Net Promoter Score (NPS) measures customer loyalty by asking one question: how likely are you to recommend us? Also called Customer Net Promoter Score (cNPS), it produces an integer between -100 and +100 by subtracting the percentage of detractors from promoters. [As of 2020, two-thirds of Fortune 1000 companies use versions of the metric] (Fortune), making it a standard benchmark for predicting growth and flagging churn risks.
What is Net Promoter Score?
Fred Reichheld developed NPS at Bain & Company and introduced it in the 2003 Harvard Business Review article "The One Number You Need To Grow." The methodology centers on a single survey item: "How likely is it that you would recommend [Organization X/Product Y/Service Z] to a friend or colleague?" Respondents rate their likelihood on an 11-point scale from 0 (not at all likely) to 10 (extremely likely).
The final score is expressed as an integer, not a percentage. It ranges from -100 (if everyone is a detractor) to +100 (if everyone is a promoter). A positive score indicates that promoters outnumber detractors; a negative score signals the opposite.
Why Net Promoter Score matters
NPS correlates with business outcomes when tracked consistently and paired with follow-up actions. Key benefits include:
- Predicting revenue growth. [Sustained value creators achieve Net Promoter Scores two times higher than the average company] (Net Promoter System). [Leaders in the Net Promoter System grow at more than twice the rate of competitors] (Net Promoter System).
- Quantifying word of mouth. [Promoters account for more than 80 percent of referrals in most businesses] (Net Promoter System). [Detractors account for more than 80 percent of negative word of mouth] (Net Promoter System). [Passives show repurchase and referral rates as much as 50 percent lower than those of promoters] (Net Promoter System).
- Closing the perception gap. Internal teams often overestimate customer sentiment. [While 49 percent of CX professionals believe satisfaction has improved, only 18 percent of consumers agree] (SurveyMonkey).
- Benchmarking by industry. Scores vary widely by sector. [Research indicates the grocery industry averages 30, video streaming averages 29, and consumer payments averages negative six] (Qualtrics). Compare your score against sector-specific norms rather than universal targets.
How Net Promoter Score works
Calculating NPS requires three steps:
- Collect responses. Ask the ultimate question on a zero-to-ten scale.
- Categorize respondents. Sort into three groups:
- Detractors (0–6): Unhappy customers who damage growth through negative word of mouth and churn.
- Passives (7–8): Satisfied but unenthusiastic customers vulnerable to competitive offers.
- Promoters (9–10): Loyal enthusiasts who fuel growth through repeat purchases and referrals.
- Calculate the score. Subtract the percentage of Detractors from the percentage of Promoters. Passives count toward the total response pool but do not enter the calculation directly. For example, with 60 Promoters, 20 Passives, and 20 Detractors out of 100 responses, your NPS is 60 minus 20, equaling 40.
Survey structure. Pair the rating question with open-ended follow-ups. Ask "What is the primary reason for your score?" to identify drivers. Ask "How can we make your experience better?" to surface specific fixes. Request permission to follow up with the respondent to enable closed-loop workflows.
Relational versus transactional. Relational NPS surveys run periodically (quarterly or annually) to gauge overall relationship health. Transactional surveys trigger after specific interactions like purchases or support calls. Qualtrics advises against transactional use because granular deployment risks annoying customers and conflating momentary friction with relationship failure.
Types of Net Promoter Score
While the core calculation remains constant, three distinct applications exist:
Relational NPS (rNPS). Deployed at regular intervals to track long-term loyalty trends and year-on-year relationship health. This acts as a barometer for the entire customer relationship.
Transactional NPS (tNPS). Sent immediately following specific touchpoints. While it captures interaction-level sentiment, critics note it violates the original intent of measuring overall relationships rather than isolated moments.
Employee Net Promoter Score (eNPS). Asks staff how likely they are to recommend the workplace. Reichheld recommends against linking eNPS to compensation or disciplinary actions, as this creates perverse incentives. Qualtrics argues eNPS lacks the complexity needed for employee experience measurement and suggests full engagement surveys instead.
Best practices
- Survey the right people in B2B. Apply the 3x3 model: target three decision-makers, three influencers, and three end users per account to avoid skewing results toward the largest user population.
- Close the loop within 48 hours. Contact Detractors to resolve issues. This retention tactic often converts critics into Promoters.
- Analyze drivers, not just scores. Segment responses by touchpoint (purchase, support, delivery) to identify which interactions lift or lower loyalty.
- Track trends, not absolutes. Compare your score against previous periods and direct competitors. [A score of 30 is excellent in consumer payments but mediocre in grocery] (Qualtrics).
