Nps Calculator
Net Promoter Score turns a pile of 0-to-10 survey answers into one loyalty number: the percentage of promoters minus the percentage of detractors. This calculator does that arithmetic two ways, from group counts or from pasted raw scores, and adds the two things most free NPS tools leave out: the approximate 95% margin of error computed from your actual response mix, and the full 0-to-10 distribution table so you can see where the score really comes from.
NPS runs from -100 to +100 and is reported as a whole number, not a percentage. The margin of error is an approximation based on your sample size and response mix.
Quick Answer
With 60 promoters, 25 passives and 15 detractors out of 100 responses, your NPS is 45 (60% − 15%), rated Great: healthy loyalty. The approximate 95% margin of error is ±14.5 points, so the true score is likely between 30 and 60. The calculator is pre-filled with these numbers, so run it once to see the full output before entering your own data.
Key takeaways
- NPS is % promoters minus % detractors, on a -100 to +100 scale. Write it as a whole number, never with a percent sign.
- Passives never appear in the formula, but they dilute the score through the denominator. Moving a 7-8 scorer to a 9-10 is usually the highest-leverage improvement.
- Small samples have wide error bands: 100 responses means roughly ±15 points, 400 means roughly ±7. Never celebrate a 5-point quarterly move on 100 responses.
- The score is a number without a diagnosis. Always pair it with the follow-up question asking why the respondent gave that score.
- Quick Answer
- How to use the NPS calculator
- The NPS formula
- Worked example: 60 promoters, 25 passives, 15 detractors
- Bands: a rough guide, not grades
- Why passives matter more than they look
- Compare two periods
- How many more promoters to reach the next band?
- What competing calculators miss
- Frequently asked questions
- Add This Calculator to Your Website
- References
How to use the NPS calculator
Pick your input mode, enter the data, click Calculate NPS. Results stay hidden until you calculate, and the 60/25/15 example is pre-filled so you can verify the full output first.
- Group counts mode: enter how many respondents scored 9-10 (promoters), 7-8 (passives) and 0-6 (detractors). Negative entries are treated as zero, and fractional counts are truncated to whole numbers, so use whole counts.
- Individual scores mode: paste raw 0-to-10 answers separated by commas, spaces, semicolons or line breaks. Anything outside 0-10 is skipped and the tool reports how many entries it ignored. This mode also builds the full 0-to-10 distribution table. Use whole-number scores: a decimal like 8.6 is grouped as a passive but lands in the 9 row of the distribution table, so decimals blur the picture.
- Read the band, then the error: the badge under the score gives this calculator’s rough loyalty band, and the margin-of-error line tells you how much the score could move by chance alone. With tiny samples the interval can stretch past the ends of the -100 to +100 scale; treat any value beyond ±100 as the scale limit, since NPS cannot physically exceed it.
- Reset restores the 60/25/15 defaults and clears the results. It does not change the mode dropdown, so switch modes with the dropdown itself.
Survey costs add up fast, so before fielding a large study it is worth pricing it with the cost per survey calculator. Once you have scores, the churn rate calculator and the net revenue retention calculator connect the loyalty picture to the revenue it predicts.
The NPS formula
NPS = % Promoters − % Detractors. Every respondent counts in the total that the percentages are built on, passives included, which is why passives dilute the score even though they never appear in the subtraction.
Promoters: scores 9-10 | Passives: scores 7-8 | Detractors: scores 0-6
The margin of error uses the standard approach for this metric: the standard error is 100 × √((pp(1−pp) + pd(1−pd) + 2pppd) ÷ n), where pp and pd are the promoter and detractor shares and n is the total responses, multiplied by 1.96 for the 95% interval. This is algebraically identical to the textbook variance method that scores each response as +1, 0 or -1 and takes the sample variance, so it is not a rough shortcut. What it is not: a correction for biased sampling, leading questions, or surveying only your happiest customers.
