NPS (Net Promoter Score)
A single-question customer loyalty metric calculated as %Promoters (9–10) minus %Detractors (0–6). NPS ranges from −100 to +100. Above 30 is good, above 50 is excellent, above 70 is best-in-class.
Net Promoter Score (NPS) measures customer loyalty with a single question: "How likely are you to recommend [product] to a friend or colleague, on a scale of 0–10?" Developed by Fred Reichheld at Bain & Company in 2003, it has become the standard loyalty metric across SaaS, retail, finance, and consumer brands.
The NPS formula
NPS = %Promoters − %Detractors
Respondents are classified into three buckets by their score:
| Bucket | Score | Behavior |
|---|---|---|
| Promoters | 9–10 | Actively recommend; drive referral growth |
| Passives | 7–8 | Satisfied but not enthusiastic; vulnerable to competitors |
| Detractors | 0–6 | Unlikely to recommend; may speak negatively |
The score ranges from −100 (everyone is a detractor) to +100 (everyone is a promoter). Passives don't enter the formula but are counted in the survey total.
NPS calculator: a worked example
Say you survey 100 customers and get:
- 60 Promoters (9–10) → 60%
- 25 Passives (7–8) → not counted in the formula
- 15 Detractors (0–6) → 15%
NPS = 60 − 15 = +45
That's a healthy score in most categories.
NPS benchmarks by score range
| Range | Interpretation | Examples |
|---|---|---|
| Below 0 | Serious problem — more detractors than promoters | Cable, legacy airlines often score here |
| 0–30 | Typical for established categories | Banks, telecoms, insurance |
| 30–50 | Good — promoters meaningfully outweigh detractors | Most SaaS products |
| 50–70 | Excellent — strong category position | Apple (historically), Costco |
| Over 70 | Best-in-class — usually category leaders | Tesla, Starbucks Rewards |
Industry NPS benchmarks: comparing inside your category
The absolute number is meaningless without the industry baseline. A +30 in cable beats a +60 in luxury retail. Below is the average NPS by major B2C and B2B category, synthesized from the published industry surveys most often cited in customer-experience reporting (Retently, Customer Gauge, Bain & Company, Qualtrics XM Institute). Use these as a category reference, not a goal.
| Industry / segment | Typical average NPS | Reading |
|---|---|---|
| Consumer technology (smartphones, wearables) | 40–60 | High — strong brand loyalty + category love |
| Streaming / digital media | 35–50 | Sticky behavior, vulnerable to price hikes |
| B2B SaaS (mid-market) | 30–40 | "Good" floor; >50 signals exceptional retention |
| E-commerce / D2C retail | 30–45 | Polarised by shipping + returns experience |
| Professional services / consulting | 40–60 | Relationship-driven; tends to score high |
| Financial services / retail banking | 25–35 | Trust ceiling caps the score even for top performers |
| Insurance | 20–35 | Claims experience dominates the signal |
| Telecom / wireless | 5–25 | Forced-contract baseline; >30 is a category win |
| Cable / pay TV | -5 to 15 | Frequently negative; switching costs mask churn |
| Airlines (legacy) | 0–20 | Highly seat-class dependent; budget carriers often lower |
How to read the table: the relevant question is not "are we above 50?" but "are we 10+ points above our category average?" That delta is the durable signal. A retail bank at +35 is outperforming most of its peers; a SaaS product at +35 is sitting at the category floor.
Methodology caveat: published industry NPS surveys use different sampling, weighting, and "industry" definitions. Treat any single source as one estimate, not the ground truth. The ranges above bracket the answers across three major published surveys, which is more defensible than quoting one to two decimal places.
The follow-up question matters more than the score
The number is a vanity metric without context. The follow-up — "What's the most important reason for your score?" — is where the value lives:
- Detractor responses pinpoint churn risks and product gaps
- Promoter responses identify the "wow moment" that turned them into advocates, which marketing can amplify
- Passive responses reveal what's missing that would push them up to Promoter
Most modern teams treat the score as a tracking metric and the free-text follow-up as the actionable artifact.
When to send NPS surveys
- Transactional NPS — triggered after a key event (purchase, support resolution, onboarding milestone). Captures recency bias but is more actionable.
- Relationship NPS — quarterly or semi-annually to all active users. Captures sentiment drift over time.
Most companies run both. Transactional informs operations; relationship informs strategy.
Common NPS mistakes
| Mistake | Why it happens | Fix |
|---|---|---|
| Treating NPS as a sole indicator | Easy single number to report up | Pair with CSAT + retention + revenue retention |
| Tying comp to NPS | Sounds customer-centric | Incentivizes gaming the survey — use cohorted retention as the comp metric instead |
| Ignoring the follow-up text | The number is easier to summarize | Tag and theme the free-text — that's where the value lives |
| Comparing absolute scores across industries | "Apple has +60, we're at +25, we're failing" | Telecom's +25 may be category-leading; benchmark inside your category |
NPS has critics
NPS has been criticized for:
- Information loss — a 6 and a 0 are both detractors, hiding meaningful differences
- Cultural bias — Americans give higher scores than Europeans; Japanese give lower than both
- Susceptibility to gaming — when used as a compensation lever, teams optimize for the number, not the customer
Most modern customer-success teams use NPS alongside CSAT and retention rather than as a sole indicator.
Related
- CSAT — a tighter-scope satisfaction metric for transactional moments
- Retention — the behavioral signal that complements NPS sentiment
- Churn — the eventual consequence detractor NPS often predicts
- Product-Market Fit — NPS is one of the loyalty signals that surfaces post-PMF
See also
- GlossaryCSAT
- GlossaryRetention
- GlossaryChurn
- GlossaryProduct-Market Fit