Customer loyalty is easier to manage when it can be measured consistently. NPS gives organizations a simple way to track willingness to recommend, identify dissatisfied customers, and monitor whether customer experience changes are improving brand perception over time.
NPS definition
Net Promoter Score (NPS) is a customer experience metric that measures how likely a customer is to recommend a company, product, or service to others. Developed by Fred Reichheld and introduced with Bain & Company and Satmetrix in a 2003 Harvard Business Review article, NPS distills customer sentiment into a single, repeatable number that organizations can track over time. Unlike one-off satisfaction surveys, NPS is designed to be measured regularly, creating a baseline against which future changes in the customer experience can be evaluated.
The NPS question and the three response categories
A standard NPS survey asks: “On a scale from 0 to 10, how likely would you be to recommend our company to a friend or colleague?” Many organizations follow this with an open-ended question, such as "What's the main reason for your score?", to capture qualitative context behind the number.
The NPS scale divides responses into three groups:
- Promoters, 9–10: Loyal customers who are enthusiastic and likely to recommend the brand.
- Passives, 7–8: Satisfied but unenthusiastic customers who may be vulnerable to competitors.
- Detractors, 0–6: Dissatisfied customers with a higher risk of churn and negative word of mouth.
This classification separates genuine advocacy (promoters) from mere satisfaction (passives) and active dissatisfaction (detractors) — a distinction that a single average satisfaction score doesn't capture.
How to calculate NPS
The NPS formula is:
NPS = % Promoters − % Detractors
Passives are counted in the total respondent base but excluded from the subtraction itself. The result is expressed as an integer (not a percentage) ranging from −100 to +100.
Worked example: A company surveys 100 customers and receives:
- 65 Promoters
- 20 Passives
- 15 Detractors
Step 1: Convert each group to a percentage of total respondents.
- Promoters: 65 ÷ 100 = 65%
- Passives: 20 ÷ 100 = 20%
- Detractors: 15 ÷ 100 = 15%
Step 2: Subtract the detractor percentage from the promoter percentage.
- 65% − 15% = 50
Step 3: Report the result as a plain number.
- NPS = +50
Passives (20%) don't factor into the final subtraction, but they remain part of the 100-respondent base used to calculate the promoter and detractor percentages in the first place.
What is a good NPS score?
A score above 0 is considered good, above 20 favorable, above 50 excellent, and above 80 world-class. However, an NPS benchmark should be interpreted within industry context. The NPS score meaning becomes more useful when performance is compared with organizations facing similar customer expectations and operating conditions.
These thresholds are a useful starting point, but industry context matters more than the absolute number. A score of 30 might be strong in a sector where competitors average 10, and mediocre in a sector where competitors average 50. NPS also shouldn't be compared across companies of very different size, geography, or business model without adjusting for those differences.
Industry NPS benchmarks provide useful context, but the figures should be treated as directional rather than universal. Scores can vary substantially between providers because datasets differ by geography, company mix, survey methodology, respondent type, and year.
Industry benchmarks are useful for context, but they're less standardized than they're often presented to be. Reported averages for the same industry can vary by 20–40 points between benchmark providers and even between different years of the same provider's report, because sample composition, survey methodology, and respondent pool all differ. For example, one widely cited Retently dataset put insurance's average NPS at 80 in one report and 74 in another; a separate report placed it at 46. Cloud & hosting has been reported anywhere from roughly 30 to 39 in the same period.
What holds up consistently across sources, rather than the exact figures, is the general pattern:
- Insurance, financial services, and B2B professional services tend to sit toward the higher end of the NPS range.
- Cloud & hosting, telecom, and airlines tend to trend lower, often reflecting more complex, ongoing service relationships rather than one-time transactions.
- B2C companies generally report higher average NPS than B2B companies, largely because B2B purchasing involves longer sales cycles, multiple stakeholders, and a more conservative scoring culture among professional buyers.
Since the specific numbers shift meaningfully year over year and by provider, organizations benchmarking their own score should pull current data directly from a benchmark source such as Retently, CustomerGauge, Bain's NPS Prism, or the Qualtrics XM Institute, rather than relying on a static table.
Types of NPS surveys
Survey timing changes what the score tells the business. The relationship NPS vs transactional NPS distinction is particularly important when separating long-term loyalty from feedback on individual interactions.
Relationship NPS captures long-term brand perception, independent of any single interaction. Transactional NPS is triggered by a specific event, such as a purchase, a support ticket resolution, an onboarding step, and is more useful for diagnosing friction at a specific touchpoint than for judging overall brand health.
NPS vs CSAT vs CES
These metrics answer different customer experience questions and can be used together rather than treated as substitutes.
Advantages and limitations of NPS
The main strengths of NPS are simplicity, repeatability, actionable customer segmentation, and standardized benchmarking. It can also flag detractors who require attention.
Its weaknesses matter equally. A single score lacks context; satisfied customers may not actively recommend brands, scoring behavior can vary across cultures or regions, and excessive survey frequency can create survey fatigue.
How to improve your NPS
Improvement starts with acting on the feedback behind the score. Close the loop with detractors, investigate what keeps passives from becoming promoters, and analyse open-ended responses for recurring friction. Segment feedback where useful, share relevant comments with responsible teams, and establish clear ownership for addressing root causes and monitoring changes.
Top NPS software tools
Your supplied material identifies CRM integration, automated distribution, trigger-based workflows, low-score alerts, and AI-enhanced sentiment analysis as useful evaluation criteria. Specific pricing and product-level feature details were not provided.
Frequently Asked Questions
What is a good score?
Above 0 is good, above 20 favorable, above 50 excellent, and above 80 world-class. Industry context should also guide interpretation.
How is it calculated?
Subtract the percentage of detractors from the percentage of promoters. Passives remain in the respondent total but not the subtraction.
What is the difference between NPS and CSAT?
NPS measures loyalty and recommendation intent. CSAT measures satisfaction with a particular experience.
How often should surveys be sent?
Relationship surveys may run quarterly or semi-annually. Transactional surveys are triggered after specific interactions.
What are promoters, passives, and detractors?
Promoters score 9–10, passives 7–8, and detractors 0–6.
How accurate is NPS?
It provides a useful snapshot but needs qualitative follow-up because the score alone does not explain customer reasoning.
What does a negative score mean?
It means detractors outnumber promoters, signalling customer experience and retention risks that warrant investigation.
What is a world-class score?
A score above 80 is classified as world-class in the framework supplied for this guide.