The Price of Getting AI Wrong in B2B Sales
Twenty-five B2B revenue leaders walked into a room in London last month having already named the lies. The week before, three workshop groups had identified the assumptions holding their sales motions together that weren’t holding up under scrutiny. This time the question was different. Not what are we getting wrong, but what is it costing us?
Three prices emerged. All three were recognised immediately around the room. None of them showed up where people expected – in the obvious metrics, the headline numbers, the reports that land on Monday morning. They showed up in the gaps. In the deals that stalled without explanation. In the pipeline that looked healthy and wasn’t. In the deal sizes that were consistently smaller than they should have been.
Price 1: The Discovery Gap
By show of hands across the first group, 40-60% of forecasted deals end in no decision. Not lost to a competitor. Not disqualified early. Worked, researched, followed up – and then nothing. The group had a clear explanation for it.
AI has made sellers better prepared than ever. It has not made them better at selling. The tools are doing the preparation, capturing the conversations, and producing the next steps. What they are not doing is the cognitive work that actually moves a deal forward – understanding what a champion needs to win an internal argument, and giving them something that survives after the seller leaves the room.
That work has atrophied. Sellers aren’t uncovering the economic buyer’s justification for change. They aren’t building the internal case the champion needs to win the decision. They are arriving better briefed, with cleaner notes and faster summaries, and still stalling at the same rate.
Gartner predicts that through 2026, atrophy of critical-thinking skills due to GenAI use will push 50% of global organisations to require AI-free skills assessments. The room arrived at the same conclusion independently. The tools got better and the thinking got worse.
The 40–60% no-decision rate isn’t a pipeline problem. It’s a discovery problem. And AI, deployed without structure, is making it worse.
Price 2: The Performance Gap
The second price is harder to see – which is precisely what makes it dangerous.
AI has not closed the performance gap between top sellers and everyone else. It has hidden it. Top performers are using AI to sharpen their judgment and pull further ahead. The middle of the team is automating the wrong behaviours and stagnating. And because CRM fields are now populated with well-constructed AI summaries, the stagnation looks like activity from above.
Roy Wood, CRO at OneAdvanced, described what he observes across his team: “Maybe 10–20% of the workforce really spend time and grasp AI. Then you’ve got a middle pack – it’s the new Google to them. And then you’ve got the others.”
Three groups. Three trajectories. One pipeline report that cannot tell them apart.
The systems that were supposed to give leaders visibility have become the reason they’ve lost it. AI tools can’t tell the difference between a conversation that moved the deal forward and one that didn’t. The data looks clean. The performance gap keeps widening.
The Deal Size Problem
The third price is the most direct and the most consistently felt. When value arrives late in a deal, deal size shrinks. The group named this finding three times before anyone had to prompt it.
The mechanism is straightforward. When value enters the conversation late, it looks like a closing tactic rather than a consultative contribution. Buyers get suspicious. They stop sharing the information needed to build a proper business case. Discovery stays surface level. The economic buyer never fully engages. And the deal closes smaller – not because the product delivered less value, but because the value case was never properly built.
Forrester’s 2026 B2B predictions put a number on the broader cost of this pattern: ungoverned use of generative AI will result in the loss of more than $10 billion in enterprise value. The room’s version of that finding was more specific. In deal after deal, value selling is being bypassed not because it doesn’t work, but because the sales motion makes feature-led selling the path of least resistance. The result is a business case built on assumptions rather than on what the buyer actually needs to justify the decision internally.
The thread running through all three
Each of these prices is different. The discovery tax shows up in the no-decision rate. The performance tax shows up in a pipeline report that looks healthier than it is. The value tax shows up in deal sizes that are consistently smaller than they should be.
But they share a common cause. AI is being deployed for efficiency without being designed for effectiveness. Efficiency without structure isn’t a productivity gain. It’s a set of problems that are harder to see because the data says everything is fine.
The room didn’t leave with a clean set of answers. But it left with a clearer picture of where to look – and a shared recognition that the price of getting this wrong is already showing up in the numbers.
The full event report here, including what the room said the lies and shift needs to look like.
