All terms

Growth

Net Promoter Score (NPS)

A customer-satisfaction metric scored as the percentage of customers rating you 9-10 on a 0-10 'how likely are you to recommend?' question, minus the percentage rating 0-6. Scores range from -100 to +100.

By Daniel Reyes · Last updated June 23, 2026

In plain English

Subtract your detractors (0-6) from your promoters (9-10). Ignore the 7-8 passives. The number is less useful as an absolute benchmark and more useful as a delta — did NPS go up or down after the last release / pricing change / support overhaul?

Example

100 responses: 45 promoters, 30 passives, 25 detractors. NPS = 45% - 25% = +20. Industry context: B2B SaaS averages +30; consumer apps span -10 to +50; world-class products (Apple, Costco) sit at +70+. The number is noisy at small sample sizes — anything below ~50 responses, treat NPS as directional not measured.

Formula

NPS = (% Promoters) − (% Detractors) where promoters = 9-10, detractors = 0-6, passives (7-8) excluded.

Why it matters

NPS is the most-used customer-sentiment metric in B2B because it's quick to ask and tracks change over time well. Its weaknesses are real (it's a single number compressing a lot of context, it's heavily sample-biased, cultural calibration varies wildly between US and Europe), but as a quarterly delta against your own baseline it's a useful signal. The mistake is treating it as a benchmark to compare against other companies — too many variables for that to be valid. Track your own NPS over time, segment it by ICP and tenure, and chase the 'why did you give this score?' open-text follow-up rather than the number itself.

Common mistakes

  • Reporting NPS as a single blended number rather than segmenting by ICP, tenure, or product surface
  • Comparing your NPS to public benchmarks across industries — too many variables to be meaningful
  • Asking the question too soon (week 1 of a customer relationship — they haven't formed an opinion yet) or too often (survey fatigue tanks response rates)
  • Ignoring the open-text follow-up; the 'why?' is where the actionable signal lives, not the number

Related