Net Promoter Score: The Complete Guide to Using It in B2B

If you’re looking to implement an NPS in a B2B company, let’s start with the bad news: with 30 responses, your score has a margin of error of 20 points. Just one person switching from promoter to detractor can shift it by nearly 7 points…

That’s no reason to give up on the Net Promoter Score, but you can’t manage it the same way you would in B2C. You’ll never survey 5,000 customers, but you can survey 100% of them: this is exactly the opposite of the problem the NPS was designed to solve, and that’s to your advantage.

In this guide, we cover the calculation, thresholds, and benchmarks—but most importantly, the protocol that transforms an unstable score into a retention system. Here is our complete method.

What is the Net Promoter Score, and how is it calculated?

The Net Promoter Score is based on a single question, rated on a scale of 0 to 10: How likely are you to recommend this company to a colleague or business partner?

The responses fall into three categories:

• 9 and 10: the developers
• 7 and 8: the passive group
• 0 to 6: Detractors

The formula excludes the passives. The NPS is equal to the percentage of promoters minus the percentage of detractors, and the result ranges from -100 to +100.

Here’s an example: out of 80 responses, you have 40 promoters, 28 passives, and 12 detractors. That’s 50% promoters minus 15% detractors, for an NPS of +35.

The point that most articles overlook is the reason for excluding intermediate ratings. It’s not just a methodological quirk. Fred Reichheld found that scores of 7 and 8 were not predictive of anything: a customer who gives an 8 will neither actively recommend you nor leave right away. They aren’t taking any action, so they don’t count in the calculation.

This exclusion has a direct consequence that you will notice very quickly. The NPS is the difference between two percentages, calculated based on only a portion of your respondents. That is what makes it so easy to understand. It is also what makes it so volatile.

Why NPS Doesn’t Work Well in B2B: 3 Methodological Limitations

The NPS was introduced in December 2003 in an article by Fred Reichheld published in the Harvard Business Review under the title “The One Number You Need to Grow.” At the time, Reichheld was a consultant at Bain & Company, and his work focused on mass-market sectors: airlines, retail banks, and telecom providers. These are sectors where surveying 3,000 people poses no problem.

In B2B, you might have 60 clients. Or 400. Rarely 50,000. Three things go wrong when you apply the same approach without adapting it.

Error 1: Your sample is too small for the score to be reliable

An NPS is never a precise measurement; it’s an estimate. Like any survey, it has a margin of error that depends on the number of responses collected relative to the size of your customer base.

We ran the calculation based on a typical B2B distribution—namely, 48% promoters, 42% passives, and 10% detractors—which yields an NPS of +38. 95% confidence interval.

Read the fourth line. A company with 800 customers that receives 100 responses—which is already a good result in B2B—has a score of roughly 12 points. Its NPS of +38 actually falls somewhere between +26 and +50.

Another way to look at it: in a survey round with 30 responses, all it takes is for a single respondent to drop from a 9 to a 6 for the score to fall by 6.7 points. A customer in a bad mood on the day of the survey triggers a crisis meeting.

Discrepancy 2: The person responding is not the person signing

In B2C, the user and the payer are the same person. In B2B, this is almost never the case.

A 2025 Gartner survey of 632 B2B buyers found that purchasing groups consist of between 5 and 16 people, spanning up to four different functions. Forrester, meanwhile, cites an average of 13 stakeholders per purchasing decision.

Who do you send your survey to? The primary contact listed in your CRM. In other words, nine times out of ten, the daily user—the person who is familiar with the tool, has a well-reasoned opinion, and has no say in the renewal decision.

So you’re measuring user satisfaction while the decision is being made elsewhere—by a CFO who has never even opened the application and who is making the final call on the cost. These two realities often diverge. A project with an 80+ user satisfaction rating could very well be scrapped simply because the sponsor has changed jobs and no one internally is advocating for the budget line anymore.

Breakdown 3: The formula treats all your customers equally

In the NPS calculation, a detractor with €250,000 in ARR exactly offsets a promoter with €4,000. One for one.

In B2C, this parity makes sense. In B2B, it’s absurd. In most portfolios, 20% of the accounts generate the bulk of the revenue, and these large accounts often have stricter requirements—and therefore stricter ratings. It’s entirely possible to see your score rise even as at-risk revenue increases.

In fact, this is the main finding of industry surveys: 70% of B2B companies never link their customer experience data to their revenue. They focus on a score, not on risk.

In B2B, the real driver isn’t the NPS score—it’s coverage

Go back to the margin-of-error table and look at the last two rows.

