Revenue optimization in a world of AI-powered buyers

Optimizing revenue when AI has flipped the information advantage from seller to buyer.

In brief

• The information asymmetry has now flipped from seller to buyer, challenging not only the sales motion but the revenue engine.
• Dashboards built for 2018 don’t show exposures in 2026. Signals show red flags like stalled deals and no decision but not where revenue is at risk.
• Revenue optimization requires firms to make a deep-dive assessment of not only the sales motion but the full revenue-creation engine.

Artificial intelligence is poised to improve productivity, revolutionize business models and democratize access to knowledge. That knowledge access transforms how we learn, how we manage our finances, how new drugs are discovered, and how both people and businesses make critical decisions. In the business-to-business (B2B) context, it means AI-powered buyers have access to information they used to rely on sellers for. That changes everything for sales teams.

According to the latest Studio+ research report, Defending Revenue when AI is changing everything about the front office, by 2026, 70% of B2B buyers report using a large language model to research a recent purchase, up from effectively zero three years ago.

Your buyer comes prepared with an understanding of your competition, pricing strategy and a comprehensive ROI analysis. What first starts showing up as a sales problem in stalled deals and no decisions is really a macro revenue problem.

Across most B2B sectors, the same patterns are showing up

The new AI-powered buyer

What’s more, the new buyer shows up in force, as a committee of 6 to 15 stakeholders according to the research, each with their own analysis, perspective and opinion formed long before your sales team entered the room. What you read as lack of forecast discipline is really a revenue drift problem disguised as a sales problem. Deals stall. No decisions mount. The one-offs are warning signals for the emerging pattern, but your dashboard wasn’t built to show this.

Revenue drift is what’s really happening

To test whether these were one-off sales patterns or a structural revenue shift, EY Studio+ built the Revenue Drift Map: an analysis of 124 publicly traded B2B companies across 21 sectors, scored over 36 months of public data. The study looked beyond traditional sales metrics and assessed five categories of drift — pipeline, expansion, demand, decision and model — to understand where AI is quietly changing how revenue is created, defended and lost.

The data shows the pattern is already visible. Acquisition costs rose 14% year over year through 2024 and another 18% through Q1 2026. Net revenue retention compressed from 106% in 2022 to 101% in 2025. Forty percent of qualified pipeline now ends in no decision rather than a competitive loss, while 89% of B2B buyers report a deal stalled in the past year. These are not isolated symptoms.

Taken together, they point to a front-office operating model built for a buyer journey that no longer exists.

To understand the drift, we built a map.

The five ways revenue starts to drift

The Revenue Drift Map breaks the problem into five distinct patterns of exposure. Each one shows a different way AI can weaken the revenue engine, and each becomes more dangerous when leaders treat it as an isolated sales, marketing or customer-success issue rather than an end-to-end commercial shift.

The point of naming the five drifts is not to create five separate problems. It is to show where the revenue engine is losing alignment. Pipeline is what leaders can see first. But demand, decision, expansion and model drift are often where the larger, less visible losses are building.

How to improve revenue optimization in an AI-first market

Revenue drift is not a reason to panic, but it is a reason to look differently at the signals already coming through the business. When AI-powered buyers learn faster, compare more deeply and challenge assumptions earlier, the answer is not simply more pipeline activity or tighter forecast discipline. Companies need to step back and examine whether their commercial model still matches how customers now discover, evaluate, decide and expand.

At a high level, that means looking across the full revenue system: how demand is created, how buying groups are engaged, how customer value is measured, how expansion is managed and how pricing or packaging may need to evolve. It also means connecting the dots across sales, marketing, customer success, product and finance so that early signs of drift do not stay trapped in separate functions.

The organizations that move first will not be the ones that treat AI as another sales enablement tool. They will be the ones that understand how AI is changing buyer behavior, where revenue is beginning to drift, and what it will take to defend growth in an AI-first market.