Worked example: 60 promoters, 25 passives, 15 detractors
These are the pre-filled values, so run the calculator and check every figure below against the tool.
| Measure | Value |
|---|---|
| Total responses | 100 |
| Promoters | 60 (60.0%) |
| Passives | 25 (25.0%) |
| Detractors | 15 (15.0%) |
| NPS | 60.0% − 15.0% = 45 |
| Band | Great: healthy loyalty |
| 95% margin of error | ±14.5 points (true NPS likely between 30 and 60) |
Hand-check the margin of error: pp = 0.6, pd = 0.15, n = 100. The standard error is 100 × √((0.6×0.4 + 0.15×0.85 + 2×0.6×0.15) ÷ 100) = 100 × √(0.5475 ÷ 100) ≈ 7.40, and 1.96 × 7.40 ≈ 14.5. Scale the sample to 10,000 with the same mix and the error collapses to ±1.5; shrink it to 20 responses and it balloons to about ±32, which is why small-sample NPS moves are mostly noise.
Bands: a rough guide, not grades
This calculator labels your score with one of five bands: 70 and above is world-class territory, 50 and above is excellent, 30 and above is great, zero and above means more promoters than detractors, and anything below zero needs improvement. Treat these as this tool’s rough reading, not universal grades.
Two caveats keep the bands honest. First, benchmarks vary enormously by industry: a score of 40 might lead in airlines and lag in software, so always compare against your own history and your sector, never against a universal bar. Second, other sources draw the lines slightly differently: one widely cited industry table puts world-class at 80, excellent at 50, favourable at 20 and good at zero. The differences are judgment calls, not math, and none of them is official. What matters is the direction of your own trend and whether the movement exceeds your margin of error.
Why passives matter more than they look
Passives are often ignored because they do not appear in the formula, but they dilute the score through the denominator. Two companies can have identical promoter and detractor counts yet different scores if one has more passives sitting in the total.
The arithmetic makes the leverage clear. Take 60 promoters and 15 detractors with 25 passives: NPS 45. Convert 10 passives into promoters and the score jumps to 55, a double gain because the promoter percentage rises while the total stays fixed. Now take the same 60 and 15 with zero passives: NPS is already 60. The lesson is practical: follow-up with 7-8 scorers is usually the highest-leverage NPS work a company can do, because those respondents are one nudge away from counting fully. Marketing teams measuring that nudge often pair NPS with efficiency metrics like the marketing efficiency ratio calculator or the affiliate marketing ROI calculator to price what the loyalty is worth.
Compare two periods
Did the score move? Run each survey through the calculator above, subtract the two scores, and compare the bucket shares. Period A of 50 promoters, 35 passives and 15 detractors scores 50% minus 15% = 35 (Great on this calculator bands, since 30 and above is great); Period B of 70 promoters, 20 passives and 10 detractors scores 70% minus 10% = 60 (Excellent, since 50 and above is excellent), an improvement of +25 points driven by promoters moving from 50% to 70% of responses. A move only means something if it exceeds the margin of error: run both periods through the tool and check the movement against each period error band, as the worked example above does for a single period.
How many more promoters to reach the next band?
Start from the current counts in the calculator above, keep your passives and detractors fixed, and solve: with P promoters, A passives, and D detractors, reaching a target score t (as a fraction) needs ceiling of ((D + t x (A + D)) / (1 – t) – P) more promoters. From 60 promoters, 25 passives, and 15 detractors (score 45), reaching 51 needs (15 + 0.51 x 40) / 0.49 – 60 = 13 more promoters. On this calculator bands, Excellent starts at 50 rather than 51, so reaching 50 from the same counts needs (15 + 0.50 x 40) / 0.50 – 60 = 10 more promoters. Each new promoter counts in both the numerator and the total, so late gains move the needle less than early ones, and converting a detractor helps twice, because it shrinks the detractor share and grows the promoter share at once.
What competing calculators miss
Most NPS tools compute the subtraction and stop. The ones that go further each leave something out:
- Simple score tools (Formbricks and similar): clean interface, correct arithmetic, but no margin of error and no distribution view. You get a number with no sense of how much it could move by chance.
- Worst-case error tools (xdage): adds a 95% margin of error, but uses the simplified 1.96 × √(0.25 ÷ n) formula, which assumes maximum variance regardless of your actual mix. For a lopsided response mix it overstates the error; this calculator uses your real promoter and detractor shares instead.
- Full statistics suites (Genroe): the strongest competitor, with margin of error plus a test for whether your score really changed between two periods. The tool itself does not run a two-period comparison in one click: the method above runs both periods through the tool and compares them, and if you need to prove a quarterly move is real, check whether the movement exceeds each quarter’s error band.
- Methodology guides (mTab, Displayr): excellent on the statistics, including the variance derivation and rounding discipline (round once, at the end), but they are articles, not tools you can paste scores into.
