The 7.2x Gap, and What It Actually Measures
Your competitors aren't using better AI than you. They're using it inside a better company.
Most leadership teams walking into Q2 2026 are squinting at the same picture. AI is everywhere in the building. Tools deployed, budgets approved, four use cases the CEO can recite from memory. And yet the impact is lumpy. A few teams sprint. Most jog. Some have quietly returned to spreadsheets and prayer.
Money in is real, value out is fuzzy. The instinct is to read the gap as a technology problem and write another cheque. That instinct is wrong. And it’s expensive.
April produced an unusually loud stack of research on this exact gap. Read alone, each report is forgettable. Read together, they tell you something your finance team won’t.
PwC surveyed 1,217 senior executives across 25 sectors and found that AI leaders deliver 7.2x more revenue and efficiency than their peers. Seventy-four percent of the economic value sits with the top 20% of organisations. That’s not a bell curve. That’s a Pareto on steroids, the kind of distribution you normally see in venture capital and Premier League goal-scoring.
In the same month, Deloitte reported that only 25% of organisations have moved 40% or more of their AI pilots into production. Stanford HAI confirmed adoption has hit 88%. Stack those numbers. Almost everyone is “doing AI.” Almost nobody is finishing it.
The distance between adoption and value is not the technology. It’s the operating model.
It’s not a budget problem. You wish it was.
If money were the answer, this would be solved already.
BCG found companies plan to roughly double AI spend in 2026, hitting around 1.7% of revenues. Bain found 42% of CFOs plan to lift AI investment by more than 30% within two years. Eighty-three percent are increasing it by at least 15%. Capability isn’t scarce either, vendors are stacked three deep at every conference.
So what separates the winners? Boring stuff. The kind of thing that doesn’t make a press release.
PwC’s leaders do three things their peers don’t.
They redesign workflows around AI, instead of bolting AI onto the workflows they already had. (Automating a broken process gets you a faster broken process.)
They push real decision-making down to the operational layer.
And they build the unsexy plumbing, data, governance, trust, so AI outputs can be acted on without a human re-checking everything five times.
None of this is a technology choice. It’s all how the company is wired. Which is precisely why most leadership teams avoid it.
The CEO is at boot camp. The board is at brunch.
The most engaged CEOs now spend more than eight hours a week on their own AI upskilling, per BCG. Meanwhile, KPMG and INSEAD report that nearly three quarters of corporate boards have only moderate or limited AI expertise.
Translate that. The chief executive is in the gym at 6 a.m. The body that is meant to hold them accountable is asking what an LLM is over coffee. This is how investments get approved without ever being interrogated.
A board sees a credible budget, a credible roadmap, a credible vendor list. What a board does not see, in most cases, is whether approval pathways have been redesigned, whether decision rights have moved closer to the work, or whether the workforce has been re-architected to handle what the technology now enables. Those are the choices that produce the 7.2x edge. They are also exactly what an AI-illiterate board cannot test.
Investment without architectural scrutiny is the most expensive form of investment a leadership team can make in 2026. It’s a Tesla with no brake pedal.
Pilots are theatre. Production is the audit.
That 25% pilot-to-production figure from Deloitte deserves a moment.
Pilots succeed because they’re built to succeed: narrow scope, motivated team, protected from the usual organisational immune system. Production is where reality returns the calls. Data ownership becomes a turf war. Approval pathways still assume someone will print it, sign it, and file it. Roles are defined as if the year is 2014.
The pilot proved the model works. Production exposes that the company isn’t built to act on it.
This is why the 7.2x gap keeps widening as adoption rises. Most organisations are succeeding at AI on a contained scale and crashing into their own org chart the moment they try to scale it. Stanford HAI’s 2026 AI Index reaches the same conclusion from a different angle.
Your workforce is not the obstacle. Your org chart is.
BCG dropped two pieces in April that, read together, explain why this gap doesn’t close on its own.
The Henderson Institute estimates 50 to 55% of US jobs will be substantially reshaped by AI within two to three years. Only 10 to 15% face full displacement, and over a longer horizon. The headline isn’t “the robots took my job.” It’s “the robots changed my job, and HR hasn’t updated my job description in eighteen months.”
The companion piece, AI Transformation Is a Workforce Transformation, makes the simple argument most companies are still ignoring. The technology only pays out when paired with workforce redesign. Roles, responsibilities, and capabilities have to move at the speed AI is changing them. Most organisations have no mechanism for this. Reskilling happens in pockets. Role redesign happens after the fact. The gap between what AI can do and what your people are configured to do compounds quarterly. Like credit card debt, but with more PowerPoint.
A reflection from the field
In our work at Zeroto100 with banks, telecoms, and industrial groups across Europe, the Middle East, and Africa, we see this pattern with almost no variation.
A bank we worked with recently arrived sure they had an AI capability problem. The diagnostic surfaced something else. Most of their AI spend was funding pilots in functions whose approval pathways had not been touched in years, and whose workflows still assumed a human checkpoint at every stage. They were paying Ferrari prices to win drag races on a dirt road.
We used to think cases like this were unusual. They aren’t. Companies arrive describing their AI challenge in technical or budgetary terms and discover, halfway through, that they have an operating-model problem wearing a capability costume. Capability is usually present in pockets. What’s missing is the architecture that turns capability into throughput. That architecture rarely shows up in a dashboard. Which is why most transformation programmes spend their first year building the wrong thing.
Three questions, before you write the next cheque
The honest question for senior teams right now isn’t whether they’ve adopted AI. They have.
The honest question is whether the operating model has been redesigned with the same seriousness as the tech investment.
Three tests stand in for the bigger question.
Has any approval pathway been redesigned around the decision pace AI now enables?
Has any role been formally redrawn in the last twelve months to assume an AI-augmented workflow rather than a human-only one?
Can the board name the three operating-model choices behind the company’s largest current AI investment?
For most organisations, the honest answer to all three is no.
Which is why your AI budget keeps growing and your AI return doesn’t.
Sources referenced
PwC. 2026 AI Performance Study (April 13, 2026). Survey of 1,217 senior executives across 25 sectors. pwc.com
Deloitte AI Institute. The hidden ROI of AI: What leaders should actually measure (April 20, 2026). Builds on the 2026 State of AI in the Enterprise survey of 3,235 leaders across 24 countries.
Stanford HAI. The 2026 AI Index Report (April 13, 2026). hai.stanford.edu/ai-index/2026-ai-index-report
BCG Henderson Institute. AI Will Reshape More Jobs Than It Replaces (April 20, 2026). Analysis of approximately 165m US jobs. bcg.com
BCG. AI Transformation Is a Workforce Transformation (April 20, 2026). bcg.com
BCG. As AI Investments Surge, CEOs Take the Lead on Decision Making and Upskilling Themselves (January 2026). bcg.com
Bain & Company. 42% of CFOs plan to increase AI investment by over 30% within two years (April 13, 2026). Survey of 100+ CFOs globally. bain.com
KPMG and INSEAD. AI Governance Principles for Boards (April 14, 2026). Drawing on the KPMG Global AI Pulse Survey. kpmg.com