NPS, or Net Promoter Score, is a customer loyalty metric derived from a single survey question: how likely a respondent is to recommend a company, product or service to a friend or colleague, answered on a 0 to 10 scale. The score equals the percentage of promoters minus the percentage of detractors, so it ranges from -100 to +100.
In market research, NPS is used as a compact indicator of customer sentiment and word-of-mouth potential, tracked over time and compared across segments, touchpoints or competitors. Its simplicity explains its popularity, and also its main limitations.
How the Net Promoter Score is calculated
Respondents are grouped by their answer. Those who answer 9 or 10 are promoters, those who answer 7 or 8 are passives and those who answer 0 to 6 are detractors. Passives count toward the total number of responses but do not enter the subtraction directly.
| Group | Score range | Typical interpretation |
|---|---|---|
| Promoters | 9-10 | Enthusiastic, likely to recommend and repurchase |
| Passives | 7-8 | Satisfied but open to competitors |
| Detractors | 0-6 | Unhappy, may discourage others |
If 400 customers respond and 200 are promoters, 120 passives and 80 detractors, the result is 50 percent minus 20 percent, giving an NPS of 50. NPS is reported as a whole number, not as a percentage.
Because it is a difference of two proportions, NPS has a wider margin of error than a simple percentage. With about 200 responses, the 95 percent confidence interval is often roughly plus or minus 10 points, so small movements between periods are frequently statistical noise, not real change.
Relational and transactional NPS
NPS surveys are run in two main ways that answer different research questions. Relational NPS measures overall loyalty toward the brand at regular intervals, while transactional NPS measures the experience of a specific interaction shortly after it happens.
| Type | Timing | What it reveals | Typical owner |
|---|---|---|---|
| Relational NPS | Quarterly or yearly | Overall loyalty and brand health | Leadership, marketing |
| Transactional NPS | Hours to days after an interaction | Quality of a specific touchpoint | Operations, support |
| Competitive benchmark NPS | Periodic market survey | Loyalty relative to competitors | Market research, strategy |
A follow-up open question, usually asking for the main reason behind the score, is essential in both types. The number shows the level of loyalty; the verbatim comments explain what drives it and therefore what to change.
The survey channel also influences results. Email, in-app prompts, text messages and phone interviews reach different customers and produce different response rates, so a change of channel can shift the score; teams should keep the method stable or run both methods in parallel for a period before switching.
NPS compared with other customer metrics
NPS is one of several survey metrics in customer and market research, and each measures a different construct. NPS targets advocacy and loyalty intention, whereas satisfaction metrics capture how well a specific experience met expectations.
| Metric | Question focus | Scale | Best use |
|---|---|---|---|
| NPS | Likelihood to recommend | 0-10, net score | Loyalty tracking, benchmarking |
| CSAT | Satisfaction with an interaction | Often 1-5 | Evaluating specific touchpoints |
| CES | Effort needed to solve an issue | Often 1-7 | Service and self-service design |
| Retention rate | Actual repeat behavior | Percentage | Validating stated loyalty |
NPS measures stated intention, not actual behavior. Its relationship with growth or retention differs by industry and should be validated against real repurchase or churn data before it is used as a primary business target.
Running and interpreting an NPS program
Reliable NPS results depend on survey design and sampling as much as on the formula. The question wording, the moment of sending and the selection of respondents all shift the score.
- Keep the standard wording and scale so results remain comparable over time and with benchmarks.
- Survey a representative sample, not only customers who recently contacted support or bought again.
- Report the response count and margin of error next to the score.
- Segment the results by product, region, customer tenure and channel to find specific drivers.
- Close the loop: contact detractors where appropriate and route recurring themes to the responsible teams.
Benchmarks vary sharply by industry, country and survey method, so comparing a score with another sector says little. The most meaningful comparisons are a company’s own trend over time and competitor scores collected with the same method, often through an independent panel survey.
Cultural response styles also matter. In some countries respondents rarely give the top scores of 9 or 10 even when highly satisfied, which depresses NPS for reasons unrelated to customer experience; multinational comparisons should account for this.
Limitations and common misuses
The single-number format hides distribution changes. A shift of customers from detractor to passive improves loyalty but may leave NPS unchanged if promoters fall at the same time, so the three group shares should always be reported alongside the score.
| Misuse | Consequence | Better practice |
|---|---|---|
| Tying staff bonuses to the score | Survey gaming, begging for high ratings | Reward improvement actions, not scores |
| Reacting to small changes | Decisions based on noise | Check confidence intervals and trends |
| Ignoring verbatim comments | Score without explanation | Code and analyze open answers |
| Surveying only happy customers | Inflated, misleading results | Random or full-population sampling |
| Cross-industry comparisons | False conclusions about performance | Benchmark within the same market |
Used carefully, NPS is a useful directional indicator and a strong prompt for qualitative follow-up, such as interviews or focus groups. It is weaker as a precise performance target or as the sole measure of customer health.
The short version
NPS asks customers one question: would you recommend us to a friend? They answer from 0 to 10. Very enthusiastic people are counted as fans, unhappy people as critics, and the score is simply the share of fans minus the share of critics.
It works like a quick pulse check. A rising score suggests that people like you more, but the real value comes from asking why they gave their answer and fixing what they mention.
Where it makes a difference
A broadband provider runs a quarterly relational NPS survey with about 1,500 responses each time. The score has been stuck around 5 for a year, and leadership wants to know which problems matter most to customers.
The research team codes the open comments of detractors and finds that installation delays appear in a large share of them, ahead of price and speed. Segmenting by tenure shows that customers in their first six months score far lower than long-standing ones.
The company introduces fixed installation time slots and proactive delay notifications. Three quarters later, NPS among new customers rises by about 15 points, the overall score reaches 12, and the change exceeds the survey’s margin of error, which confirms that the improvement is real. Churn data in the same period moves in the same direction.