The Gap Is Bigger Than Most Executives Realize
In June 2026, McKinsey published research that quietly reframed how boards should be thinking about AI strategy. The study analyzed 471 privately held, PE-backed companies across 30 countries and 31 industries, and it found that companies which broadly embed AI into their products and offerings trade at a median revenue multiple approximately 130 percent higher than companies that use AI primarily as an internal productivity tool.
One hundred and thirty percent. That is not a rounding difference or a sector-specific anomaly. It is a structural valuation divide, and it is already baked into how investors are pricing companies today.
The finding matters because it runs directly against how most executive conversations about AI are still framed. Most leadership teams talk about AI in terms of efficiency: faster workflows, reduced headcount for routine tasks, better reporting, quicker content production. Those outcomes are real and the cost savings are measurable. But according to McKinsey's data, markets are not rewarding them. The valuation premium is reserved almost entirely for companies where AI changes what they sell, not just how they operate.
Three Levels, One Clear Cliff
McKinsey's framework in the research organizes companies into three levels of AI adoption. Level one is opportunistic use, where teams adopt AI tools as they become available without a central strategy. Level two is integration into operations, where AI is embedded into workflows, functions, and internal processes in a structured way. Level three is embedding AI into what the company actually offers to customers.
The finding that stops most executives is this: the valuation difference between level one and level two is negligible. Markets do not materially reward a company for having built disciplined AI-powered internal operations compared to one that is just dabbling. The premium only appears at level three, when AI is part of the product itself, changing the value the customer receives rather than just reducing the cost to deliver it.
Companies at level three trade at a median revenue multiple of 20x. Companies at level two trade at 14x. That 43 percent gap between levels two and three is where the real money is. And it confirms something that most AI roadmaps are quietly ignoring: operational efficiency alone does not move valuation in a meaningful way.
The Market Is Already Pricing This In
The McKinsey research does not exist in isolation. The same pattern shows up across multiple data sources from the first half of 2026. EY's analysis of industrial M&A deals describes AI as creating a new fault line between what it calls AI-ready assets, which command premiums, and AI-exposed assets, which absorb discounts. The EY CEO Outlook shows that 49 percent of US chief executives cite accelerated AI adoption as their single biggest growth factor for 2026, and the M&A market is beginning to price the convergence between traditional industry and technology company positioning.
PwC's 2026 AI predictions research captures the same divide from the investor side. Only a small number of companies are realizing what PwC calls extraordinary value from AI, including surging revenue growth and significant valuation premiums. The majority are capturing modest efficiency gains that pay for themselves but do not add up to transformation. The technology, PwC's analysis notes, delivers only about 20 percent of an initiative's value. The other 80 percent comes from redesigning what the company does with it.
On the venture side, AI startups captured 81 percent of all global venture capital in Q1 2026. AI seed rounds now run 42 percent above non-AI baselines. That capital is not flowing evenly. It concentrates in companies with proprietary data advantages and AI embedded in their core product, not in companies that are using AI for faster internal operations.
What the Distinction Actually Looks Like
The difference between level two and level three is not always obvious from the outside, and many companies that believe they are at level three have not actually crossed the line. A useful test is asking a single question: if you removed the AI component from what you offer customers, would their experience or outcome change in a material way? If the honest answer is no, because the AI is mostly improving your internal delivery of something the customer was already getting, then you are operating at level two regardless of how the company describes its AI strategy externally.
Companies on the right side of the divide have made AI part of what customers pay for, not just part of how the product is built. That can look like an AI-powered insight layer inside a SaaS product that customers actively use. It can look like a service that delivers outputs customers could not generate on their own without the AI behind it. It can look like a workflow tool that has become functionally irreplaceable because the AI component compounds over time with customer data. In all three cases, AI is the offering, not just the engine behind it.
This distinction is becoming an M&A diligence question, not just a strategy question. The EY and McKinsey data reflects deals that have already closed. Investors examining targets in 2026 are asking directly where the AI sits, whether it changes customer outcomes or only internal costs, and what happens to the product if the AI component is removed. The answers are showing up in multiples.
How KAIDATA Thinks About This
This is the framing we bring to clients who are trying to figure out where they actually sit on that curve. Most companies we work with have meaningful AI activity happening across multiple functions, but the strategic question of whether any of it is changing what they offer versus just how they operate it often has not been asked clearly.
The work starts with an honest audit of where AI is touching the business and what it is changing. Most of that activity sits at level one or level two, which is fine as a foundation but not where the valuation premium is. The more important question is what the company owns, in terms of data, workflows, customer relationships, or operational knowledge, that could become the basis of an AI-embedded offering rather than an AI-assisted internal process.
That is a business strategy question as much as a technology question, and it is one most organizations have not answered with the same rigor they applied to their initial AI adoption. The 130 percent gap is not closing. The window to position on the right side of it is not permanent.
