---
name: net-promoter-score
description: Use when asked to set up, calculate, or interpret Net Promoter Score (NPS) — a customer-loyalty metric from a single 0-10 "how likely to recommend" survey question sorting respondents into promoters, passives, and detractors.
---

# Net Promoter Score

Net Promoter Score (NPS) is a customer-loyalty metric built from a
single survey question: "How likely are you to recommend [product/
company] to a friend or colleague?", scored 0 to 10. Its appeal is
simplicity — one question, one number, cheap to run repeatedly and easy
to trend over time.

## Scoring method

Respondents are sorted into three groups by their score:

- **Promoters (9-10)** — loyal, enthusiastic customers likely to keep
  buying and to refer others.
- **Passives (7-8)** — satisfied but unenthusiastic; not actively
  promoting, not actively at risk.
- **Detractors (0-6)** — unhappy customers, at risk of churning and
  capable of actively discouraging others.

The score itself is the percentage of respondents who are promoters
minus the percentage who are detractors (passives are counted in the
total but don't otherwise enter the formula). The result ranges from
-100 to +100; there's no universal "good" number — what matters is the
score relative to the organization's own history and, where available,
its industry.

## What it's good for, and what it isn't

NPS is good for what it's designed for: a simple, cheap, trackable
trend that can be run repeatedly (monthly, quarterly, per interaction)
and compared over time or across segments. It is not a substitute for
understanding *why* customers scored the way they did — the number by
itself says nothing about what's driving loyalty or dissatisfaction.
Pairing the score with a follow-up open-ended question ("What's the
main reason for your score?") is what turns a trend line into
something the business can act on.

## Common pitfalls

- **A single point-in-time score treated as meaningful on its own** — an
  NPS of 40 means little without a trend or a comparable baseline to
  read it against; the value of NPS is mostly in its movement over
  time, not any one reading.
- **Biased or too-small a sample** — a survey sent only to happy,
  engaged customers (or with a response rate so low that only the most
  motivated respondents answer) produces a score that doesn't represent
  the broader customer base.
- **No follow-up question** — collecting the 0-10 score with nothing
  else leaves a low score with no clue what to actually fix, and a high
  score with no clue what to protect.
- **Chasing the score itself** — optimizing survey timing or wording to
  nudge the number up (e.g. surveying only right after a good
  experience) produces a flattering number that no longer reflects
  actual customer sentiment.
- **Ignoring segment differences** — one blended NPS across all
  customers can hide that a specific plan tier, region, or cohort is
  far less satisfied than the overall number suggests.

## Learn more

- [[customer-success-plan]] for using NPS as one ongoing health
  indicator within a customer relationship.
- [[churn-analysis]] for correlating low scores and detractor status
  with actual churn outcomes.
- [[customer-journey-map]] for deciding at which points in the journey
  to actually ask the NPS question.
- [[customer-persona]] for interpreting NPS results by the type of
  customer responding rather than as one undifferentiated group.