Benchmarks

What is a good NPS score?

Net Promoter Score compresses customer sentiment into one number from −100 to +100. Here’s how it’s calculated, the heuristic bands for what counts as good, and why you should read it as a direction, not a verdict.

Net Promoter Score is everywhere because it’s simple: one question, one number, easy to track over time. That simplicity is also its trap, a single figure invites over-reading. Here’s what NPS is, what counts as “good,” and how to keep it honest.

How NPS is calculated

Ask one question (“how likely are you to recommend us: 0 to 10?”), and bucket the answers:

  • Promoters (9-10): loyal enthusiasts.
  • Passives (7-8): satisfied but unenthusiastic.
  • Detractors (0-6): unhappy, and a churn and word-of-mouth risk.
NPS = % promoters − % detractors

Passives count toward the totals but not the score, so the result ranges from −100 (everyone a detractor) to +100 (everyone a promoter).

What counts as good

The commonly-cited heuristic bands:

  • Above 0: more promoters than detractors. The minimum bar.
  • 30+: generally considered good; most customers are happy.
  • 50+: excellent; a strong base of advocates.
  • 70+: world-class, the territory of the most loved brands.

These bands vary a lot by industry (sectors with naturally low goodwill score lower across the board), so the useful comparison is within your own sector and, above all, against your own trend over time.

Why to read it as a direction, not a verdict

NPS compresses a whole distribution into one figure, drops the passives entirely, and shifts with who you ask and when. A score with no context (sample size, segment, timing) is easy to misuse. Two honest habits help: segment it (new vs long-tenured customers can differ sharply) and prefer transactional NPS (asked right after a specific interaction) over a single relational number, because it ties sentiment to a concrete moment you can act on.

Where NPS fits with analytics

NPS is stated sentiment; product analytics is revealed behavior. The two are strongest together: a detractor who also shows declining usage and rising churn risk is a different, more urgent story than a detractor who’s still highly active. Putting a survey score next to a unified profile turns an abstract number into a person you can understand.

Calculate yours

Use the free NPS calculator to turn promoter and detractor counts into a score. Then don’t stop at the number: segment it, watch the trend, and read it alongside what those customers actually do. For related targets, see what a good retention rate is.

FAQ

Common questions

What is a good NPS score?

NPS runs from −100 to +100. As widely-cited heuristics: anything above 0 means you have more promoters than detractors, above ~30 is generally considered good, above ~50 excellent, and above ~70 world-class. These bands vary by industry, so compare within your sector and against your own trend.

How is NPS calculated?

Ask “how likely are you to recommend us?” on a 0-10 scale. Promoters score 9-10, passives 7-8, detractors 0-6. NPS = % promoters − % detractors. Passives count toward the total but not the score.

Why can NPS be misleading?

It compresses a 0-10 distribution into one number, ignores passives, and is sensitive to who you survey and when. A single NPS figure with no context (sample size, segment, trend) is easy to over-read. It is a directional signal and reads best as one.

What is the difference between relational and transactional NPS?

Relational NPS surveys overall sentiment periodically; transactional NPS asks right after a specific interaction (a purchase, a support ticket). Transactional NPS is more actionable because it ties the score to a concrete moment.

Connect sentiment to behavior.

Pug ships per-person profiles and behavioral analytics, so a survey score can sit next to what users actually did. Open source, self-hostable, free in open beta.

Questions? Email hello@pug.sh