What is Net Promoter Score (NPS), and why is it worth measuring?
By Matúš Marcin, founder of Sensa · Updated
Net Promoter Score (NPS) is a customer-loyalty metric built on one question: “How likely are you to recommend us to a friend or colleague?”, answered on a 0–10 scale. People who answer 9–10 are promoters, 7–8 are passives and 0–6 are detractors. NPS is the percentage of promoters minus the percentage of detractors, so it ranges from −100 to +100.
It was introduced by Fred Reichheld of Bain & Company in a 2003 Harvard Business Review article, “The One Number You Need to Grow”, and has since become one of the most widely used customer metrics in the world.
What the tablet shows once it's locked down
The Sensa kiosk in the two most common setups — a smiley scale and a 0–10 scale. Tap one to try it; nothing is submitted.
Smiley scale, 0–10 NPS, or Likert. Your choice, changeable anytime.
The question
The standard wording
How likely are you to recommend [this place] to a friend or colleague?
with a scale from 0 (“not at all likely”) to 10 (“extremely likely”). Keep the wording close to this — it’s what makes your score comparable over time and roughly comparable with others. Replace “us” with something concrete: “this library branch”, “this store”, “this museum”.
The three groups
| Score | Group | What it usually means |
|---|---|---|
| 9–10 | Promoters | Enthusiastic; likely to come back and to recommend you |
| 7–8 | Passives | Satisfied but unenthusiastic; easily lured elsewhere |
| 0–6 | Detractors | Unhappy enough to warn others away |
Note that 7 and 8 — scores that feel positive — don’t count in your favour. NPS deliberately rewards enthusiasm, not mere satisfaction.
How to calculate NPS
- Count all responses to the NPS question.
- Count promoters (9–10) and detractors (0–6). Ignore passives except in the total.
- Divide each by the total and multiply by 100 to get percentages.
- Subtract the detractor percentage from the promoter percentage. The result is your NPS — a whole number between −100 and +100, written without a % sign.
Worked example: 200 visitors answered. 110 gave 9 or 10, 50 gave 7 or 8, and 40 gave 6 or lower. Promoters are 55%, detractors 20%, so NPS = 55 − 20 = +35.
Why NPS is useful
- It’s one tap. A single 0–10 question is the shortest meaningful survey there is, so response rates stay high — especially on a kiosk, where people give you seconds, not minutes.
- It’s a trend line. One number, measured the same way every week, shows whether things are getting better or worse long before revenue or visitor counts do.
- It’s comparable. Branches, stores or rooms measured with the same question can be ranked side by side.
- It’s understood. Boards, funders and staff already know what NPS is, so you spend less time explaining the metric and more time acting on it.
What NPS doesn’t tell you
NPS tells you how people feel, not why. A score of +20 doesn’t say whether the problem is queues, prices or signage. Researchers have also questioned how well it predicts growth on its own. Treat it as a headline, not a diagnosis:
- Pair it with a few specific questions — 0–10 agreement statements about the things you can change (“Checkout was quick”, “It was easy to find my way around”). The lowest-scoring statement is usually your first fix.
- Optionally add one open question at the end (“What’s one thing we should improve?”) for the reasons behind the numbers.
- Watch the sample size. Below about 30 responses, a handful of answers can swing the score by 20 points. Read trends, not single days.
What is a good NPS?
Any score above 0 means you have more promoters than detractors. Broadly, above +30 is strong and above +50 is excellent — but published averages vary widely by industry, country and survey method, and a score collected at the exit of a building isn’t directly comparable with one collected by email weeks later. The most useful benchmark is your own: the same question, the same place, over time.
NPS in physical spaces
Most NPS programmes run by email, which works when you know who your customers are. Libraries, museums, shops and service counters mostly don’t — visitors walk in and out without leaving a name. For them, the practical way to measure NPS is a tablet at the exit:
- Ask on the spot. The visit is fresh, and you reach the people who’d never open a survey email.
- Keep it anonymous. No login, no email, no personal data — which also keeps GDPR simple.
- Tag every answer with its location, so each branch or store gets its own score.
- Put NPS first. If you ask more than one question, the first one gets the most answers. Make it the one you’d report upward.
Frequently asked questions
- Is NPS the same as customer satisfaction (CSAT)?
- No. CSAT asks how satisfied someone was with a specific interaction, usually on a 1–5 scale, and reports the share of satisfied answers. NPS asks about likelihood to recommend, on 0–10, and subtracts detractors from promoters. Many teams use both: NPS as the headline, satisfaction statements to explain it.
- Why is NPS not a percentage?
- It’s the difference between two percentages, so it’s reported as a plain number from −100 to +100.
- Can NPS be negative?
- Yes — whenever detractors outnumber promoters.
- How many responses do I need?
- Aim for at least 30 before reading the score, and more if you want to compare months or locations with confidence.
- Should I change the wording?
- Swap “us” for the specific place, but keep the structure and the 0–10 scale; otherwise you lose comparability.
- How often should I measure NPS?
- In a physical space, continuously. A kiosk that’s always on gives you a live trend instead of a once-a-year snapshot.
Want NPS for a library, museum or shop? Sensa turns any tablet into an anonymous NPS kiosk — one tap, results by location, and the free plan includes 1 NPS question at 1 location.
Start free with one location