. That is Gallup's item 4, it is licensed,[6] and it is a good question with one job: it gives you the rate. All twelve above are written to give you the reason.
Add two fields the twelve do not include: team or function, and how long somebody has been here. Items 4 and 6 only mean anything split by one of those, so add them as optional fields at the end rather than opening with them.
Why a recognition rate cannot tell you what to fix
Four national workforce surveys, none of them selling a recognition platform, publish their own recognition results and their own item wordings. Read together they settle the question.
One survey, four questions, and the gap between them is the whole argument. In the 2025 NHS Staff Survey, answered by 766,285 people, 72.84 percent agreed their immediate manager values their work and 67.34 percent said the people they work with show appreciation to one another. In the same survey, 53.10 percent were satisfied with the recognition they get for good work and 42.95 percent with the extent to which their organization values their work.[1]
None of the four has moved more than two points in five years, so this is a settled shape and not a bad year. They say something a single rate cannot: the manager is usually not where the problem sits, and the organization usually is.[1] A survey that asks only whether you get recognized reports the average and points at nobody.
| What one recognition rate tells you | What it cannot tell you |
|---|---|
| That recognition is thin, and roughly how thin, in a number you can trend | Whether it is thin for everyone or thick for a few, which is a different problem with a different fix |
| A figure you can set against a published benchmark | Whether the people who want to give recognition are able to, or know how |
| That something is wrong | Whether it is wrong with the managers or wrong above them |
In the United States the recognition item is the worst question on the whole survey. "In my work unit, differences in performance are recognized in a meaningful way" scored 47 percent positive in the 2024 Federal Employee Viewpoint Survey, and no other item out of ninety scored lower.[2] Satisfaction with recognition, asked separately of the same 674,000 people, scored 57 percent.[2] Ten points apart: how people feel about recognition and whether it tracks what they did are two measurements, not one, which is why questions 1 and 2 are both here.
What a good answer looks like, from two more surveys that publish theirs. In the 2024 Civil Service People Survey, 83.7 percent of respondents in the median organisation answered positively on the manager recognition item, on a series running since 2009, while the item beside it about regular feedback on performance scored 73.6 percent.[3] In Canada, 64 percent of federal public servants said they receive meaningful recognition for work well done.[4] The items that ask about a manager score high. The items that ask about the organization are the ones that do not, and question 6 is how you tell which one you have.
Where recognition lands matters as much as how much of it there is. In a field experiment with more than 300 people on a short paid data-entry task, a thank-you card given to everyone raised later output by 5.2 percent and a card given to only the best three of eight raised it by 7.3 percent, with most of that second increase coming from the people who did not get one.[7] Three hours of data entry is not a career, and the authors put it down to comparison between colleagues rather than to motivation.[7] It is still the best published reason to ask who recognition reaches.
Reading the answers, and the three changes worth making
Twelve questions means a short reading. In this order it produces a decision rather than a dashboard.
Read the first block apart from the second
Items 1 to 3 say whether recognition lands. Items 4 and 5 say whether it is fair and whether anybody can explain it. A place can score well on one pair and badly on the other, so averaging them into one score hides the thing you came for.
Split question 6 by the answers to questions 1 and 2
If people name their own manager and nothing else, whatever the organization runs is not reaching them, and the fix sits above the manager. If nobody at all is named, the problem is wider than recognition and the engagement survey with a recognition driver will show you how much wider.
Change three things, wait thirty days, re-ask what you changed
One item from each of the first two blocks, one thing from the open answers. Then send the same wording to the same people: a comparison only holds if the wording did not move between waves.
When this is the wrong survey. If what you already suspect is that people are leaving, recognition questions will not tell you: run the employee attrition survey template and let it say whether recognition is one of the reasons. If what you want is mood rather than a scheme, the employee morale pulse asks it in plainer language. If the question is what this place rewards in general, that is the company culture survey template.
The manager question is a different survey. Whether a named manager notices good work, and whether they say what was good about it, belongs on the manager feedback survey, which asks a whole team about one person. This page names the manager once, as one option of six in question 6, and never rates them.
All twelve are already built. Open them in the editor, change any wording and send them today, or start from an empty page and build a survey for free.
Questions people ask about recognition surveys
Is the seven-day recognition question the Gallup one, and can I use it?
Yes, it is. The question asking whether recognition arrived inside the last week is item 4 of the Gallup Q12, and Gallup's Product Terms of Use state that the Q12 items are its copyrighted material and cannot be reproduced or used outside a participating company without express permission.[6] Nothing here reproduces it, rewords it or rebuilds it in another shape: no question on this page counts, rates or recalls recognition inside a time window, because the window and the receipt framing are the item. The wider answer on reusing published instruments is on the employee experience survey template.
How many questions should an employee recognition survey have?
Twelve here, which takes about three minutes. Lists of forty, fifty and seventy questions on this subject are menus to choose from rather than instruments to send; twelve fits one phone screen and still separates the five failures the blocks are named after. If that is more than you want, send one block.
