What NPS (Net Promoter Score) Is: Definition, Formula & How to Improve It
- NPS (Net Promoter Score) measures customer loyalty with one question, how likely are you to recommend us on a 0 to 10 scale, sorting customers into Promoters (9-10), Passives (7-8), and Detractors (0-6).
- The formula is % Promoters minus % Detractors (Passives count toward the total but are not added or subtracted), giving a score from minus 100 to plus 100; e.g. 60% minus 15% = 45.
- There is no universal good score, read NPS against your own trend over time and across your own segments, not a generic global benchmark; treat any external number as loose context.
- The number alone does not explain itself, the open-ended follow-up (why did you give that score?) is what makes NPS actionable; it also measures stated intent, not actual behaviour.
- Improve NPS by acting on causes, close the loop with Detractors and win over Passives; Nexus shows NPS by segment and ties it to retention, repeat-purchase rate, and CLV.
One question, one number, and a loyalty metric used across almost every industry: that is the appeal of NPS. Ask customers how likely they are to recommend you, subtract the unhappy from the enthusiastic, and you have a single score you can track over time. But the number is only as useful as what you do with it, and the most valuable part of an NPS survey is not the score at all, it is the follow-up question that asks why. This guide explains what NPS is, how it is calculated, how to read it, its limits, how to improve it, and how Nexus turns it into an actionable loyalty signal, drawing on 13 years of data across 7,000+ websites in 15+ industries [CROBenchmark Report 2026, Omniconvert].
One idea runs through all of it: NPS is a pulse, not a diagnosis. It tells you the balance of loyalty at a glance, but you act on the reasons behind it, not the number.
What NPS is
NPS, or Net Promoter Score, is a widely used measure of customer loyalty based on a single question: how likely are you to recommend this company, product, or service to a friend or colleague? Customers answer on a scale from 0 to 10, and their answers sort them into three groups. Promoters score 9 or 10 and are loyal enthusiasts likely to keep buying and refer others. Passives score 7 or 8 and are satisfied but unenthusiastic, easily tempted away by a competitor. Detractors score 0 to 6 and are unhappy customers who can damage your reputation through negative word of mouth.
The score itself is the percentage of Promoters minus the percentage of Detractors, which produces a single number from minus 100 to plus 100. Because it comes from one simple question, NPS is quick to collect and easy to track over time, which is why so many companies use it as a headline loyalty metric. Its real value comes not from the number alone but from the follow-up question that asks customers why they gave the score they did. To see where the number comes from, it helps to walk through the calculation.
How NPS is calculated
NPS is calculated as the percentage of Promoters minus the percentage of Detractors, ignoring the Passives in the subtraction. The three groups are scored like this:
| Group | Score | What they are | How to act |
|---|---|---|---|
| Promoters | 9–10 | Loyal enthusiasts who keep buying and refer others | Nurture and encourage referrals |
| Passives | 7–8 | Satisfied but unenthusiastic, easily tempted away | Find the one fix that wins them over |
| Detractors | 0–6 | Unhappy customers, a source of negative word of mouth | Close the loop; fix the problem and show it |
To calculate the score, convert the Promoter and Detractor counts into percentages of all respondents, then subtract. For example, if 100 people respond, 60 are Promoters, 25 are Passives, and 15 are Detractors, then Promoters are 60% and Detractors are 15%, so the NPS is 60 − 15, which equals 45. The result is a whole number, not a percentage, and ranges from minus 100 to plus 100. Note that Passives count toward the total respondents, which lowers both percentages, but they are not added or subtracted directly, so a large satisfied-but-unenthusiastic middle can quietly hold your score down. Once you have a number, the next question is how to read it.
What is a good NPS score?
There is no single number that counts as a universally good NPS, because scores vary widely by industry, region, and how the survey is run, so a score that is strong in one sector can be ordinary in another. As a rough orientation, any positive score means you have more Promoters than Detractors, a score above a broad benchmark such as 50 is generally considered excellent, and a negative score is a warning sign that unhappy customers outnumber loyal ones. But comparing your number against a generic global benchmark is the least useful way to read it.
The most reliable reading comes from two comparisons. First, compare your NPS to your own past scores: is the trend rising or falling over time? A steady climb is a better sign than any single reading. Second, compare across your own segments: which customer groups, products, or regions have high scores and which drag the average down? That is where NPS becomes actionable. When you measure the score also shapes what it tells you, which is where the two main types of NPS come in.
Relational vs transactional NPS
Relational NPS and transactional NPS use the same question and scale but ask it at different moments and answer different questions. Relational NPS is measured at regular intervals, such as quarterly or annually, and asks about the customer's overall relationship with your brand. It gives you a high-level, stable read of loyalty over time and is good for tracking the big trend. Transactional NPS is measured right after a specific interaction, such as a purchase, a delivery, or a support ticket, and asks about that experience. It is more tactical, pinpointing which moments in the journey create Promoters and which create Detractors.
