Net promoter score calculator for survey results
This NPS calculator works out the net promoter score from a survey that asks “How likely are you to recommend us to a friend or colleague?” on a 0 to 10 scale. Either type the number of promoters, passives and detractors, or paste the raw scores straight from your survey export and let the tool sort them.
The result is the score from −100 to +100, the percentage in each group, and a margin of error, so you can tell a real change from noise. The calculation runs in your browser and nothing is uploaded.
How to use the Net Promoter Score Calculator
- Choose Counts if you already have totals, or Paste scores to paste raw 0–10 answers.
- Enter the three counts or paste the scores, separated by spaces, commas or new lines.
- Pick the confidence level for the margin of error (95% is the usual choice).
- Read the score, the split and the likely range. A warning appears when there are fewer than 30 responses.
Net Promoter Score Calculator features
Raw scores or counts
Paste a column of answers or type totals; both give the same result.
Group split
Promoters, passives and detractors as percentages and as counts.
Margin of error
A range around the score, based on the variance of the three possible outcomes.
Input checks
Scores outside 0–10, text and empty input are flagged instead of being counted silently.
The NPS formula
Respondents are grouped by their answer: promoters score 9 or 10, passives score 7 or 8 and detractors score 0 to 6. The score is NPS = % promoters − % detractors. Passives count in the total but not in the difference.
Example: of 200 responses, 120 are promoters (60%), 60 passives (30%) and 20 detractors (10%). NPS = 60 − 10 = +50. It is written as a whole number, not a percentage.
The margin of error uses the fact that each answer is worth +1, 0 or −1. The variance of the score is (pP + pD − (pP − pD)²) ÷ n. For the example that is 0.00225, a standard error of 4.7 points and a 95% margin of ±9.3 points. With only 20 responses the same percentages would give a margin near ±29.
Reading the result
Compare scores only when the survey wording, timing and audience match. A change smaller than the margin of error is not evidence that customers feel differently. To plan how many responses you need, use the sample size calculator; to see how retention changes alongside, try the churn rate calculator. This tool shows no industry benchmarks, because published averages differ widely by method and year.
| Group | Score | Counted in NPS as |
|---|---|---|
| Promoters | 9–10 | +1 |
| Passives | 7–8 | 0 |
| Detractors | 0–6 | −1 |
Limits and trademark note
- NPS summarises one question. It does not explain why people scored as they did; read the comments too.
- Response bias matters: if unhappy customers rarely answer, the score will look better than reality.
- The margin of error assumes a random sample of respondents and does not correct for non-response.
- Net Promoter, NPS and Net Promoter Score are trademarks of Bain & Company, Satmetrix Systems (NICE) and Fred Reichheld. This tool is an independent calculator and is not affiliated with or endorsed by them.
Net Promoter Score Calculator FAQ
How do you calculate NPS?
Percentage of promoters (scores 9–10) minus percentage of detractors (scores 0–6). Passives (7–8) are ignored in the subtraction. 60% promoters and 10% detractors gives +50.
What is a good NPS?
Any positive score means more promoters than detractors, but what counts as good depends on your industry and method. Track your own score over time and judge changes against the margin of error.
Can NPS be negative?
Yes. The range is −100 (all detractors) to +100 (all promoters). A negative score means detractors outnumber promoters.
Is NPS a percentage?
No. It is the difference between two percentages, written as a number from −100 to +100 without a percent sign.
How many responses do I need?
More is better. With about 100 responses the 95% margin is roughly ±13 to ±15 points, depending on the split. Fewer than 30 gives only a rough idea.
Should I include blank answers?
No. Only count respondents who gave a score from 0 to 10; leave out non-responses.