- Treat non-respondents as Detractors. An absence of signal often indicates disengagement. Including them in analysis prevents false optimism.
- Keep surveys short. Three questions suffice: the rating, the reason, and the improvement suggestion. Additional questions reduce completion rates.
Common mistakes
- Including Passives in the calculation. Passives represent 7–8 scores. They do not factor into the math, though you should monitor their migration toward Promoter or Detractor status.
- Surveying too frequently. Reichheld describes over-surveying as a tragedy of the commons. Excessive requests for ratings train customers to ignore surveys or answer randomly.
- Using averages instead of categories. Do not average raw scores. Stick to the discrete 0–6, 7–8, 9–10 buckets to maintain benchmark comparability.
- Weighting all responses equally in B2B. If one account sends fifty responses and another sends five, the larger account skews results. Calculate NPS per account first, then average across accounts, or use the 3x3 model to standardize inputs.
- Ignoring the open text. The numeric score indicates what is happening. The verbatim explains why. Analyze text for recurring themes to prioritize fixes.
- Linking scores to employee bonuses. This incentivizes gaming the system, encouraging employees to beg for high scores rather than improve service. Reichheld explicitly recommends decoupling NPS from compensation.
Examples
- Scenario: SaaS company. One hundred responses break down as 50 Promoters, 20 Passives, and 30 Detractors. Subtract 30 percent Detractors from 50 percent Promoters for an NPS of 20.
- Scenario: Industry comparison. A grocery chain scores 30, matching its industry average. A consumer payments company scores negative six, also matching its average. Both perform at par despite the sixty-point gap.
- Scenario: B2B technology. A vendor surveys three contacts at Client A (all Detractors) and three at Client B (all Promoters). They calculate NPS per account (Client A: negative 100; Client B: positive 100) then average to zero, revealing the true relationship split rather than blending into a misleading aggregate.
NPS vs Customer Satisfaction (CSAT)
NPS and CSAT measure different facets of the customer experience. Choose the metric that aligns with your specific goal.
| Goal | NPS | CSAT |
|---|---|---|
| Primary use | Measure loyalty and likelihood to recommend | Gauge immediate satisfaction with a specific interaction |
| Timing | Periodic relationship health checks | Post-interaction or post-support touchpoints |
| Scale | 0–10, converted to -100 to +100 | Usually 1–5 or satisfied/neutral/dissatisfied |
| Predictive power | Indicates future retention and growth | Reflects past performance on a specific event |
| Key risk | Lacks detail without driver questions | Does not predict future buying behavior |
Choose NPS to benchmark overall brand health and predict revenue. Choose CSAT to diagnose the quality of a specific transaction or support ticket.
FAQ
What is the formula for Net Promoter Score? Subtract the percentage of Detractors (scores 0–6) from the percentage of Promoters (scores 9–10). The formula is NPS equals percent Promoters minus percent Detractors. Passives (7–8) are excluded from the numerator but included in the total response count.
What is a good Net Promoter Score? A good score exceeds your industry average. Bain suggests above zero is good, above twenty is favorable, above fifty is excellent, and above eighty is world-class. However, benchmarks vary significantly by sector. Compare against direct competitors and track your own trend lines rather than chasing absolute numbers.
How often should I survey customers? Run relational surveys quarterly or annually to track trends. Avoid surveying the same customer more than once per quarter unless triggering a transactional survey, which Qualtrics generally advises against. Frequent surveying causes respondent fatigue and degrades data quality.
Why are Passives excluded from the calculation? Passives neither strengthen nor weaken growth momentum. Their repurchase and referral rates run roughly 50 percent lower than Promoters, placing them in a neutral zone. Excluding them sharpens the contrast between advocates and critics, creating a clearer signal for action.
Can I use NPS for employee feedback (eNPS)? You can, but experts caution against it. eNPS lacks the depth of full engagement surveys and may create liability if tied to compensation. Reichheld recommends tracking referrals and retention through other means rather than relying on a single eNPS question.
How do I analyze open-ended feedback at scale? Use text analysis tools to categorize verbatim responses into themes. Look for recurring drivers associated with Promoter scores (strengths to maintain) and Detractor scores (friction points to fix). This transforms the numeric signal into actionable fixes.
Should I include demographic questions in my NPS survey? Only if you cannot source that data elsewhere. Minimizing survey length improves completion rates. If you already know a respondent's industry or company size from your CRM, skip the demographic screen.
What is the difference between relational and transactional NPS? Relational NPS measures overall relationship health at regular intervals. Transactional NPS triggers after specific interactions. Qualtrics recommends relational deployment because NPS works best as a relationship metric, not a transaction-level diagnostic.