The two work best together: relational NPS tells you whether loyalty is improving overall, while transactional NPS tells you which touchpoints are driving the change, so you know where to act. A common approach is to run relational surveys on a fixed schedule for the trend and layer transactional surveys on key journey moments to diagnose the causes behind it. Whichever you run, it helps to be clear-eyed about what the metric can and cannot do.
The limits of NPS, and how to improve it
NPS is popular because it is simple, but that simplicity is also its main limit. A single number tells you the balance of loyalty, but on its own it does not tell you why customers feel that way or what to do about it, which is why the open-ended follow-up question is essential. It can also mislead if you treat it as precise: small samples produce noisy scores, and because Passives are excluded from the subtraction, two companies with very different customer mixes can share the same headline number. It is prone to non-response bias, and it measures stated intent to recommend, not actual behaviour, so a high score does not guarantee real referrals or repeat purchases.
None of this means NPS is worthless, it is a useful, easy-to-track pulse, but you improve it by acting on the reasons behind the score:
-
Read the open-ended answersThe comments beside the scores tell you what is creating Detractors and what is delighting Promoters.
-
Close the loop with DetractorsFollow up, fix their specific problem, and show them it was addressed; this turns some around and stops negative word of mouth.
-
Win over the PassivesFind the one improvement that moves the large satisfied-but-unenthusiastic middle toward promoting you.
-
Segment, then targetUse segmentation to see which groups score low and fix the causes there, not with a blanket change.
-
Act, then re-measureFeed what you learn back into the product and experience, then measure again, so improvement becomes a loop.
Because the score reflects real customer feeling, the durable way to raise it is to remove the friction and disappointments that create Detractors in the first place, which is far easier when you can see the score by segment.
NPS with Nexus
Nexus is a customer value optimization platform that helps you move NPS from a headline number to an actionable loyalty signal by connecting it to the rest of your customer data. Instead of a single company-wide score, Nexus lets you see NPS by customer segment, so you can tell which groups are Promoters and which are Detractors, and read the score alongside the metrics that reflect real behaviour: retention, retention rate, and customer lifetime value. That matters because stated intent to recommend is most useful when checked against what customers actually do.
By grouping customers with RFM segmentation and tying feedback to value, Nexus helps you focus improvement where it pays off most, on the high-value segments whose loyalty drives revenue, and close the loop with the segments most at risk of churning. Drawing on 13 years of data across 7,000+ websites and 15+ industries, with 248+ audit criteria, Nexus helps turn the NPS pulse into decisions that protect and grow customer lifetime value.
Want to see which customer segments are Promoters and which are quietly churning?
See how Nexus turns NPS into action →Frequently Asked Questions
NPS, or Net Promoter Score, is a widely used measure of customer loyalty based on a single question: how likely are you to recommend this company, product, or service to a friend or colleague? Customers answer on a scale from 0 to 10, and their answers sort them into three groups. Promoters score 9 or 10 and are loyal enthusiasts likely to keep buying and refer others. Passives score 7 or 8 and are satisfied but unenthusiastic, easily tempted away by a competitor. Detractors score 0 to 6 and are unhappy customers who can damage your reputation through negative word of mouth. The score itself is the percentage of Promoters minus the percentage of Detractors, which produces a single number from minus 100 to plus 100. Because it comes from one simple question, NPS is quick to collect and easy to track over time, which is why so many companies use it as a headline loyalty metric. Its real value comes not from the number alone but from the follow-up question that asks customers why they gave the score they did.
NPS is calculated as the percentage of Promoters minus the percentage of Detractors, ignoring the Passives in the subtraction. First, survey customers with the recommend question and sort each response into a group: Promoters (9 to 10), Passives (7 to 8), and Detractors (0 to 6). Next, convert the Promoter and Detractor counts into percentages of all respondents. Then subtract: NPS equals the percentage of Promoters minus the percentage of Detractors. For example, if 100 people respond, 60 are Promoters, 25 are Passives, and 15 are Detractors, then Promoters are 60% and Detractors are 15%, so the NPS is 60 minus 15, which equals 45. The result is expressed as a whole number, not a percentage, and ranges from minus 100 (every respondent is a Detractor) to plus 100 (every respondent is a Promoter). Passives count toward the total number of respondents, which lowers both percentages, but they are not added or subtracted directly. This is why a large group of merely satisfied Passives can hold your score down even when you have few outright Detractors.
