Net Promoter Score (NPS)

NPS turns one survey question, "How likely are you to recommend us to a friend or colleague?" on a 0–10 scale, into a single score. Answers of 9 and 10 are promoters, 7 and 8 are passives, 0 to 6 are detractors:

$\text{NPS} = 100 \times \left(\frac{\text{promoters}}{n} - \frac{\text{detractors}}{n}\right)$

The result runs from −100 (everyone a detractor) to +100 (everyone a promoter) and is written as a plain number, "NPS 20", not a percentage. Passives count in $n$, so adding passives pulls the score towards zero without being positive or negative themselves.

NPS from a survey is an estimate. Another sample of the same size would give a different score, and the spread is wider than most people expect because the score is the difference of two shares from the same respondents. Score every answer +1, 0 or −1: the NPS is 100 × their average, and its standard error is $100\sqrt{V/n}$ with $V = p_P + p_D - (p_P - p_D)^2$. With 200 answers the 95% range is roughly ±11 points; a ±5 point range takes about 1,000 to 1,500 answers.

The simple ± 1.96 × standard error range breaks down at small samples and lopsided answers (with every answer a promoter it has zero width), so a reliable calculator adds a few imaginary answers first, three quarters of an answer to promoters and to detractors, before building the range.

The same score can hide very different mixes: 40% promoters and 20% detractors, or 20% and 0%, both give NPS 20. Report the three shares next to the score.