A customer base of 300 with 200 responses yields a score of plus or minus 5 points. The same is true for a base of 60 customers with 55 responses. These two companies have a more reliable NPS than a B2C company that collected 400 responses from 100,000 customers.

This is the key insight to keep in mind. As soon as you survey a significant portion of your total population, the uncertainty drops dramatically. Statisticians call this the correction for a finite population. In practical terms: the small size of your customer base, which seemed to be a disadvantage, becomes an advantage as soon as you increase the response rate.

Except that no one does. The average response rate for B2B NPS surveys hovers around 12.4%, ranging from 4.5% to nearly 40% depending on the company. Mature programs, on the other hand, exceed 60%.

The two figures that should always be displayed alongside the score

An NPS on its own is incomplete information. Always include these two indicators along with it:

• The response rate, expressed as a percentage of the contacts approached. This determines the margin of error—and thus the credibility of the figure.
• The revenue coverage rate, which is the portion of your total ARR represented by the accounts that responded. This is the more useful of the two, and almost no one calculates it.

An NPS of +62 based on 12 accounts that account for 18% of your revenue tells you nothing about the remaining 82%. And in B2B, an account that no longer responds is not a neutral account. It’s often the first sign of disengagement, long before the issue is escalated to the support or a portfolio review.

How to Increase the Response Rate

• Send the survey from the personal email address of the customer success manager or the sales representative in charge of the account—never from a generic address.
• Limit the survey to two questions: the rating, followed by “What would it take to earn one more point?”
• Include the request in the quarterly progress update rather than sending it as a cold email.
• Follow up by phone with strategic accounts that haven’t responded by day 10.
• Explain what you’ll do with the feedback, then actually follow through.

The last point is the one that’s most often overlooked, and it’s the one holding back the second wave. A customer who took five minutes last year and never heard back won’t respond this year.

B2B NPS Survey: Who to Survey and How Often?

Two Decisions to Make…

How many contacts per account

The rule we apply: a minimum of two contacts per account, three for strategic accounts. The daily user, the operational manager who champions the issue internally, and the budget sponsor when one exists.

It’s more work to manage, of course. It’s also the only way to identify the critical difference—the one between a delighted user and a lukewarm sponsor. This difference is a much more reliable indicator of churn than the average account rating.

Be careful of a common mistake: do not average the responses from the same account before consolidating your score. Doing so would obscure the very information you’re interested in. Maintain both levels of analysis: the NPS per respondent for the overall trend, and the internal variation within the account as a risk indicator.

Relational NPS or Transactional NPS

The two scores never overlap. The transactional NPS is structurally higher because it surveys people who have just interacted with you, and most interactions go smoothly. Averaging them would produce a number that means nothing.

If you’re just starting out, begin with the relationship-building aspect. It will show you which touchpoints really matter to your customers, and you can then incorporate transactional elements where it’s worth it.

The Calendar and Its Pitfalls

Three Things to Avoid:

• Immediately after resolving a ticket, if you’re aiming for a relationship score, you’ll be measuring the quality of the support, not the quality of the relationship.
• Three weeks before a renewal. The customer fully understands the intent and responds strategically, sometimes preparing for a negotiation.
• During onboarding. A B2B customer’s perception doesn’t stabilize until after several months of actual use.

For relationship management, a twice-yearly frequency is sufficient in the vast majority of B2B contexts. A quarterly frequency is only worthwhile if your customer base is large enough for variations to stand out from statistical noise—which is rare.

What Is a Good NPS in B2B? Industry Benchmarks

Let’s start with the numbers.

The largest dataset available today is Survicate’s 2025 NPS Barometer, based on 5.4 million responses collected from 599 companies.

It puts the median across all sectors at 42. When breaking down the economic
, the gap becomes clear: 49 for B2C versus 38 for B2B. In the software
, the gap widens even further, with 47 for B2C versus 29 for B2B.

The sector-specific medians below are taken from the same survey.

Now, a word of caution. All of these medians come from a single dataset
, and changing the source will change the result.

Take B2B software, for example: Survicate puts the median at 29, Retently at around 41, and Userpilot’s analysis of 229 SaaS providers places the median in the mid-30s.

A twelve-point gap in the same sector, in the same year. This isn’t a
measurement error; it’s a difference in methodology: different panels, different
response rates, different survey scopes, and a significant portion of
self-reported scores that no one has ever audited.

