eNPS Survey Template
An 8-question eNPS survey built on one 0 to 10 question. The result is a number between -100 and +100, not a percentage. Take it here, work out your own figure in the calculator, or open it in the editor.
- 8questions
- About 2 minto complete
- 0 to 10one scored question
- -100 to +100the possible range
How eNPS is worked out, and why the answer is not a percentage
One question makes the number. Three bands turn it into two percentages, and one subtraction turns those into a figure.
The question. "How likely are you to recommend this organization as a place to work?", answered from 0 to 10. Ask it in words close to those, or next quarter is not a comparison.
The bands, and why the answer is not a percentage. A 9 or a 10 is a promoter, a 7 or an 8 is a passive, and 0 to 6 is a detractor. Subtract the detractor share from the promoter share. One percentage taken from another gives a difference, not a proportion, so a result of 27 is not 27 percent of anything.[4] It runs from -100 to +100.
Check which recommend question you inherited. Some tools ask employees about the products and services instead of the workplace and call the result by the same name.[13] Find out which one produced any figure you were handed.
All eight questions are built into the free survey creator, so you can change the wording before anyone sees it. The marks are not ours and we hold no licence; the attribution is in the methods block at the foot of the page.
If what you want is the reason the figure moved, this is the wrong survey. The employee engagement survey asks about the manager, growth, recognition, workload and fairness, and says which of them changed.
How to calculate eNPS from your own responses
Count how many people gave each answer and let the box do the subtraction.
The same sum by hand
Say 40 people answered: 18 gave a 9 or a 10, 8 gave something from 0 to 6, and the other 14 sat in the middle. Promoters are 45 percent, detractors are 20 percent, and your eNPS is +25.
Two ways people get it wrong. Averaging the 0 to 10 answers gives a completely different number, usually around 7. Leaving passives out of the total inflates the result, and on these same 40 answers it would report +38 instead of +25. Write the figure down with a plus or minus sign in front of it and no percent sign after it.
What one answer is worth at your headcount
The fewer people reply, the further a single answer moves the result. This part is arithmetic, so your own figure is exact.
The rule. With n responses, one person moving from a 6 to a 9 changes the result by exactly 200 divided by n. They leave the detractor share, worth 100 divided by n, and join the promoter share, worth the same again.
| Responses | One person moving from a 6 to a 9 moves your result by |
|---|---|
| 15 | 13.3 points |
| 25 | 8.0 points |
| 50 | 4.0 points |
| 100 | 2.0 points |
| 250 | 0.8 points |
Why that matters. Hive HR described a 10-point move in one sector as one of the biggest shifts in its published figures.[6] At 25 responses one person is worth 8 points, so a 10-point move is one and a bit people changing their minds.
So: below fifty responses one person is worth four points or more. Publish the three band counts and leave the single figure out. Above fifty, ignore any change smaller than your own row in the table.
Two teams can be far apart and land on the same result. A team that all answers 6 and a team that all answers 1 both come out at -100, because 1 and 6 are both detractor answers.[3][12] The bands throw away everything inside them, which is what the open boxes recover.
The 8 eNPS survey questions
The recommend item first, because it is the number. Then the reason in two boxes and again in one click, and a closing check on whether people could answer freely.
The recommend question
Asked first, so somebody who gives up after ten seconds still leaves you a usable number.
- How likely are you to recommend this organization as a place to work?A 7 or an 8 counts for nothing and a 6 counts against you, so say what the bands do when you send it.
Two boxes for the reason
An anonymous survey has no follow-up call, so whatever people type here is the only follow-up you get.
- What is behind the number you chose?Expect one line, not a paragraph. Give each answer a single theme and count the themes.
- What would have to change for you to give a higher number?A blank here from somebody who scored you 3 is the worst answer on the page. They have stopped expecting to be asked.
The same reason, in one click
Written answers are thin when twenty people reply. This asks the same thing in a form you can still count at that size.
- Which of these is closest to the reason for your number?If most people pick how the place is run, no amount of manager training will move the figure.
One cut, and only one
People in their first year answer this differently from people in their fifth, and it is the only cut this survey makes.
- How long have you worked here?Tenure plus a small team identifies somebody. Drop this before you drop the promise you made about anonymity.
Three questions that tell you whether to believe the figure
These say how much weight the number can carry, and they close the survey on a scale rather than a text box.
- When people outside ask about working here, I speak well of it.Stated willingness and actual behaviour differ. A gap between this and question 1 is the interesting part.
- I expect something to change because of this survey.Low here means people are answering about the last survey rather than about the job.
- How freely could you answer these questions?If people say they could not answer freely, your result is a floor rather than a measurement.
Where the wording comes from. The recommend question is the standard one in plain English.[1] One item is adapted from the public-domain Federal Employee Viewpoint Survey; the rest were written for this page.
What is deliberately missing. No name and no email field: a survey that asks who you are is not anonymous, whatever the invitation says. No questions about pay, the manager or training either, because asking those properly takes twenty questions rather than eight.