- The gap this page fills: both input modes in one tool, the mix-based margin of error that matches the textbook variance method exactly, the full 0-to-10 distribution table from pasted scores, honest bands labelled as a rough guide, and entries outside 0-10 skipped and reported rather than silently absorbed.
Frequently asked questions
Anything above zero means you have more promoters than detractors, which is the first bar to clear. As a rough guide, 30 and above is great, 50 and above is excellent, and 70 and above is world-class territory few companies reach. Benchmarks vary enormously by industry, so compare your score against your own history and your sector rather than a universal bar, and only trust a move that exceeds your margin of error.
No. NPS is reported as a whole number on the -100 to +100 scale, never with a percent sign. The percentages cancel out in the subtraction (60% promoters minus 15% detractors = NPS 45, not 45%), so writing 45% is a common mistake that misstates what the number is. This calculator reports the score as a plain integer.
The Net Promoter framework, developed by Fred Reichheld and Bain and Company, found that 7 and 8 scorers behave differently from true advocates: they are satisfied but unenthusiastic, unlikely to refer you, and vulnerable to competitive offers. So they count in the respondent total, which dilutes the score, but add nothing in the subtraction. That is exactly why converting passives into promoters is the highest-leverage NPS work.
It depends how narrow you need the error band. As a rule of thumb, 50 responses gives roughly plus or minus 22 points, 100 gives about plus or minus 15, 200 about plus or minus 10, and 400 about plus or minus 7. Quadrupling the sample roughly halves the error. This calculator computes the exact band for your mix and sample size, so run your expected numbers before fielding the survey and see whether the precision justifies the cost.
Yes. The scale runs from -100 (every respondent a detractor) to +100 (every respondent a promoter). A negative score means detractors outnumber promoters, and this calculator labels it as needing improvement. Negative scores are common in industries with structural dissatisfaction, which is another reason to benchmark within your sector.
The interval is a normal approximation and the tool does not clip it to the ends of the -100 to +100 scale. With tiny samples the band is very wide, so a result like NPS 67 with plus or minus 53 points can display a range of 14 to 120. Treat anything beyond the scale limit as the limit itself: the true score cannot exceed 100 or fall below -100. Wide bands like that are the tool telling you the sample is too small to trust.
NPS measures loyalty: how likely a customer is to recommend you, which predicts referrals and growth. CSAT (customer satisfaction) measures how satisfied someone was with a specific interaction or purchase. A customer can be satisfied with one support call (high CSAT) while still being unlikely to recommend the company (low NPS). Use NPS for the relationship and CSAT for the transaction.
No. The score is the number this calculator produces. The Net Promoter System is the wider operating model Bain and Company and Fred Reichheld built around it: leadership commitment, one shared metric, closed-loop follow-up with detractors and passives, and learning from promoters. Net Promoter, NPS and Net Promoter Score are registered trademarks of Bain, Satmetrix and Reichheld; this page is an independent explanation and is not affiliated with them.
The formula is identical. Transactional NPS is measured right after a specific interaction (a purchase, a support call), while relational NPS measures overall brand loyalty on a schedule. This calculator computes the same score either way; the difference is when you survey, not how you calculate.
Converting a detractor into a promoter moves the score twice as far as converting a passive, because it shrinks the detractor share and grows the promoter share at once. Follow up with detractors first. Converting passives is the other high-leverage move, since a 7-8 scorer is one nudge away from counting fully, as the passives section above shows.
Add This Calculator to Your Website
Embed this calculator on your own site with the code below. Pick a size to preview it live, then copy the code.
Powered by Calculator Zilo
References
- QuestionPro. Net Promoter System: What It Is and How to Build One (Bain and Company and Fred Reichheld origin). Accessed October 23, 2026.
- Wikipedia. Net Promoter (score computation and methodology overview). Accessed October 23, 2026.
- mTab. How to Calculate Net Promoter Score: Formula, Steps (rounding discipline, sample sizes). Accessed October 23, 2026.
- Displayr. Sample Sizes for Net Promoter Score (variance formula, worked example). Accessed October 23, 2026.
- TLF Research. Understanding Net Promoter Score: Top 5 Questions Answered (rule-of-thumb error bands). Accessed October 23, 2026.
- Reichheld, F. The One Number You Need to Grow, Harvard Business Review, December 2003 (original Net Promoter Score publication; score definition). Accessed October 23, 2026.