A global survey by McKinsey among 1,993 respondents shows: 88 percent of organizations use AI, but only around one third has left the pilot phase. Only 39 percent report any EBIT effect at all — mostly under 5 percent. A small group of 6 percent "high performers" captures the value. The difference lies not in tool use, but in workflow redesign and leadership accountability.

1. The Finding

  • 88 percent of surveyed organizations use AI regularly in at least one business function, up from 78 percent in the prior year.
  • Around one third of organizations are in the scaling phase; the majority remains in the experimentation or pilot stage.
  • 39 percent of respondents attribute any effect on company EBIT to AI at all; for most of these respondents, the EBIT contribution is under 5 percent.
  • 23 percent report scaling at least one agentic AI system somewhere in the company — mostly in only one to two functions.
  • In marketing and sales, 67 percent of respondents report revenue increases from AI use cases, of which 10 percent report over 10 percent revenue growth.
  • The group of "AI high performers" encompasses around 6 percent of the sample (defined by >5 percent EBIT contribution and self-reported "significant value").
  • High performers redesign workflows 2.8 times more often fundamentally than the rest.
  • Among high performers, 3.0 times more respondents strongly agree that top management takes responsibility for AI initiatives.

2. What This Means for Sales Organizations

The finding is decisive for the management of mid-market B2B companies because it corrects a widespread misunderstanding: AI use and AI impact are not the same thing. When 88 percent "use" but only 39 percent see any EBIT effect at all, then the majority is working with tools that run parallel to the existing sales process — not within it.

In practice this manifests as: sales reps generate email drafts via generative AI, but the pipeline logic, lead scoring, and forecasting remain unchanged. Marketing creates content faster, sales conducts discovery calls as in 2019. The study shows exactly why this generates little impact: the largest explanatory factor for value creation is not the tool, but the redesign of the workflow.

For marketing and sales in B2B, the spread is particularly large: the function ranks among the three with the highest reported revenue effect — but only when AI is integrated into the process, not placed alongside it.

3. The Structural Consequence

The data leads to a concrete requirement for management: the question "which AI tools do we use?" is the wrong question. The right one is: "At which points in our sales and growth process does AI change the way we work — and who carries responsibility for making that happen?"

Three structural consequences:

  1. Efficiency alone is not enough. The study shows that organizations aiming exclusively at cost reduction more rarely realize value than those that also set growth or innovation as an additional goal. An "AI cost reduction project" in sales is structurally underdimensioned.
  2. Workflow redesign is a management task, not an IT task. When the 2.8x higher workflow redesign share among high performers is the strongest differentiator, this is an organizational and process question, not a technology question.
  3. Without visible management commitment, no effect. The 3.0x difference in senior leadership engagement shows: delegating to an "AI lead" is not sufficient. Responsibility for changing the sales mechanics lies structurally at the top.

For mid-market B2B CEOs this means: AI is not an efficiency topic for the sales manager. It is an intervention in the operating model of revenue.

4. The metodic Approach

Instead of layering AI tools on top of an existing sales organization, metodic systematically redesigns the go-to-market system: pipeline logic, forecasting, account development, marketing-sales handoff, and AI agent use case integration are treated as a single process.

This addresses precisely the two levers that the McKinsey study identifies as decisive for value realization: fundamental workflow redesign and sustained management accountability.

Source note: All data is from The state of AI in 2025: Agents, innovation, and transformation, McKinsey & Company (QuantumBlack, AI by McKinsey), November 2025. McKinsey & Company is not affiliated with metodic GmbH. This article paraphrases key findings of the study and links them with metodic GmbH's own assessments. It does not constitute a complete reproduction of the study.