What is NPS survey scoring: the question, the scale and the calculation

An NPS survey is one recommendation question on a 0 to 10 scale

An NPS survey asks customers a single question: how likely are you to recommend us to a friend or colleague? They answer on a scale from 0 to 10. NPS stands for Net Promoter Score, and the score is the share of customers who answer 9 or 10 minus the share who answer 0 to 6.

The method comes from Frederick F. Reichheld's article "The One Number You Need to Grow", published in Harvard Business Review in December 2003. Bain & Company, where Reichheld worked, describes how he and a Bain team tested many survey questions against real customer behavior and found the recommendation question was the best single predictor in most of the industries they studied. Net Promoter and NPS are registered trademarks of Bain & Company, NICE Systems and Fred Reichheld.

Promoters, passives and detractors on the 0 to 10 scale

Every answer falls into one of three groups. The cut-off points are fixed, and Bain's own guide gives them as follows.

Bain gives two reasons for the eleven-point scale. Starting at zero stops people from mixing up which end is good. And the wide range leaves room to tell a truly keen customer from one who is merely content. That is why a 7 or 8, which sounds like a good mark, is not counted as a promoter.

How to calculate NPS, with a worked example

1. Count the total number of people who answered the question.
2. Count the promoters (9 or 10) and the detractors (0 to 6).
3. Turn each count into a percentage of the total.
4. Subtract the detractor percentage from the promoter percentage. Passives count in the total but are not added or subtracted.

Here is an example. A dental practice gets 200 answers. 110 people answer 9 or 10, 50 answer 7 or 8, and 40 answer 0 to 6.

The score is 55 minus 20, which is 35. NPS is written as a whole number, not a percentage, and it can run from minus 100, where everyone is a detractor, to plus 100, where everyone is a promoter. Always report the number of responses next to the score. Our guide on counts before percentages explains why that matters with small samples.

Common mistakes when reading an NPS survey

• Averaging the answers. An average of 7.9 is not an NPS. The score is a difference between two percentages.
• Treating 7 and 8 as promoters. They are passives and do not add to the score.
• Changing the scale. A 1 to 5 or 1 to 10 scale gives a number that cannot be compared with a standard NPS.
• Asking the score and nothing else. The number tells you how things stand, not why. Add one open question asking for the main reason. See the wording in our NPS survey examples.
• Comparing against another company's number. Different industries, customers and survey moments give different scores. Your own trend is the fair comparison.

NPS is also only one of several measures. If you want to rate a single visit or support call, a satisfaction question may suit better, as explained in NPS or CSAT.

When a paid NPS tool is worth it

You can run an NPS survey with any form that offers a 0 to 10 scale and work out the score in a spreadsheet. That is fine for a first survey or a few dozen answers.

A paid tool becomes worth it when the sorting and follow-up take real time. Yay! Forms, for example, has a dedicated NPS question that arrives with the standard wording and scale, calculates the score automatically, and can route promoters and detractors to different follow-up questions with logic. Its Starter plan was $19 a month billed monthly, or $9 a month billed yearly, when read on 3 October 2026. For other options, including a free one that calculates the score, see our comparison of NPS survey tools.