NPS calculator online
NPS, the Net Promoter Score, comes from a single question: "How likely are you to recommend us to a friend?" Paste 0–10 scores from Excel or just enter the count for each group to see your NPS, the share of promoters and detractors, and the confidence interval. The interval tells you how far you can trust the number with this many responses.
Calculate your own NPS
Score 9 and 10
Score 7 and 8
Score 0 to 6
NPS
20
Good
95% confidence interval: 8.7 to 31.3 (±11.3)
- Detractors
- 25%(50)
- Passives
- 30%(60)
- Promoters
- 45%(90)
NPS formula
Formula
NPS = %Promoters − %DetractorsPercentage of promoters minus percentage of detractors
| Score | Group | Typical behaviour |
|---|---|---|
| 9 and 10 | Promoter | Happily recommends you and is more likely to stay. |
| 7 and 8 | Passive | Satisfied but not loyal, and leaves for a better offer. Not in the numerator, but counted in the total. |
| 0 to 6 | Detractor | Unhappy, and may talk others out of buying too. |
For example, out of 200 responses, 90 are promoters (45%), 60 passives (30%) and 50 detractors (25%): NPS is 45 minus 25, i.e. 20. NPS runs from −100 to +100 and is written without a percent sign: "NPS is 20", not "20%".
NPS is not the average score. An average of 7 out of 10 can come with a negative NPS, because NPS only counts the two ends of the scale. We explain why with a worked example in What is NPS and why averaging it is wrong.
Margin of error: how certain is your NPS?
NPS is calculated from a sample and, like any estimate, has error. The calculator above gives the confidence interval with this formula:
Standard error and confidence interval
SE = √((p + d − (p − d)²) ÷ n)p and d are the shares of promoters and detractors between zero and one; n is the total number of responsesNPS ± z × SE × 100z is 1.96 for 95% confidence
In the example above (200 responses, NPS 20), the 95% interval runs from about 9 to 31; the margin of error is about 11 points. With the same percentages but only 50 responses, the interval widens to −3 to 43, and even a positive NPS is not certain.
In practice: if NPS was 22 this quarter and 18 last quarter, with about 200 responses each time, the difference is within the margin of error and you cannot say customer loyalty improved. Before announcing "NPS growth", put the intervals side by side. To compare the share of promoters between two periods you can also use the Significance test for two percentages.
This interval uses the normal (Wald) approximation. With fewer than about 30 responses, or when almost everyone is in one group, the approximation is poor and the calculator warns you.
What is a good NPS?
| NPS | Interpretation |
|---|---|
| Above 50 | Excellent; promoters clearly outnumber detractors. |
| 20 to 50 | Good; word of mouth is working for you. |
| 0 to 20 | Acceptable but fragile; a little dissatisfaction turns it negative. |
| Below zero | Detractors outnumber promoters; a serious warning, not just a low score. |
These cut-offs are approximate. NPS levels vary hugely between industries, and comparing a bank's NPS with a café's is meaningless. The most reliable comparison is your own NPS over time, or across your own branches and products, with the same question, the same timing and the same collection method.
Running an NPS survey properly
- Don't change the standard question: "How likely are you to recommend [your name] to a friend or colleague?" on a 0–10 scale labelled "Not at all likely" and "Extremely likely" at the ends.
- Follow it immediately with an open question: "What is the main reason for your score?" The number tells you where you stand; this question tells you why.
- Keep the timing fixed, e.g. one week after purchase. Asking right after a good experience inflates the score.
- Think about those who didn't respond. If only satisfied customers answer, NPS looks better than reality.
- For the margin of error to reach about ±10, a distribution like the example above needs at least 254 responses.
In Porsino the NPS field builds this 0–10 scale, and the NPS report shows each group's share automatically. The ready-made template Loyalty survey (NPS) also has separate follow-up branches for detractors, passives and promoters. To start from scratch, see Build an online survey for details.
Frequently asked questions
What does a negative NPS mean?
It means the percentage of detractors is higher than the percentage of promoters. It doesn't necessarily mean most customers are unhappy, since most may be passives; but it shows word of mouth is working against you.
Why are passives not in the NPS formula?
They are not in the numerator but they are in the denominator: NPS is the percentage of promoters minus the percentage of detractors out of all responses. Passives have no direct effect, but the more there are, the smaller the other two shares become.
Should NPS be reported as a percentage?
NPS is a number between −100 and +100 and is usually written without a percent sign, as in "NPS is 32". Report the shares of promoters and detractors as percentages.
How many responses are enough for NPS?
It depends on the precision you need. With 50 responses the margin of error in a typical distribution is about ±23 points, and with a few hundred responses it gets down to about ±10. The calculator above shows the margin of error for your own data.
Is the average 0–10 score the same as NPS?
No. An average of 7 can come with a negative NPS. NPS counts only the shares at the two ends of the scale, which is why it is more sensitive to changes in customer behaviour.
Where do I get the scores?
From the score column of your survey's Excel export. Copy the whole column and paste it into the tool; the column heading, empty cells and values outside 0–10 are set aside and counted separately.
Related pages
- Build an online surveyCustomer and employee satisfaction surveys with invitations and ready-made reports.
- Statistical significance testIs the gap between two percentages real or noise? A two-proportion z-test with p-value and confidence interval.
- Sample size calculation (Cochran)Cochran's formula calculator and the Morgan table.
- Build an online questionnaireThesis questionnaires with Likert scales, matrices, conditional logic and direct SPSS export.
Survey & analysis guides
All posts- What Is NPS, and Why Averaging the Scores Is WrongNPS is not the average of the scores; it's the percentage of promoters minus the percentage of detractors. Take the average and you produce a number that has no interpretation.2 min read
- The Complete Guide to Creating an Online Survey: From Question Design to Analyzing ResultsNine steps to an online survey that actually leads to a decision: defining the goal, sample size with Cochran's formula, rules for writing questions, length and structure, distribution, raising the response rate, sound analysis, and a 20-point checklist before you hit send.15 min read
- Questionnaire Design: A Guide to Standard Questionnaires for Theses and Research (With Sample Questions)The standard structure of a questionnaire, choosing a scale, validity (face, content with Lawshe's CVR table, construct) and reliability (Cronbach's alpha with a worked example), sample size, a complete 15-item sample, running it online, and exporting to SPSS.14 min read
Send your NPS survey online
Porsino has an NPS field and a ready-made loyalty survey template. Build the survey for free, send the link to your customers and see the report instantly.