Tickbox

Free NPS calculator

Paste raw scores, choose one column of a CSV file, or enter counts per rating. No signup. All processing is in your browser.

Put one score on each line, or separate the scores with commas, semicolons, tabs or spaces.

Compare two NPS results

Find out if the difference between two NPS results is larger than chance. Use it for two periods, for example two quarters, or for two groups of different customers. Each person must be in one result only.

Result A

For example, the earlier period or the first group.

Result B

For example, the later period or the second group.

What this NPS calculator does

Net Promoter Score (NPS) comes from one survey question: how likely are you to recommend us, on a scale from 0 to 10? This calculator takes the answers and gives the score, its margin of error, the three groups and each calculation step. It can also compare two NPS results and tell you if the difference is statistically significant. Your scores and files stay in your browser.

How NPS is calculated

Each response goes into one of three groups:

  • Promoters: Ratings 9 and 10
  • Passives: Ratings 7 and 8
  • Detractors: Ratings 0 to 6

NPS = % promoters − % detractors

Passives count in the total, but you do not add or subtract them. The result is a number from −100 to 100, not a percentage.

Example: 10 responses, with 5 promoters, 3 passives and 2 detractors. Promoters are 50% and detractors are 20%, so the NPS is 50 − 20 = 30.

How the margin of error and the comparison are calculated

An NPS from a sample is an estimate. The margin of error shows how much it can change from one sample to a different sample. This calculator uses the adjusted-Wald method that Jim Lewis and Jeff Sauro of MeasuringU give for NPS:

  1. Add 3 to the number of responses. Add 0.75 to the promoters and 0.75 to the detractors.
  2. Divide the adjusted promoters and the adjusted detractors by the adjusted total. The adjusted NPS is the adjusted promoter proportion minus the adjusted detractor proportion.
  3. Variance = adjusted promoter proportion + adjusted detractor proportion − (adjusted NPS)².
  4. Margin of error of one result = z × √(variance ÷ adjusted total). z is 1.645 at the 90% confidence level and 1.96 at the 95% confidence level.
  5. For two results, the standard error of the difference is √(variance A ÷ adjusted total A + variance B ÷ adjusted total B). The test value z is the difference divided by the standard error. The two-tailed p-value comes from the standard normal distribution.

Example from the MeasuringU article: result A has 7 promoters, 12 passives and 12 detractors (NPS −16). Result B has 15 promoters, 13 passives and 8 detractors (NPS 19). The adjusted difference is 0.33, or 33 points. z is 1.81 and p is 0.07. The 90% confidence interval of the difference is 3 to 62 points. The difference is significant at the 90% confidence level, but not at the 95% level.

How to use the calculator

  1. Choose the tab that matches your data: Paste scores, CSV file or Counts per rating.
  2. Enter your data. If your CSV file has more than one column, choose the score column.
  3. Select Calculate NPS.
  4. Read the score, the margin of error, the three groups and the calculation steps. Select 90% or 95% to change the confidence level of the margin of error. If an entry is not valid, a message shows where it is. Correct it and calculate again.
  5. To compare two results, go to Compare two NPS results. Enter counts or raw scores for result A and result B, select the confidence level, then select Compare results.

NPS calculator FAQ

What is a good NPS?

A score above +44 is above the median in SurveyMonkey's 2025 benchmark data, from more than 150,000 organizations. In that data, the global average is +32, the bottom quartile is 0 or lower and the top quartile is +72 or higher. These figures are general. The better comparison is organizations in the same industry and your own earlier scores, from the same question asked in the same way.

How do I know if a change in NPS is real?

Use Compare two NPS results on this page. Enter the counts or the raw scores of the two periods. If p is less than 0.05, the difference is statistically significant at the 95% confidence level, so chance alone is an unlikely cause. With small samples, chance can cause large changes. In the MeasuringU example, about 30 responses in each result and a change of 35 points were significant at the 90% level only. Ask the same question in the same way in the two periods, and do not put the same people in both results.

Can NPS be negative?

Yes. NPS is negative when you have more detractors than promoters. The minimum is −100, when all responses are detractors. The maximum is 100, when all responses are promoters.

Do passives change the score?

Yes, but only through the total. More passives make the percentages of promoters and detractors smaller, so the score moves closer to 0.

Which CSV files can I use?

Use a text file with values separated by commas, semicolons or tabs. If you use a spreadsheet, export the sheet as CSV first. The score column must contain only whole numbers from 0 to 10. A heading in row 1 and blank cells are accepted.

Is my data sent to a server?

No. The calculation runs in your browser, and this page does not send or save your scores or files. The site uses Plausible analytics to count page visits. Plausible does not receive what you enter.

Do I need an account?

No. The calculator is free and needs no signup.

Related pages

Method source: Measuring your Net Promoter Score, Bain & Company. Benchmark source: 2025 NPS benchmarks, SurveyMonkey. Margin of error and comparison method: How to Statistically Compare Two Net Promoter Scores, Jim Lewis and Jeff Sauro, MeasuringU, 2021.