How to Make the Most of NPS Survey Results

The score isn’t the end result. It’s the starting point. What creates value is what you do within 48 hours of receiving a response.

The Short Loop and the Long Loop

Two mechanisms, two time frames:

• The short loop addresses individual cases. A detractor responds, and the account owner calls them back within 48 hours—not an automated thank-you email, but a phone call. The goal isn’t to boost the rating; it’s to understand and show that we’ve heard them. Programs that uphold this commitment regularly turn detractors into advocates. This is the best return on investment for an NPS program.
• The long loop addresses the root cause. When the same issue arises across eight accounts, it’s no longer a customer relationship issue. It’s a product, pricing, or process issue. It falls outside the scope of customer success and is escalated to a committee, with the ARR amount included as an attachment. It’s this figure that determines the outcome, not the number of mentions.

Prioritize by exposed revenue, not by volume

Here’s how we organize the processing of a wave:

• Detractors in the top 20% of ARR: follow-up call from the account owner within 48 hours; written action plan shared with the client within one week.
• Detractors in the rest of the portfolio: a personalized email from the customer success manager, followed by a call if the response warrants it.
• Passive accounts: These are your most vulnerable and neglected accounts. A passive account is on the verge of switching to a competitor. Treat them as a distinct segment, with an upsell or re-engagement strategy as appropriate.
• Referrers: Ask for a recommendation, testimonial, or customer review, but never in the same email as your thank-you note. Wait a week.

Referrals are the only area where NPS directly generates revenue. A referred customer costs significantly less to acquire than a cold prospect, which lowers your customer acquisition cost, and they stay longer.

The AI prompt for analyzing your verbatim comments

The score accounts for 5% of the value of an NPS survey. The open-ended comments make up the rest. And that’s exactly what nobody ever deals with, because reading 140 open-ended comments and organizing them properly takes a whole day—a day that nobody has.

Here is the prompt you can use. What makes it unique is that it sorts by reported income rather than by the number of mentions, which completely changes the resulting priorities.

The result isn’t a definitive analysis—it’s an initial screening. You save a day’s worth of manual sorting and go into the portfolio review with actual figures, not just impressions.

Connect the NPS to the CRM

An NPS that lives in a spreadsheet doesn’t trigger anything.

The minimum viable setup consists of two fields on your CRM’s company profile: the latest NPS score and the date of the last response, both of which are populated automatically. From there, you create a workflow that generates a task assigned to the account owner as soon as a score of 6 or lower is recorded. It also triggers an alert when an account has not responded in two consecutive rounds.

Most CRMs and customer success tools can do this natively or via a connector. This isn’t a tooling issue—it’s a discipline issue. As long as the score doesn’t create a task assigned to someone with a due date, it remains just a slide.

The Limitations of the NPS and the Metrics That Complement It

Let’s be clear: the NPS does not predict your growth.

Yet that was the initial promise, summarized in the title of the 2003 article. It was never confirmed. In 2007, Timothy Keiningham and his co-authors published a replication of Reichheld’s study in the Journal of the Academy of Marketing Science, using a larger sample and a more rigorous methodology.

Their conclusion: NPS does not predict growth any better than other satisfaction metrics. The article received the H. Paul Root Award from the Marketing Science Institute, but that hasn’t stopped NPS from becoming the global standard.

More recently, several studies have shown that NPS correlates poorly with gross retention—that is, with the pure renewal rate. It performs significantly better when it comes to net retention, which takes expansion into account. This makes sense when you think about it: a promoter isn’t someone who just stays—it’s someone who buys more and brings in new customers.

The most interesting thing is that Reichheld himself published a correction. In 2021, in an article titled “Net Promoter 3.0,” he acknowledged that self-reported, unaudited scores had become vanity metrics and proposed a complementary metric derived from accounting rather than surveys: the Earned Growth Rate. His formula combines net revenue retention and the portion of growth coming from customers acquired through referrals.

Take a look at this formula. It’s a B2B metric. Twenty years after inventing the NPS, its creator proposes supplementing it with what B2B companies are already measuring.

We’re not suggesting you abandon the NPS. It has a quality that few other metrics share: it fits into a single question, it can be understood in three seconds during an executive committee meeting, and it provides a scalable reason to start a conversation with a customer you wouldn’t have called otherwise. That’s already a lot.

But treat it for what it is. A trigger, complete with a margin of error, to be displayed alongside its coverage rate and interpreted in conjunction with more concrete metrics. The day your NPS is no longer used to comment on a trend but to decide whom to call back on Monday morning, it will have started to serve a purpose.