Net Promoter Score (NPS) Calculator

Background.

The Net Promoter Score is defined by Bain & Company in one sentence: ask "How likely are you to recommend us to a friend or colleague?", score the answers on a zero-to-ten scale, and "your Net Promoter Score is simply the percentage of customers who are promoters (those who scored 9 or 10) minus the percentage who are detractors (those who scored 0 to 6)". Scores of 7 and 8 are passives. They are counted in the denominator and never in the numerator, which is the single most misunderstood feature of the metric — drop them from your survey base and your NPS jumps without a single customer changing their mind.

That definition has a consequence worth knowing before you read your score. NPS is a difference of two percentages, so many quite different customer bases produce the same number. Forty responses split evenly between promoters and detractors scores 0. Forty responses that are all passives also score 0. The first company has an angry half of its customers and an enthusiastic half; the second has nobody who cares either way. One number, two completely different problems, and no way to tell them apart from the score alone. That is why this calculator always prints the band breakdown next to the result.

The second thing your score hides is how little of it is signal. NPS is a sample statistic and, with the response rates most surveys achieve, usually a small one. Sixty-two promoters, twenty-five passives and thirteen detractors — a clean 100-response quarter — gives an NPS of +49, but the 95% confidence interval around it runs from +33.7 to +61.5. A rival team reporting +38 is not measurably behind you. Collect a hundred times as many responses in the same proportions and the interval tightens to +47.6 to +50.4, which is finally narrow enough to detect a real change. The interval shown here is the adjusted-Wald AW(3,T) method that Brendan Rocks found to perform best for this statistic in The American Statistician; his paper is equally blunt that the plain Wald interval most spreadsheets would produce "should be avoided".

The third thing is the evidence base. Reichheld introduced NPS in Harvard Business Review in December 2003 as "the one number you need to grow", and Bain has built a management system around it. That growth claim has been formally contested: Keiningham, Cooil, Andreassen and Aksoy re-ran the analysis on 21 firms and more than 15,500 interviews from the Norwegian Customer Satisfaction Barometer and reported in the Journal of Marketing that their research "fails to replicate his assertions regarding the 'clear superiority' of Net Promoter compared with other measures in those industries". NPS is a perfectly reasonable, comparable-over-time loyalty summary. Treating it as a proven predictor of revenue growth goes beyond what the published evidence supports, and you should know that before it becomes a board KPI.

Two input modes are provided. Enter group counts when your survey platform has already bucketed people. Paste the raw 0–10 ratings when you have individual answers — the calculator does the bucketing, which also means it catches the ratings that should never have been there, such as an 11 or a 4.5 imported from a five-star widget.

Finally, what no calculator can repair: your score describes the people who answered, not your customers. If unhappy customers churn before the survey reaches them, or a support agent asks for a 10 while standing over the customer, the arithmetic below will be perfect and the number will still be wrong.

What is net promoter score calculator?

Net Promoter Score is a single summary statistic for survey answers to one recommendation question on an 11-point scale. Respondents scoring 9 or 10 are promoters, 7 or 8 are passives, and 0 through 6 are detractors; the score is the promoter percentage minus the detractor percentage. Because it is a difference of two shares of the same base, its arithmetic range is −100 (everyone a detractor) to +100 (everyone a promoter), and it is conventionally written as a whole number without a percent sign — it is a net difference, not a proportion. Rocks (2016) gives the equivalent formulation that is easier to reason about statistically: code each promoter +1, each passive 0 and each detractor −1, and NPS is the mean of those codes, times 100. Net Promoter®, NPS® and Net Promoter System® are registered trademarks of Bain & Company, NICE Systems and Fred Reichheld; Net Promoter Score℠ is a service mark of Bain & Company.

How to use this calculator.

1. 01Choose your input mode. Group counts if your survey tool reports promoters, passives and detractors; raw ratings if you have the individual 0–10 answers.
2. 02In group-count mode, enter all three counts — including passives. Leaving passives out shrinks the denominator and silently inflates the score; it is the most common way an NPS gets overstated.
3. 03In raw-ratings mode, paste the ratings separated by commas, spaces or new lines. Only whole numbers 0 to 10 are accepted, so anything that arrived from a five-star or 1–7 widget will be rejected rather than quietly mis-banded.
4. 04Read the score together with the 95% interval below it. Two periods whose intervals overlap have not been shown to differ, however satisfying the headline movement looks.
5. 05Check the band breakdown line. A score of 0 made of all passives and a score of 0 made of an even promoter/detractor split call for completely different responses.
6. 06Keep the question wording, the scale and the sampling method fixed between periods. Changing any of them makes the comparison meaningless in a way no confidence interval will warn you about.

The formula.

