What is considered a good NPS score?

Above 0 is good (more promoters than detractors), above +30 is great, above +50 is excellent, and above +70 is world-class. But the only honest answer is industry-relative: +20 is excellent for an ISP and below average for a luxury hotel.

Is an NPS of 50 good?

Yes — an NPS of +50 falls in Bain & Company's 'excellent' band and is in the top quartile for most industries. The exceptions are transactional home services, professional services, and luxury hospitality, where +50 is closer to average and the top operators clear +70.

What is the average NPS score?

There is no single global average, but most synthesized benchmarks land between +25 and +40 for B2C and +25 and +45 for B2B. Sources disagree by 10–15 points in any given industry, which is why ranges are more useful than point averages.

Is a negative NPS bad?

Negative NPS means more detractors than promoters, which is a clear warning sign in most industries. However, telecom, utilities, healthcare payers, and government services routinely run near or below zero as a category — for those industries, the trend matters more than the absolute score.

What is a realistic NPS target for the first year of a program?

If you're starting near or below the industry average, a 10 to 15 point improvement in the first 12 months is realistic when you close the loop on every detractor. Mature programs typically move 5 to 10 points per year. Anything faster than 20 points is usually sample-size noise, not real improvement.

Should I compare my NPS to the global average or my industry?

Your industry, every time. Comparing a healthcare NPS to a SaaS benchmark will make you look bad even when you're outperforming direct competitors. Pick three to five named peers and track relative position instead of an absolute number.

What Is a Good NPS Score in 2026? Bands, Benchmarks & How to Set a Target

The most common question after running an NPS survey for the first time is, "is this a good score?" The honest answer has two parts: a universal interpretation that applies everywhere, and an industry-relative reality that matters far more. This guide covers both, plus a five-step framework for setting a target you can actually defend in a board meeting.

The universal NPS bands

Before any industry context, NPS comes with a built-in interpretation that hasn't changed since Bain & Company published the original methodology in 2003.

Score	Band	What it means
+70 to +100	World-class	Top 1% globally. Strong word-of-mouth growth engine.
+50 to +69	Excellent	Top quartile in most industries.
+30 to +49	Great	Above most industry averages.
0 to +29	Good	More promoters than detractors. Room to improve.
−30 to −1	Needs work	More detractors than promoters — investigate root causes.
−100 to −31	Critical	Serious trust gap. Likely churn pressure.

Why "good" is industry-relative

A +25 NPS for an ISP would be a flagship operating result. The same +25 for a luxury hotel would be a five-alarm fire. The universal bands above are useful for a sniff test, but every serious NPS target has to be set against a relevant peer group.

A few quick illustrations of how dramatically "good" shifts:

Industry	Typical "good" NPS
Home services (HVAC, plumbing, electrical)	+60 to +80
Professional services	+40 to +60
B2B SaaS (mid-market)	+30 to +50
Property management (multifamily)	+10 to +35
ISPs & broadband	0 to +20
Healthcare payers	−5 to +20

Full breakdown across 17 verticals in the NPS benchmarks by industry guide.

Transactional NPS scores higher than relationship NPS

A post-job survey ("how was your install today?") almost always scores 15 to 30 points higher than an annual relationship survey ("how likely are you to recommend us?") at the same company. If you compare a transactional score to a relationship benchmark — or vice versa — you'll draw the wrong conclusion. Always label the question type and compare like-for-like.

Five steps to set a defensible NPS target

Find your industry band. Use the benchmarks above or the full industry table as a starting point. Cross-check with one or two source-specific reports (Retently, CustomerGauge, Qualtrics).
Anchor on the top of the band, not the average. "Average" is the median of every operator in your category, including failing ones. A reasonable 24-month goal is the top quartile of your range.
Set a trend target alongside the score. "Move from +18 to +30 in 12 months" is more actionable than "hit +30." It also stays useful when sample size is too small for confident absolute reporting.
Pair it with a response-rate floor. An NPS goal without a response-rate target is trivially gameable. Most service businesses should aim for at least a 20% response rate on transactional surveys (see our response rates guide).
Track per-segment NPS, not just the company score. Per-location, per-technician, per-product NPS will move long before the company average does. The company score is a lagging indicator of segment-level work.

Common mistakes when judging NPS

Treating the global average as a goal. "Average" includes everyone — including the worst operators in your category.
Reporting a score without sample size. A +60 from 25 responses is noise. Always show n.
Comparing one quarter in isolation. NPS naturally swings 5–10 points quarter to quarter; only rolling trends are reliable.
Ignoring geography. European and Japanese respondents typically score 10–20 points lower than North American respondents on the same experience.
Mixing transactional and relationship NPS. Different cadences, different baselines — never pool them.

Frequently asked questions

Keep reading

• NPS benchmarks by industry — full 17-vertical breakdown
• How to calculate NPS — formula and worked example
• What is NPS? — the full explainer
• Free NPS Calculator