NPS Formula: How Net Promoter Score Math Works
- Subtract the percentage of detractors (0 to 6) from the percentage of promoters (9 to 10) to get your NPS.
- Count passives (7 to 8) in the total number of responses, even though they do not appear in the subtraction.
- Report NPS as a number from -100 to +100, not as a percentage.
- Multiply by 100 only when you start from raw counts, never when you start from percentages.
- Weight segment scores by their number of responses before you combine them into one NPS.
- Use the 0 to 10 scale every time, because a 1 to 10 or 5-point version is not comparable with NPS benchmarks.
The NPS formula is the percentage of promoters minus the percentage of detractors: NPS = % promoters - % detractors. Promoters answer 9 or 10 to the question "How likely are you to recommend us to a friend or colleague?", detractors answer 0 to 6, and the result is a number between -100 and +100.
Fred Reichheld introduced the metric in The One Number You Need to Grow [Harvard Business Review, 2003] and later expanded it, with Rob Markey, into the book The Ultimate Question. The formula itself is short, but small slips in it change the score a lot. This article covers the formula, the score range, one worked example and the math errors to avoid. For the full step-by-step method, from collecting answers to segmenting them, read how to calculate NPS. For everything else about the metric, start with our Net Promoter Score guide.
What is the Net Promoter Score formula?
The Net Promoter Score (NPS) is built on one question, asked on an 11-point scale from 0 ("not at all likely") to 10 ("extremely likely"). Every answer falls into one of three groups. Reichheld named them after how the customers behaved in his research: promoters reported the highest repurchase rates and made more than 80% of referrals, while detractors produced more than 80% of negative word-of-mouth comments [Bain & Company, 2006].
- Promoters (9 to 10): enthusiastic customers who are likely to buy again and recommend you.
- Passives (7 to 8): satisfied but unattached customers who could switch for a better offer.
- Detractors (0 to 6): unhappy customers who are at risk of leaving and of warning others.
The formula in words: NPS = (promoters ÷ total responses × 100) - (detractors ÷ total responses × 100). Each group and how to treat it is explained in promoters, passives and detractors in NPS.
NPS score range: why it runs from -100 to +100
The two ends of the scale are easy to picture. If all 200 people who answered your survey score 0 to 6, the share of promoters is 0% and the share of detractors is 100%, so NPS = 0 - 100 = -100. If all 200 score 9 or 10, NPS = 100 - 0 = +100. A score of 0 means you have as many promoters as detractors, whatever the share of passives.
That is also why NPS is not a percentage. A score of 35 does not mean that 35% of customers did anything. It means promoters outnumber detractors by 35 percentage points. Write it as "NPS 35" or "+35", and keep the percent sign for the two inputs.
Passives sit in the middle and shape the score only through the total. You can find their share by subtraction: % passives = 100 - % promoters - % detractors.
NPS formula example: a worked calculation
| Group | Scores | Responses | Share of total |
|---|---|---|---|
| Promoters | 9 to 10 | 110 | 110 ÷ 200 = 55% |
| Passives | 7 to 8 | 50 | 50 ÷ 200 = 25% |
| Detractors | 0 to 6 | 40 | 40 ÷ 200 = 20% |
| NPS | 200 | 55 - 20 = 35 |
You get the same answer from raw counts, which is how a spreadsheet usually does it: NPS = (promoters - detractors) ÷ total responses × 100 = (110 - 40) ÷ 200 × 100 = 35. Both versions are correct. Problems start when people mix them.
Common NPS formula mistakes and how to fix them
| Mistake | What it looks like | Fix |
|---|---|---|
| A stray × 100 | (55 - 20) × 100 = 3,500, or 0.35 shown as the score | Multiply by 100 only when you start from counts or decimals, never from percentages |
| Passives left out of the total | (110 - 40) ÷ 150 × 100 = 46.7 instead of 35 | Divide by every response, passives included |
| A 1 to 10 scale | The 0 option is missing, so the scale has 10 points, not 11 | Always offer 0 to 10, so your score is comparable with benchmarks |
| Wrong cut-offs | 8 counted as a promoter, or 6 as a passive | Promoters 9 to 10, passives 7 to 8, detractors 0 to 6 |
| Averaging the raw scores | "Average score 8" reported as the NPS | Group the answers first, then subtract the shares |
| Averaging segment scores | Two segment scores averaged with equal weight | Weight each segment by its number of responses |
The average-score trap is easy to show. Ten customers who all score 8 have an average of 8 and an NPS of 0, because all of them are passives. Six customers who score 10 and four who score 5 also average 8, but their NPS is 60 - 40 = 20. Same average, different loyalty.