There is no single number that counts as a universally good NPS, because scores vary widely by industry, region, and how the survey is run, so a score that is strong in one sector can be ordinary in another. As a rough orientation, any positive score means you have more Promoters than Detractors, a score above a broad benchmark such as 50 is generally considered excellent, and a negative score is a warning sign that unhappy customers outnumber loyal ones. But comparing your number against a generic global benchmark is the least useful way to read it. The most reliable reading comes from two comparisons. First, compare your NPS to your own past scores: is the trend rising or falling over time? A steady climb is a better sign than any single reading. Second, compare across your own segments: which customer groups, products, or regions have high scores and which drag the average down? That is where NPS becomes actionable. Treat any external benchmark as loose context, not a target, and focus on your own trend and your own segments.
Relational NPS and transactional NPS use the same question and scale but ask it at different moments and answer different questions. Relational NPS is measured at regular intervals, such as quarterly or annually, and asks about the customer's overall relationship with your brand. It gives you a high-level, stable read of loyalty over time and is good for tracking the big trend. Transactional NPS is measured right after a specific interaction, such as a purchase, a delivery, or a support ticket, and asks about that experience. It is more tactical, pinpointing which moments in the journey create Promoters and which create Detractors. The two work best together: relational NPS tells you whether loyalty is improving overall, while transactional NPS tells you which touchpoints are driving the change, so you know where to act. A common approach is to run relational surveys on a fixed schedule for the trend and layer transactional surveys on key journey moments to diagnose the causes behind it.
NPS is popular because it is simple, but that simplicity is also its main limit. A single number tells you the balance of loyalty, but on its own it does not tell you why customers feel that way or what to do about it, which is why the open-ended follow-up question that asks for the reason behind the score is essential. NPS can also mislead if you treat it as precise: small samples produce noisy scores, and because Passives are excluded from the subtraction, two companies with very different customer mixes can share the same headline number. It is prone to non-response bias, since the customers who bother to answer may not represent the silent majority. And it measures stated intent to recommend, not actual behaviour, so a high score does not guarantee real referrals or repeat purchases. None of this means NPS is worthless, it is a useful, easy-to-track pulse of loyalty, but it works best as one signal among several, read alongside retention, repeat-purchase rate, and lifetime value, and always paired with the qualitative why.
You improve NPS by acting on the reasons behind the score, not by chasing the number itself. Start by reading the open-ended answers, because the comments that accompany the scores tell you what is creating Detractors and what is delighting Promoters. Close the loop with Detractors: follow up with unhappy customers, fix their specific problem, and show them it was addressed, which can turn some of them around and stops negative word of mouth. Look at the Passives, the large satisfied-but-unenthusiastic middle, and find the one improvement that would move them toward promoting you, since converting Passives often lifts the score more than chasing the already-loyal. Use segmentation to see which groups score low and target the fixes there rather than applying a blanket change. And feed what you learn back into the product and experience, then measure again, so improving NPS becomes a loop: listen, act, close the loop, and re-measure. Because the score reflects real customer feeling, the durable way to raise it is to remove the friction and disappointments that create Detractors in the first place.
Nexus is a customer value optimization platform that helps you move NPS from a headline number to an actionable loyalty signal by connecting it to the rest of your customer data. Instead of a single company-wide score, Nexus lets you see NPS by customer segment, so you can tell which groups are Promoters and which are Detractors, and read the score alongside the metrics that reflect real behaviour, retention, retention rate, and customer lifetime value. That matters because stated intent to recommend is most useful when checked against what customers actually do. By grouping customers with RFM segmentation and tying feedback to value, Nexus helps you focus improvement where it pays off most, on the high-value segments whose loyalty drives revenue, and close the loop with the segments most at risk of churning. Drawing on 13 years of data across 7,000+ websites and 15+ industries, with 248+ audit criteria, Nexus helps turn the NPS pulse into decisions that protect and grow customer lifetime value.
NPS distils customer loyalty into one number by asking a single question, how likely are you to recommend us, and subtracting the percentage of Detractors (0 to 6) from the percentage of Promoters (9 to 10), which produces a score from minus 100 to plus 100. Its strengths are its strengths and its weaknesses at once: it is quick to collect and easy to track, but a lone number cannot tell you why customers feel as they do, which is why the open-ended follow-up is the part that actually drives improvement. Read it well by comparing against your own trend over time and across your own segments, not against a generic global benchmark, and treat it as one signal among several, checked against retention, repeat-purchase rate, and lifetime value rather than trusted in isolation. The durable way to raise it is not to chase the score but to act on its causes: close the loop with Detractors, win over the satisfied-but-unenthusiastic Passives, and remove the friction that creates unhappy customers. Nexus turns the pulse into action by showing NPS per segment and tying it to the behaviour and value that loyalty is supposed to predict.
Turn NPS into loyalty you can act on with Nexus
A company-wide score hides the segments that matter. Nexus shows NPS by customer segment and reads it alongside retention, repeat-purchase rate, and lifetime value, so you close the loop where it protects the most revenue.