NPS = %promoters − %detractors, where both percentages are taken over all respondents including passives. With the worked example — 62 promoters, 25 passives, 13 detractors from 100 responses — that is 62% − 13% = +49. The equivalent coded form is often clearer: score each promoter +1, each passive 0 and each detractor −1, take the mean, and multiply by 100. Here (62 − 13) ÷ 100 = 0.49, so +49. The passives are not ignored; they contribute zeros, which is what drags the score toward the middle.

The confidence interval uses the adjusted Wald AW(3,T) method that Rocks (2016) identified as the best-performing practical option for this statistic. It adds 0.75 to the promoter count, 0.75 to the detractor count and 1.5 to the passive count, giving an adjusted base of n + 3, then builds an ordinary Wald interval on those adjusted figures. The variance of the score is p_promoters + p_detractors − (p_promoters − p_detractors)², which is exactly the variance of the +1/0/−1 coding described above. For the worked example this gives an interval of +33.69 to +61.45. Two features are deliberate and worth flagging. The interval is centred on the adjusted estimate rather than on your observed score, so it will not be symmetric about the headline number. And because an adjusted-Wald interval can mathematically extend past ±100 when almost everyone lands in one band, the bounds are truncated at −100 and +100, which the score itself can never exceed.

Rounding stage: nothing is rounded on the way through. The shares, the score, the variance and the square root are all computed at full decimal precision and rounded once, at the point results are returned, to ten decimal places. The single exception is the "NPS as usually reported" output, which applies the whole-number reporting convention with halves rounded away from zero — so an exact +0.5 reports as +1 and an exact −0.5 reports as −1. That output is separated from the primary score on purpose: Bain's published definition specifies the bands and the subtraction but not a rounding rule, so the convention is presented as a convention.

A worked example.

Frequently asked questions.

References& sources.

1. [1]Bain & Company, "Measuring Your Net Promoter Score℠" (netpromotersystem.com, current edition, retrieved 2026-07-29). The issuing body's own definition: "To calculate NPS, start with the ultimate question, 'How likely are you to recommend us to a friend or colleague?' and score the answers on a zero-to-ten scale. Your Net Promoter Score is simply the percentage of customers who are promoters (those who scored 9 or 10) minus the percentage who are detractors (those who scored 0 to 6)." Also the source of the band labels Promoters (9 to 10), Passives (7 or 8), Detractors (0 to 6), the rationale for the 11-point scale, the statement that passives' "repurchase and referral rates are as much as 50% lower than those of promoters", and the trademark notice. Publicly accessible.
2. [2]Reichheld, F. F. (2003). "The One Number You Need to Grow." Harvard Business Review, December 2003 issue. The original publication of the metric and of the growth claim attached to it. Author, title and issue date verified on the publisher's page 2026-07-29; the full article text is paywalled, so this is cited as the origin of the metric rather than for any specific figure.
3. [3]Rocks, B. (2016). "Interval Estimation for the 'Net Promoter Score'." The American Statistician, 70(4), 365–372, DOI 10.1080/00031305.2016.1158124; open preprint arXiv:1601.07235, retrieved and text-extracted 2026-07-29. Supplies the variance of the score, σ = p_pro + p_det − (p_pro − p_det)², the finding that "The Wald and Goodman tests perform poorly; their use should be avoided", and the recommended method implemented here: "The best performing Adjusted Wald is AW(3,T) which can be used by adding 3/4 to the counts of both Promoters and Detractors, and 3/2 to the count of Passives, before construction of a Wald interval." Journal version paywalled; preprint open.
4. [4]Keiningham, T. L., Cooil, B., Andreassen, T. W. & Aksoy, L. (2007). "A Longitudinal Examination of Net Promoter and Firm Revenue Growth." Journal of Marketing, 71(3), 39–51, DOI 10.1509/jmkg.71.3.039 (abstract retrieved 2026-07-29). Consulted as an independent authority on the metric's predictive claim rather than its arithmetic. Using 21 firms and more than 15,500 interviews from the Norwegian Customer Satisfaction Barometer, "the research fails to replicate his assertions regarding the 'clear superiority' of Net Promoter compared with other measures in those industries." Abstract public; full text paywalled.
5. [5]Agresti, A. & Coull, B. A. (1998). "Approximate Is Better Than 'Exact' for Interval Estimation of Binomial Proportions." The American Statistician, 52(2), 119–126, DOI 10.1080/00031305.1998.10480550. The origin of the adjusted-Wald construction that Rocks (2016) adapts to the trinomial Net Promoter case; cited here because the AW(3,T) interval on this page is a direct descendant of it. Bibliographic reference — the publisher's copy is paywalled and was not fetched; the method as implemented is taken verbatim from Rocks §2.1.3 and §4.