NPS = % Promoters (9–10) − % Detractors (0–6). Passives (7–8) count in the total but don’t affect the score. Range −100 to 100.
Step 1 — Enter response counts
Type how many people picked each score (0–10). Not sure? .
Step 2 — Your score
- Detractors: scores 0–6 → lower NPS
- Promoters: scores 9–10 → raise NPS
- Passives: scores 7–8 → don’t change NPS but count in total
- Formula: NPS = %Promoters − %Detractors
We tested the NPS calculator with real survey data from a 200-person customer base. One thing that trips people up: the score swings dramatically at low response counts. With fewer than 50 responses, a handful of detractors can tank your NPS by 20 points. Wait until you have at least 100 responses before drawing conclusions about trend direction.
What is NPS and how does it work?
Net Promoter Score boils customer loyalty down to a single question: how likely are you to recommend this company to a friend or colleague, on a scale of zero to ten? It sounds almost too simple to be useful, and that is exactly why it caught on. One number, tracked over time, that tells you whether people would put their own reputation on the line to vouch for you.
The math sorts every respondent into three groups. Anyone who answers 9 or 10 is a promoter — the loyal, enthusiastic customer who actually brings you referrals. Scores of 7 or 8 are passives: satisfied enough to stick around, but not excited enough to talk about you. Everyone from 0 to 6 is a detractor, and that band is wider on purpose, because a 6 is not a mild compliment — it is a warning. Your score is the percentage of promoters minus the percentage of detractors. Passives count toward your total response volume but never move the number directly.
That subtraction is what makes NPS swing harder than people expect. A score of +50 is not “50% happy” — it could mean 60% promoters and 10% detractors, or 50% promoters and no detractors at all. Enter your response counts in the calculator above to get your score instantly, and watch how a handful of detractors pulls the needle down faster than the same number of promoters pushes it up.
What is a good NPS score?
There is no universal pass mark. Any score above zero technically means you have more promoters than detractors, and anything north of +50 is genuinely strong — but “good” only means something next to your own industry, where customer expectations and switching costs differ wildly. A bank sitting at +70 and a telecom at +30 can both be winning their category.
| Industry | Average NPS |
|---|---|
| Technology / SaaS | 40–55 |
| Ecommerce | 50–56 |
| Financial services | 70–75 |
| Healthcare | 50–55 |
| Telecommunications | 25–35 |
| Hospitality | 40–50 |
| B2B services | 25–45 |
Read these ranges as gravity, not goalposts. Industries with high switching friction and emotional buy-in, like financial services and hospitality, tend to post higher averages, while sectors people love to complain about, like telecom, sit far lower without any individual company doing much wrong. The useful comparison is always the same one: are you above or below the typical score for the businesses your customers would consider switching to?
How to interpret your results
A negative score is not a catastrophe, but it is a clear instruction: more people are actively steering others away from you than recommending you, and no growth plan survives that for long. Treat it as a fire to put out before you invest in anything fancier. The fastest signal hides in your detractors’ written comments — the score tells you something is wrong, the open-text answers tell you what, and that is where your reading time pays off most.
What matters more than any single number is the direction it is moving. A flat +20 that has held for a year is a very different story from a +20 that was +35 last quarter. Track the trend across the same survey design and the same customer segments, and let the slope, not the absolute figure, drive your decisions.
NPS survey best practices
The single highest-leverage thing you can do is ask “why?” immediately after the rating. The 0–10 number is a symptom; the follow-up comment is the diagnosis, and a survey without it hands you a scoreboard with no game footage. Make that second question open-ended and optional, and resist the urge to bolt on ten more — every extra field you add quietly drops your completion rate.
Timing is the next thing people get wrong. Fire a relational survey too early and you are measuring a first impression, not loyalty; send transactional ones too often and you train customers to ignore you. Survey the same people no more than once a quarter for relationship NPS, and tie transactional surveys to a real moment — a finished onboarding, a resolved support ticket — when the experience is fresh enough to rate honestly.
Then actually close the loop. Reaching back out to a detractor within a few days does two things at once: it occasionally rescues the relationship, and it consistently teaches you more about your product than any dashboard will. The companies that treat NPS as a conversation starter rather than a report card are the ones that watch the number climb.
Finally, do not let NPS stand alone. It captures loyalty, but it says nothing about whether a specific interaction was easy or satisfying. Pair it with CSAT for transactional happiness and CES for effort, and you get a fuller picture than any one metric can offer on its own.
Best tools to run NPS surveys
You can run NPS in a spreadsheet, but dedicated tooling handles the segmenting, follow-up triggers, and trend tracking that make the metric worth collecting in the first place. Three tools we have reviewed stand out for NPS specifically:
Best for conversational NPS with high completion rates.
Its chat-style surveys consistently pull higher response rates than static forms, which matters most when your volume is low.
Starting at $19/mo
Read our review →Best for simple, fast NPS deployment.
Purpose-built for NPS and CSAT, it is the quickest way to get a survey live without configuration overhead, and the free tier is genuinely usable.
Free to start
Read our review →Best for advanced analytics and research-grade NPS programs.
The deepest reporting of the three, built for teams running large or methodologically rigorous programs.
Starting at $99/mo
Read our review →Compare all NPS-focused tools →
Frequently asked questions
How many responses do I need for an accurate NPS?
Aim for 30–50 responses as an absolute minimum, 100 or more for a number you can trust quarter to quarter, and 500+ if you are running a high-volume program and want to segment with confidence. Below 50, a single cluster of detractors can swing your score by 20 points or more.
How often should I run NPS surveys?
Run relational NPS quarterly to annually for the same customers, and transactional NPS after specific interactions such as onboarding or a support resolution. Avoid surveying the same people monthly — survey fatigue drops response rates significantly and skews who bothers to reply.
What's the difference between NPS, CSAT, and CES?
NPS measures long-term loyalty and likelihood to recommend, usually as one company-wide number. CSAT (Customer Satisfaction) measures how happy someone was with a specific interaction, typically on a 1–5 scale right after the event. CES (Customer Effort Score) measures how easy it was to get something done, and it is often the strongest predictor of repeat business. Each answers a different question, so the strongest programs run more than one — see our CES calculator and CSAT calculator.
Can I share my NPS results from this calculator?
Yes — use the "Copy Link" button in the calculator above. The URL captures your exact inputs, so you can paste it to a teammate and they will see the same score and breakdown you do.
