How Do You Benchmark Your NPS Against Your Industry?

You just calculated your Net Promoter Score and got a 38. Is that good? On its own, the number tells you almost nothing. A 38 could be leading in telecommunications and trailing in ecommerce. The only way to know is to compare it against the right reference: your own trend first, then your industry.

To benchmark your NPS against your industry, do four things. Compare your relationship NPS (not a transactional one) against a current, clearly sourced industry average for your exact sector. Use a relative lens, not the universal "good/bad" bar, because what counts as strong varies widely by industry. Where you can, use a neutral third-party benchmark that surveys competitors' customers the same way you survey yours, so you are comparing like with like. Then, and this is the part most teams skip, find out why the gap exists. A benchmark tells you that you are 12 points behind the leader. It does not tell you which themes in the feedback are costing you those 12 points. Thematic closes that gap by quantifying how much each theme moves your score, so a benchmark becomes a prioritized action list instead of a scoreboard.

Below is how to run the comparison properly, a snapshot of reported industry averages, the caveats that make raw comparison unreliable, and how to turn a benchmark gap into something you can act on.

What "benchmarking your NPS" actually means

Benchmarking your NPS means judging your score against a relevant reference point rather than in isolation. There are two references that matter.

• Absolute benchmark. Your score against a universal scale. Bain, which created NPS, puts it at above 0 is good, above 20 is favorable, above 50 is excellent, and above 80 is world-class.
• Relative benchmark. Your score against competitors in your specific industry. This is the lens that matters most, because a score that looks weak on the universal scale can be strong for your sector, and the reverse.

The single most important reference is your own score over time. Track your relationship NPS at regular intervals first. Industry comparison is the second question, not the first.

How to benchmark your NPS against your industry

1. Benchmark on relationship NPS, not transactional NPS. Relationship NPS (rNPS) measures overall loyalty and is run quarterly or annually. Transactional NPS (tNPS) is fired after a specific touchpoint like a support call. Benchmark the relationship score against your industry. Use transactional scores to diagnose which touchpoints are moving it.
2. Find a current, sourced industry average for your exact sector. Use a figure with a named source and a year attached, not a number floating in a slide. Then segment. B2C programs average higher than B2B, so a blended cross-industry number will mislead you.
3. Prefer a neutral competitive benchmark. The most rigorous method, defined by Bain, uses a neutral third party to survey both your customers and your competitors' customers in a double-blind design. That removes the sampling bias you get when a company surveys only its own customers.
4. Compare the drivers, not just the scores. The score gap is the symptom. The reasons your competitors' customers are more loyal live in the open-ended feedback. That is the comparison that tells you what to do.

A snapshot of reported industry NPS averages

Here is a snapshot of average NPS by industry as reported by Retently's 2026 benchmark. Read it as a directional guide, not gospel, for reasons the next section explains.

The same story holds in other datasets, with different numbers. Qualtrics XM Institute has reported grocery around 30 and consumer payments at −6. Bain's NPS Prism put average grocer relationship NPS at 34 in 2026. The disagreement is the point: healthcare shows as 37 in Retently's panel and as a median of 58 in CustomerGauge's B2B data. Pick your sector's figure, cite who reported it and when, and treat it as a range.

Why raw cross-industry comparisons mislead

A benchmark is only as trustworthy as the methods behind the numbers being compared. Most published comparisons quietly break at least one of these rules.

• Survey method has to match. Compare a 0 to 10 email survey against a 1 to 5 in-app prompt and you are not comparing NPS at all.
• Culture skews the score. NPS runs higher in the United States and lower in Japan and France. Some markets skew toward passive 7 and 8 ratings, others toward extreme 9 and 10 ratings, even when real satisfaction is equal.
• Self-selection inflates in-house numbers. A company surveying its own customers gets a friendlier sample than a neutral third party would. That is why Bain built the neutral-intermediary model.
• The published tables disagree. As the healthcare example shows, the "industry average" depends entirely on whose panel you cite and in what year.

There is also a deeper caveat. Research by Keiningham and colleagues, and a later replication by MeasuringU, found NPS is not a better predictor of growth than other satisfaction measures. Comparing a contested number across companies compounds the uncertainty. None of this makes benchmarking useless. It means the raw number deserves less weight, and the reasons behind it deserve more.

How Thematic turns a benchmark gap into an action list

A benchmark gap is a question, not an answer. Knowing you sit 12 points below your industry leader does not tell you what to fix. The answer is in the themes inside your feedback, and Thematic is built to quantify them.

Thematic reads every open-ended response behind your score, groups it into themes, and measures each theme's NPS impact, how many points each is adding or subtracting. Instead of "we are behind," you get "onboarding friction is our single largest detractor driver, and it is the theme separating us from the sector leader." That turns the benchmark into a ranked list of what to fix and in what order. Thematic can also generate a predicted NPS from unstructured feedback such as calls and chats, so touchpoints that never carried a survey still contribute to the picture you benchmark.

The benchmark tells you where you stand. The driver themes tell you why, and what moving up the ranking will take.

What this looks like in practice

• Vodafone New Zealand analyzed the themes inside its NPS and transactional NPS feedback and benchmarked its performance against Vodafone's global counterparts. Theme analysis showed that cross-training staff drove the biggest score lifts. The team saw a double-digit increase in transactional NPS over nine months and saved 60 hours a month on the analysis itself. Telecommunications is a low-NPS industry, which makes knowing the specific drivers more valuable than the headline number.
• Mitre10, the New Zealand hardware retailer, analyzes 20,000 verbatim comments a month across 84 stores. Thematic showed that stock availability alone was taking half an NPS point off its overall score, a driver-level finding precise enough to shape a new trade strategy.
• Atom Bank, the UK digital bank, unified feedback from seven channels across three product lines and acted on the driver themes behind its scores. It became the highest-rated UK bank on Trustpilot, at 4.8 out of 5 across Trustpilot, the App Store, and Google Play, and cut calls on unaccepted mortgage requests by 69%.

In each case the benchmark set the direction. The theme-level drivers set the work.

A benchmarking checklist

Before you present a benchmark to leadership, confirm you can answer each of these:

1. Is this a relationship NPS compared against a relationship NPS, not a transactional one?
2. Does the industry figure have a named source and a year?
3. Is the comparison segmented to your sector and your B2B or B2C model?
4. Were the two scores collected with a comparable survey method?
5. Do you know the top three themes driving the gap, and how many points each is worth?

If you can answer the first four, you have a defensible benchmark. If you can answer the fifth, you have a plan.

The short answer

Benchmark your relationship NPS against a current, clearly sourced average for your specific industry, use a neutral comparison method where you can, and always read the number as a range rather than a verdict. Then go past the number. A benchmark tells you that you are behind. The themes in your feedback tell you why, and which ones to fix first. That is the difference between knowing your score and improving it.