The weighting error is the one we see most often in real reports, because it hides inside a dashboard that looks correct. A store sees an NPS of 60 among 100 repeat buyers and 10 among 900 first-time buyers, averages the two segment scores and reports 35. The true NPS across all 1,000 answers is (100 × 60 + 900 × 10) ÷ 1,000 = 15. The team celebrates a score more than twice as high as the real one, and the first-purchase experience, where most customers sit, gets no attention. Recalculate from the pooled counts, or weight each segment by its responses.
How to read your NPS result
A single number tells you the balance between promoters and detractors, not the reason for it. Follow the 0 to 10 question with an open one, such as "What is the main reason for your score?". Code each verbatim answer into a theme, such as product quality, delivery or customer service, and count the themes for each group. If most detractor comments are about delivery, the fix sits with your shipping partners, not with your website.
Then look past the average. Split the score by pre-delivery and post-delivery surveys, location, brand, category and customer segment; the customer segmentation tools you already use can usually supply the segments. Omniconvert Commander (formerly Nexus) connects each NPS answer to the customer's RFM segment, so you can see whether your most valuable customers are among your detractors. For a step-by-step reading method, see NPS analysis, and to judge the number against others, see what a good NPS score is.
Survey timing alone can move the result. In an Omniconvert project with a large online marketplace, the pre-delivery NPS was 71 and the post-delivery NPS was 82 in the first month. The gap pointed to the website, not the products: open answers named search, filters, payment and site speed. Four months after the fixes, the pre-delivery score reached 78 and the post-delivery score 84 [Omniconvert, marketplace NPS case study]. The full story is in our NPS case study.
NPS formula FAQs
The Net Promoter Score formula is NPS = % promoters - % detractors. Promoters answer 9 or 10 to the question "How likely are you to recommend us to a friend or colleague?", detractors answer 0 to 6, and passives (7 or 8) count in the total but not in the subtraction.
NPS is the difference between two percentages, so the result is a plain number on a scale from -100 to +100. If 55% of respondents are promoters and 20% are detractors, the NPS is 35, not 35%. Writing it as a percentage suggests that 35% of customers did something, which is not what the score means.
The NPS score range runs from -100, when every respondent is a detractor, to +100, when every respondent is a promoter. A score of 0 means promoters and detractors are equal in number. Any score above 0 means you have more promoters than detractors.
Subtract the number of detractors from the number of promoters, divide by the total number of responses, including passives, and multiply by 100. With 110 promoters, 50 passives and 40 detractors, NPS = (110 - 40) / 200 x 100 = 35. Multiply by 100 only when you start from counts, never when you start from percentages.
Not as a true NPS. The Net Promoter Score is defined on an 11-point 0 to 10 scale, and the 9-10, 7-8 and 0-6 cut-offs only work on that scale. You can apply the same promoters-minus-detractors logic to a 5-point question, but the result is a different metric and you cannot compare it with NPS benchmarks.
There is no single good score. A positive NPS means you have more promoters than detractors. Judge your score against your own past results, your industry and your direct competitors, and check the score inside your most valuable customer segments before you celebrate the average.
Before your next NPS report goes out, rebuild the number once by hand from raw counts: promoters minus detractors, divided by every response including passives, times 100. If your dashboard and your hand calculation disagree, one of the six errors above is in the formula.
Then break the score down by segment and weight each segment by its responses. A correct NPS is only the start: the follow-up answers from your detractors tell you what to test next, and Omniconvert Explore runs the NPS survey and the A/B test in the same tool.
Measure NPS where the experience happens
Omniconvert Explore runs NPS and on-site surveys, A/B tests and personalization in one tool, so you can collect the score, read the reasons behind it and test the fix.