What is a normal score, and compared to what?
National workforce surveys publish theirs, so no vendor benchmark is needed. Question 2 is printed word for word from the US federal survey, where it scored 47 percent positive.[2] The nearest comparisons for the rest are 64 percent in the Canadian federal public service,[4] 83.7 percent on the UK civil service manager recognition item,[3] and 53.10 percent for satisfaction with recognition in the NHS.[1] Four populations and four wordings: a range, not a target. Compare your second wave against your first.
What kind of recognition do people actually want?
Nothing published will tell you, and that is the honest answer: preference varies person by person and cannot be inherited from somebody else's benchmark. Questions 7 and 8 exist for that, and 8 is the more useful of the pair, because a form of recognition somebody dreads costs more than one they shrug at. If the answer that keeps coming back is money, that is a pay question and the pay fairness survey can settle it.
Does a recognition survey have to be anonymous?
It has to be honest about who reads it, which is not the same thing. Put one line at the top saying who will see the answers, and if a manager will, say so. That line is worth more than a promise of anonymity nobody can keep on a team of six. When the answers you need are ones people will not give to somebody who knows their name, the setup for that is on the anonymous employee feedback template.
Recognition covers everything from a thank you to a formal award. Where a company runs a named monthly scheme and the question is whether staff think it is fair and worth keeping, the employee of the month survey questions ask about the programme itself.
A survey collects opinion from everybody. A nomination is one person putting forward one colleague with evidence, signed and never anonymous, which is what the employee of the month nomination questions collect.
Methods and sources
What this template is based on
The five failure modes the questions are grouped by come from four national workforce surveys that publish their own recognition results and their own item wordings.[1][2][3][4] Ten of the twelve questions are written for this page.
Question licensing
Question 2 is US Office of Personnel Management, Federal Employee Viewpoint Survey 2024, question 17, reprinted as a work of the United States government.[2] Question 1 is adapted from the Public Service Employee Survey 2024 under the Statistics Canada Open Licence, which requires the line that rides with the copy button: this does not constitute an endorsement by Statistics Canada of this product.[5] The Gallup Q12 is named and cited and never reproduced.[6] The NHS Staff Survey is quoted for its published results only, because its questionnaire carries an all rights reserved notice and no reuse licence.[1] Everything else here is yours to copy, edit and translate.
Limits and disclosure
The four national comparisons come from public-sector workforces in three countries and use four different wordings, so they set a range and not a target. The field experiment behind the distribution question ran for three hours on a paid data-entry task and its own authors attribute the result to comparison between colleagues.[7] The 2024 Federal Employee Viewpoint Survey is the most recent governmentwide US result. No figure on this page comes from SuperSurvey.
SuperSurvey makes the editor this template opens in. The twelve questions are free to copy, edit and translate with or without it.
References
- NHS England and the Survey Coordination Centre. 2025 NHS Staff Survey: National results briefing, published March 2026, section "We are recognised and rewarded". 766,285 respondents, 49 percent response rate. nhsstaffsurveys.com. Quoted for its published results only: the site carries an all rights reserved notice and no reuse licence, so none of its questions is reproduced here.
- U.S. Office of Personnel Management. 2024 Federal Employee Viewpoint Survey Results: Governmentwide Management Report. Over 674,000 respondents, 41 percent response rate. opm.gov. A work of the United States government.
- Cabinet Office. Civil Service People Survey 2024: Benchmark Results, measures B13 and B14, and the accompanying Quality and Methodology Information. 354,962 respondents across 103 organisations, 61 percent response rate, fieldwork 10 September to 8 October 2024. gov.uk. Crown copyright, licensed under the Open Government Licence v3.0.
- Statistics Canada. "Public Service Employee Survey, 2024." The Daily, 23 June 2025. Close to 187,000 respondents across 93 departments and agencies, 50.5 percent response rate, fielded 28 October to 31 December 2024. statcan.gc.ca
- Statistics Canada. Public Service Employee Survey 2024 questionnaire, IMDB instrument 1541199, question JOB_Q60, and the Statistics Canada Open Licence. statcan.gc.ca/en/reference/licence
- Gallup. Product Terms of Use, updated 17 October 2025: "The report template and the Q12 items are the copyrighted material of Gallup and cannot be reproduced or used, except within the participating company, without the express permission of Gallup." gallup.com
- Bradler, C., Dur, R., Neckermann, S. and Non, A. "Employee Recognition and Performance: A Field Experiment." Management Science 62(11), 2016, 3085-3099; IZA Discussion Paper 8311, July 2014. iza.org (PDF)
- Stajkovic, A. D. and Luthans, F. "Behavioral management and task performance in organizations: conceptual background, meta-analysis, and test of alternative models." Personnel Psychology 56, 2003, 155-194. 72 studies, 350 effect sizes, total N = 13,301.
Send the twelve and find out which half of recognition is broken
Twelve questions, three minutes, free to copy or open in the editor.
Use this template