How it is calculated
The Net Promoter Score answers a single question — “How likely are you to recommend us?”, on a scale from 0 to 10. The answers are split into three groups and subtracted from one another:
| Score | Group | Effect |
|---|---|---|
| 9 – 10 | Promoters | counts positive |
| 7 – 8 | Passives | does not count — but dilutes both shares |
| 0 – 6 | Detractors | counts negative |
NPS = share of promoters − share of detractors, in percentage points. The result sits between −100 (everyone a detractor) and +100 (everyone a promoter). It is not a percentage and not a grade; it is a difference.
Why passives count after all
They are absent from the formula but present in the denominator. Many passives shrink both shares and pull the score towards zero. A survey full of sevens and eights produces a clean zero — and that means something entirely different from a zero made of half promoters and half detractors.
Where the calculation happens
Entirely in your browser. There is no transmission, no server and no database behind this page — the responses you paste never leave your device.
What the score does not tell you
With few responses it is a coin toss. That is why the calculator shows how far a single additional response would move the result. With twenty responses that is several points — reporting a three-point rise on such a basis is reporting noise.
It does not compare across industries. A good score in a business with few, close customer relationships would be weak in a volume business, and the other way round. The only comparison that holds is against your own previous year, with the same question and the same way of asking.
It does not say why. The number is a thermometer, not a diagnosis. The insight sits in the free-text field next to it, not in the score.
Best practice
The companies that get something out of it treat the NPS not as a report but as a trigger: every answer in detractor territory produces a phone call within days — not from marketing, but from the people who can fix the case. What then gets measured is no longer only the score, but the time to that call and how many of those cases answer differently next time.
Lifting the number onto a quarterly slide instead leaves you with a metric that decides nothing — and that is decoration.