What a good eNPS is, and who is telling you
Three organizations publish an overall figure. None of them measured you.
| Who publishes it | What they measured | Their overall figure | Why it is not your benchmark |
|---|---|---|---|
| Hive HR | Employee answers collected through one vendor, from UK organizations, over 500,000 responses | 27 in Q3 2024[5] | UK only, one vendor and its own clients, and its sector figures moved about 10 points between October and March[6] |
| QuestionPro | A panel of 5,000 individual employees in full-time jobs, December 2024 to January 2025 | 32 in 2025, 25 in 2024[7] | A pool of individuals from many employers rather than a set of organization results, with no sample size given per industry |
| Perceptyx | Its own client database, with no window and no method stated | Low to mid 20s[8] | The page gives its own sample size two different ways and in two different units, so the figure cannot be placed |
The band tables disagree as well. A result of +5 is good to Matter, typical to Hive HR, and room for improvement to HiBob.[4][6][12] Leapsome and SurveyMonkey leave +5 out altogether.[14][9] None of the five publishes a source for its boundaries.
Compare against your own last result, on the same wording and the same people. Above zero means more promoters than detractors, and that is most of what a single figure can honestly say. A systematic review found the research does not support the measure for broad benchmarking, and the fullest review of the academic work puts it level with a satisfaction score rather than ahead of it.[11][10]
Never put it beside your customer number. Bain warns that employee results run substantially lower, because people hold their own employer to a higher standard.[2] If the customer side is what you are measuring, the NPS survey template is the one to send.
Qualtrics, which sells employee surveys, does not recommend the metric for employee programmes at all.[3] Advocacy is not contentment either, which is the ground the employee satisfaction survey covers. If the figure falls and nobody can say why, the employee morale survey reads the mood directly.
How to run an eNPS survey without fooling yourself
Three decisions that only this metric forces on you. Everything a survey of any kind needs is in the survey research guide.
Tell people what a 7 does
Only the two ends of the scale carry labels. Say in the invitation that 7 and 8 count for nothing, or people will guess differently.
Pick a frequency and hold it
Publishers put this anywhere between monthly and twice a year. Pick one and keep it; a shorter rhythm is what the employee pulse survey is for.
Sort the written answers by band first
The figure tells you the weather; the sentences tell you what to do on Monday.
eNPS survey FAQ
What does eNPS stand for?
Employee Net Promoter Score. It is the workplace version of a measure built for customers, and the marks belong to Bain and Company, NICE Systems and Fred Reichheld rather than to any survey tool.
What is a good eNPS?
Anything above zero means more promoters than detractors. The published overall figures land between about 20 and 32, but each comes from a different population measured a different way, so your own previous result is the comparison worth making.
Are the answers anonymous?
That depends on how you run it rather than on the survey itself. Whoever owns the survey can read the answers. What buys you anonymity is putting it with somebody outside the reporting line, asking for no identifying details, and refusing to report any group small enough to name.
One number tells you the score moved and nothing about why. Where recognition is the suspected cause, the employee appreciation survey asks the questions the score cannot.
Methods and sources
What this template is based on
Built on the trademark holder's own published guidance: keep it short, ask the recommend item first, and add a second scaled question because one item predicts less.[2] One item is adapted from the public-domain Federal Employee Viewpoint Survey.
Question licensing
Net Promoter(R), NPS(R), NPS Prism(R), and the NPS-related emoticons are registered trademarks of Bain & Company, Inc., NICE Systems, Inc., and Fred Reichheld. Net Promoter Score(SM) and Net Promoter System(SM) are service marks of Bain & Company, Inc., NICE Systems, Inc., and Fred Reichheld.[1] SuperSurvey holds no licence, certification, endorsement or affiliation; the question wording is plain English and free to use, and nothing here reproduces a licensed instrument.
Limits and disclosure
None of the published benchmarks is a norm for your organization, and the band tables in circulation cite no source. None of this is legal advice.
SuperSurvey makes the editor this template opens in; everything here is free with or without it.
References
- Bain and Company. Trademarks and licensing. Net Promoter System. netpromotersystem.com
- Bain and Company. The Employee Net Promoter System. Net Promoter System. netpromotersystem.com
- Herbert, C. "Seven Limitations of eNPS as an Employee Experience Metric." Qualtrics, 20 December 2021. qualtrics.com
- Matter. Employee Net Promoter Score (eNPS). Matter help centre. help.matterapp.com
- Hive HR. The Employee Engagement Benchmarks: Q3 2024. Hive, 2024. hive.hr
- Hive HR. The Employee Engagement Benchmarks: Q1 2026. Hive, 2026. hive.hr
- QuestionPro. eNPS Industry Benchmarks 2025 and Year on Year Trends. QuestionPro Employee Experience, 2025. questionpro.com (PDF)
- Perceptyx. "Employee Net Promoter Score." Perceptyx blog. perceptyx.com
- SurveyMonkey. "Employee NPS benchmarks." SurveyMonkey. surveymonkey.com
- Sauro, J. Review of the Net Promoter Score research literature. MeasuringU. measuringu.com
- Adams, C., Walpola, R., Schembri, A.M. and Harrison, R. "The ultimate question? Evaluating the use of Net Promoter Score in healthcare: a systematic review." Health Expectations 25 (2022): 2328-2339. ncbi.nlm.nih.gov
- HiBob. Employee net promoter score (eNPS). HiBob HR glossary. hibob.com
- Qualtrics. Employee Net Promoter Score (eNPS), employee experience article. qualtrics.com
- Leapsome. "Employee net promoter score." Leapsome blog. leapsome.com
Send the eNPS survey this week
Eight questions, about two minutes, no name and no email field. Open it in the editor, change the wording, and share one link.
Use this template