Share article
At customer experience conferences over the past few years, the same strange scene keeps playing out. Managers at large B2B firms complain that their CX (customer experience) scores refuse to budge. They are investing more than ever. They are hiring specialists. They are buying platforms. And still nothing moves.
McKinsey’s data shows that the average B2B CX score has long been stuck below 50 out of 100, while consumer brands routinely sit between 65 and 85. A gap that has barely narrowed over the past decade.
The explanation seems obvious: B2B is simply different. More complex, slower, less emotional. That explanation is correct. It just isn’t enough. The real question isn’t why B2B is more complicated. It’s why everything we have known about that complexity for forty years still hasn’t filtered through to the tools we use to measure it.
Seven people, one signature
Let’s start with a number that changes everything. Long-running Gartner research shows that the average B2B purchase decision involves six to ten people. Each of them independently consults four to five of their own information sources. And here comes the genuinely uncomfortable figure: B2B buyers spend roughly 17 per cent of the entire purchase cycle talking to suppliers. When they’re comparing several offers, that number drops to five or six per cent per supplier.
In plain English: the supplier gets to speak with the decision-making group for only a sliver of the process. Most of the time, the group decides without them. And that group is far from uniform. The classic buying centre model, described by Thomas Bonoma and Benson Shapiro in the Harvard Business Review back in 1981, identifies six distinct roles inside a single company: someone initiates the purchase, someone will use the product, someone recommends it, someone approves it, someone pays, and someone watches the risks. Each one sees the same thing through an entirely different lens.
The initiator is solving a problem. The user cares about productivity. The CFO cares about price and return on investment. The lawyer cares about risk. And none of the metrics most companies use today — NPS (Net Promoter Score, willingness to recommend), CSAT (satisfaction with a specific interaction) or CES (how much effort the customer had to put in) — was designed with any of this in mind. All three were born in the B2C world, where the decision-maker, the user and the buyer tend to be the same person.
The double blind spot of NPS
Fred Reichheld, the creator of NPS and a partner at Bain & Company, has flagged this problem himself. In his book The Ultimate Question 2.0, he admits that NPS in B2B suffers from what he calls a double blind spot. The customer knows their sales rep. The supplier knows their account manager. But neither of them sees the full relationship between the two organisations. A score from a survey sent to one contact then often says more about how well that particular pair works together than about the health of the business relationship as a whole.
Forrester’s B2B Customer Experience Index repeatedly documents the same thing: in B2B, the link between one respondent’s NPS and the actual likelihood of contract renewal is far weaker than in the consumer world. The reason isn’t methodology. It’s what we’re actually measuring. B2C measures the experience of an individual. B2B sells to an organisation. These two things are not the same.
Picture a familiar situation. A satisfied champion on the customer’s side may lose their position or leave, taking with them a business relationship that looked rock-solid in the reports. Conversely, a frustrated user with a low NPS may mean very little if the CFO views the supplier as a strategic partner. In both cases the metric fails, because it measures a voice rather than a relationship.
The five per cent we see. And the 95 we don’t.
The second blind spot is time. In 2021, Professor John Dawes of Australia’s Ehrenberg-Bass Institute published an observation that has since become folk wisdom among B2B marketers as the 95/5 rule. At any given moment, only around five per cent of potential B2B customers are actively in buying mode. The remaining 95 per cent aren’t requesting quotes, aren’t comparing options, aren’t choosing anything. Yet their perception of the brand is still forming, because the purchase decision will come in two, five, sometimes ten years.
This is awkward for CX strategy. Most B2B customer experience programmes today measure only what happens inside an active relationship. After the contract is signed, during implementation, at renewal. In other words, in exactly that fifth or tenth of the time when the customer is genuinely engaged. The rest of the period, when the relationship is being maintained and when the question of whether to even invite the supplier to the next tender is being decided, is handled by marketing and sales. Usually with no connection whatsoever to CX data.
The key point gets overlooked: in B2B, customer experience doesn’t only happen in moments of contact. It happens in the organisation’s memory. And that memory has its own inertia, its own turnover, and its own myths about “who’s good to work with and who isn’t”.
What the successful firms actually do
A 2023 McKinsey study tracked B2B firms with above-average customer experience results. They share one common trait. And it isn’t a bigger budget or a better survey. It’s the way they define the unit of measurement.
Instead of NPS from one respondent, they track what’s known as an account health score. This is a composite indicator that combines several inputs at once: feedback from different roles within a single client (typically the user, the economic decision-maker and the sponsor), behavioural data on how the product is actually being used, financial indicators from the contract, and qualitative assessment from the account manager. None of those inputs is new on its own. What is new is that none of them is considered sufficient on its own.
The same study shows that B2B firms with above-average CX grow their revenue roughly twice as fast as the average for their industry. The correlation is strongest in companies that have separated measurement of the relationship (at the account level) from measurement of the transaction (at the level of a single interaction). Most B2B organisations still mix the two together. And then they wonder why their NPS is rising while customers are leaving.
What is actually being sold
Perhaps the most interesting shift currently underway in B2B CX is a shift in thinking. For years we were trained to see customer experience as the sum of individual touchpoints: every email, every ticket, every meeting. That view is a legacy of B2C thinking, where it makes sense. A consumer really does decide from contact to contact.
In B2B, however, the main thing a supplier delivers is the relationship itself. Software, service, component. All of that is, in a sense, just the physical carrier. What the customer is really buying is predictability. Certainty that the other side will respond, learn, deliver. And that’s precisely the quality no classic metric captures.
So the real question for anyone running a B2B CX programme today isn’t which metric to add next. It’s a different one. Do my tools actually measure what really matters in a B2B relationship? If the answer is “partially” — and for most firms that’s exactly how it sounds — then perhaps it’s time to stop tuning the survey and step back. To the question of what, and whom, we’re really measuring.
Because in B2B there is no single customer. There is an organisation. And organisations cannot be measured by an average.
