Free tool
Online NPS Calculator with Margin of Error
Enter your promoter, neutral and detractor counts to see your score and its margin of error, then compare two samples to find out if a change is real or just statistical noise.
Your NPS and margin of error
Net Promoter Score
33
Margin of error: ±10.1 points
With 240 responses, an NPS of 33 has a margin of error of ±10.1 points, so the true score is likely between 23.2 and 43.4.
Compare two samples
Has your score really changed between two periods, or two segments?
Margin of error test: has it changed?
No
The 4.2-point difference is within the 4.3-point margin of error, so this could just be sampling noise.
Chi-squared test: has the mix changed?
No
We can't be 95% sure this is a real shift, it could still just be sampling noise.
Want to see the working? Read how these formulas work.
Free NPS Statistics SpreadsheetThe NPS Formula
The NPS formula subtracts the percentage of Detractors from the percentage of Promoters.
The Net Promoter Score formula
Neutrals are not subtracted. They still count in Total Respondents, so every Neutral pulls the score towards zero.
The formula uses the answers to one survey question: how likely are you to recommend us to a friend or colleague, on a scale from 0 to 10. The Net Promoter Score question explains where it came from and how it fits into a customer survey.
Worked Example: NPS From 240 Responses
A survey with 240 responses gives an NPS of 33. These are the figures the calculator opens with.
| Group | Rating | Responses | Share of responses |
|---|---|---|---|
| Promoters | 9 or 10 | 130 | 54.17% |
| Neutrals | 7 or 8 | 60 | 25.00% |
| Detractors | 0 to 6 | 50 | 20.83% |
| Total | 240 | 100% |
The percentage of Promoters is 130 divided by 240, or 54.17%.
The percentage of Detractors is 50 divided by 240, or 20.83%.
54.17 minus 20.83 is 33.33, reported as an NPS of 33.
The Margin of Error on an NPS of 33
With 240 responses, an NPS of 33 has a margin of error of ±10.1 points at 95% confidence. The true score is likely between 23.2 and 43.4.
A margin of error that wide changes how you read the next result. A move from 33 to 38 next quarter, on a similar number of responses, sits inside it. It is not evidence that anything changed. The calculator’s comparison panel tests two samples for you, and how to calculate the NPS margin of error shows the working.
The margin of error is also why I advise against hard NPS targets tied to bonuses. Rewarding or penalising someone for a movement inside the margin of error is rewarding noise.
Calculate NPS in Excel or Google Sheets
To calculate NPS in a spreadsheet, count each group with the COUNTIF function, then apply the formula above. How to calculate NPS in Excel or Google Sheets builds it step by step.
To work out the margin of error and compare two samples offline, download the free NPS statistics spreadsheet.
Frequently Asked Questions
How is NPS calculated?
- NPS is calculated by subtracting the percentage of Detractors from the percentage of Promoters. Customers rate how likely they are to recommend you on a 0 to 10 scale. Ratings of 9 and 10 are Promoters, 0 to 6 are Detractors and 7 and 8 are Neutrals. Divide the Promoter and Detractor counts by the total number of responses, subtract, and multiply by 100.
What is the NPS formula?
- The NPS formula is the percentage of Promoters minus the percentage of Detractors. Written out, NPS equals the number of Promoters divided by Total Respondents, minus the number of Detractors divided by Total Respondents, multiplied by 100. Neutrals appear only in Total Respondents. An NPS of 33 means Promoters outnumber Detractors by 33 percentage points.
What range can NPS take?
- NPS ranges from -100 to +100. A score of -100 means every respondent is a Detractor. A score of +100 means every respondent is a Promoter. A score of 0 means Promoters and Detractors are equal in number, which can hide very different mixes of customers. NPS is a difference between two percentages, so it is reported in points, not as a percentage.
Has my NPS really changed?
- An NPS has really changed only when the difference between two scores is larger than random sampling variation. Enter both periods in the calculator to test it. The margin of error test checks whether the score moved. The chi-squared test checks whether the mix of Promoters, Neutrals and Detractors moved, which catches two samples with the same score but different customers.
Did my NPS really, really change? · NPS margin of error and chi-squared