Wondering if your score is any good? A good NPS score benchmark is your industry’s average, not one global figure, so compare against peers first and your own past results second.
This guide sits within our wider coverage of customer experience metrics.
| Item | Value |
|---|---|
| Overall average in Simplesat’s benchmark | NPS 52 |
| Share of companies scoring 50 or above | 52.3% |
| Highest industry average (Healthcare & Life Sciences) | NPS 74 |
| Lowest industry average (Marketing & Media) | NPS 43 |
| Average for companies with 1,001 or more employees | NPS 35 |
| Range SurveyMonkey calls “good” | NPS 0-30 |
Figures come from Simplesat’s 2026 benchmark report and SurveyMonkey’s NPS benchmarks page. They can change, so check the sources before you rely on them.
How an NPS score is calculated
The score starts with one question. SurveyMonkey describes it as asking, on a scale of 0 to 10, how likely a person is to recommend the company to a friend or colleague. Retently notes that the concept was first introduced in 2003.
In Simplesat’s definition, promoters answer 9 or 10, detractors answer 0 to 6, and the score is the percentage of promoters minus the percentage of detractors, which gives a range from -100 to 100. Everyone in between is a passive.
To work it out by hand, follow the steps SurveyMonkey lays out:
- Sort every response into promoters, passives or detractors and count each group.
- Divide each group’s count by the total number of responses to get percentages.
- Subtract the detractor percentage from the promoter percentage.
SurveyMonkey’s worked example uses 100 customers: 15 detractors, 25 passives and 60 promoters. The sum is 60% minus 15%, so the score is 45.
What score counts as good in general?
The most practical definition comes from SurveyMonkey’s calculation guide: a good score is one higher than your industry average. That makes the peer group the real yardstick.
For a rough all-industry scale, SurveyMonkey offers these labels.
| NPS range | Label |
|---|---|
| -100 to 0 | Needs improvement |
| 0-30 | Good |
| 30-70 | Great |
| 70-100 | Excellent |
Labels come from SurveyMonkey’s NPS benchmarks page; check it before relying on them, as the scale may be revised.
Published averages sit well above the bottom of that scale. SurveyMonkey’s data from over 150,000 organizations gives an average of 32 and a top 25% starting at 72, while Simplesat reports an overall benchmark of 52 from 605,443 responses. By SurveyMonkey’s own labels, 52 would count as great rather than merely good, so a benchmark reflects who was surveyed.
Simplesat’s data also shows the spread: 52.3% of companies scored 50 or above, and a quarter scored 70 or above, as its report puts it.
How much industry changes the answer
Simplesat found industry averages running from 43 in Marketing & Media to 74 in Healthcare & Life Sciences, so the same score can be strong in one context and weak in another. Several of its groups sit in between.
| Industry | Average NPS |
|---|---|
| Healthcare & Life Sciences | 74 |
| Professional Services | 60 |
| Consumer Products | 60 |
| Technology | 45 |
| Marketing & Media | 43 |
Figures are taken from Simplesat’s 2026 benchmark report, which covers responses collected during 2025 and lists more industries than shown here. Check the source for updates.
Other datasets paint a different picture. Retently’s five-year data puts the healthcare average at 34 and Communication & Media at 19, and IBM cites a 2024 Retently study with an average of 80 for insurance and 39 for cloud and hosting.
The same number can mean different things. Retently notes that an NPS of 30 might be the worst result in one industry and place a business among the market leaders in another.

How company size and small samples shift a score
Simplesat reports that companies with 1,001 or more employees average an NPS of 35, which is 17 points below the band before it. A company of that size can trail the overall figure of 52 and still match or beat its size peers.
Sample size matters too. HubSpot warns that your most loyal customers may simply forget to respond while one unhappy customer gives a zero, which can swing a score when responses are few.
To pick the right comparison, follow Simplesat’s guidance:
- Use the benchmark for your own industry when one is published.
- Treat company size as a second reference.
- Where the two disagree, trust the industry figure, because scores vary more by industry than by size.
- Fall back on the overall average only if your industry is not listed.
Using a benchmark without chasing it
Retently’s guide argues that the best benchmark is your own past performance, and that consistent progress is the real win. An external average tells you where you stand; your own trend tells you whether you are improving.
A number alone does not explain itself. SurveyMonkey says open-ended answers uncover the “why” behind a score, while the quantitative data feeds the calculation. The guide to customer satisfaction survey questions covers wording for those follow-ups.
A voice of customer program is one way to route those comments to the teams that can act on them.
When you report a score internally, record three things alongside it:
- The date and number of responses behind the score.
- The peer group you compared it with.
- The main themes in the comments.
Frequently asked questions
Is a negative NPS always bad?
Show answer
A score below zero means detractors outnumber promoters, and SurveyMonkey reads that as a sign of work to do on loyalty. Simplesat found about one in twelve companies in that position.
Is 50 a good NPS?
Show answer
It lands a little under Simplesat’s overall benchmark, yet above the technology average and far below healthcare and life sciences. The answer depends on your peer group.
How often should NPS surveys run?
Show answer
IBM advises running them frequently enough to set a baseline and see whether changes to the customer experience move the score.
Does NPS replace a satisfaction score?
Show answer
No. Simplesat points out that a customer can give a support interaction top marks and still hold back from recommending the company as a whole, so the two measures answer different questions.



