How to calculate Net Promoter Score (NPS): a practical worked example
A customer says they would recommend your product. Another gives a low score but no explanation. Net Promoter Score can help you summarise those responses, provided you keep the calculation separate from the conclusions you draw.
Net Promoter Score, usually shortened to NPS, uses a likelihood-to-recommend question and a scale from zero to ten. This guide follows Bain’s published calculation method, using an original fictional workshop-platform example.
You will learn how to group the responses, calculate the score and decide what to investigate next. You do not need a complicated dashboard to practise it. You need a consistent question and a clear count of valid responses.
Calculate NPS in three steps
1. Group the responses
Ask how likely someone is to recommend your product or service to a friend or colleague. Explain the ends of the scale and keep the question consistent when comparing survey rounds.
- Promoters: respondents who select nine or ten.
- Passives: respondents who select seven or eight.
- Detractors: respondents who select a score from zero to six.
These labels describe the survey categories. A rating does not tell you exactly what that person will buy, say or do next. Follow-up questions help you understand the experience behind it.
2. Calculate two percentages
Divide the number of promoters by the total number of valid responses, then multiply by 100. Repeat that calculation for detractors. The denominator is the total number you divide by.
Include passive responses in that total even though their percentage is not subtracted directly. Removing them changes the calculation and makes the result difficult to compare with NPS calculated correctly.
Keep the original counts alongside the percentages. They make the arithmetic easier to check and show how much evidence sits behind the score.
3. Subtract detractors from promoters
NPS = percentage of promoters − percentage of detractors.
The result ranges from minus 100 to plus 100. Report it as a score, rather than writing it as a percentage. A score of zero means the two percentages are equal; it does not mean nobody was satisfied.
A negative result means the detractor percentage exceeds the promoter percentage. That is a reason to investigate the responses, rather than guessing the cause from the number alone.
Try the worked example
A fictional workshop platform
The platform receives 80 valid responses: 36 promoters, 28 passives and 16 detractors. What is its NPS? Write down the two percentages first.
Reveal the calculation
Promoters: 36 ÷ 80 × 100 = 45%. Detractors: 16 ÷ 80 × 100 = 20%. NPS: 45 − 20 = 25.
All 80 responses belong in the denominator, including the 28 passives. These numbers are invented for practice. They are not Coded Female survey results or a recommended benchmark.
What if you leave out the passives?
You would divide by 52 instead of 80. The resulting score would be approximately 38.5, which is not the correct NPS for these responses. Keep the passive responses in the total.
Turn the score into a useful question
Record the sample size, collection dates and who was invited. A small or selective sample can give a misleading impression of your wider audience. When comparing rounds, check whether the sampling method or customer group changed.
Read the comments for recurring problems and differences between groups. Choose one issue to investigate and decide what evidence would help you act. Avoid treating a higher score as proof that a particular product change caused the improvement.
NPS can be one input into a product decision. Combine it with evidence about the task people are trying to complete, the problems they encounter and what changes after you address them.
For your next survey review, bring the response counts, your calculation and one question the score cannot answer. That gives the team a practical starting point for investigating the experience behind the number.
You May Also Like
Stop reopening the same product decision: use a six-part decision log
Your Ultimate Guide to a Successful Product Launch: Step-by-Step with Real-World Example
