The hardest part of AI transformation isn't the AI — it's re-engineering how your organisation thinks. Tools dropped into old thinking just produce old outcomes faster.
- AI transformation is a thinking problem wearing a technology costume. The hardware is rarely the constraint; the way the organisation thinks is what holds it back.
- The companies that thrive do not add AI to what they already do; they re-engineer how they think, with the customer and human judgement at the centre.
- The average S&P 500 company now stays on the index for about 15 years, down from 33 in the 1960s. "Change or become irrelevant" is no longer rhetoric.
- You cannot future-proof a business. You can only build a culture that keeps re-thinking.
The hardest part of AI transformation is not the AI. It is re-engineering how your organisation thinks. You can buy the best models on earth and still fail, because intelligence dropped into a business that thinks the old way just produces old outcomes faster. This is why I have always said digital, and now AI, is not a bolt-on. It has to be woven into the centre of how the business thinks, decides and serves customers. And that is a human transformation, not a technical one.
Years ago Tony Hsieh, the late founder of the online shoe retailer Zappos, showed me around his company. He told me about their longest customer call: eight and a half hours, with a single mother who had been home all week with a sick child and had not spoken to another adult for days. The rep simply stayed on the line. She bought a pair of trainers; Zappos sent flowers and chocolates; I am still telling the story today. No technology did that. A culture did. Zappos had digital at its core and the human judgement to use it well, and that combination is the whole game.
Why is AI transformation really a thinking problem?
Because the technology is the easy 10%. The research keeps proving the point. BCG (Boston Consulting Group, a global management consultancy) estimates that roughly 70% of the value from AI comes from people and process, about 20% from data and technology, and only around 10% from the algorithms themselves. The bottleneck is no longer the technology. It is the thinking around the technology.
| Organisations that bolt AI on | Organisations that re-engineer their thinking |
|---|---|
| Ask "what can this tool do?" | Ask "how should we think differently now this exists?" |
| Put the technology at the centre | Put the customer and human judgement at the centre |
| Run 18-month projects, then ship | Run short iterations, learn, adjust continuously |
| Use accountability to punish | Use accountability to learn against outcomes |
| Become a slower version of themselves | Become a technology company that happens to do what they do |
The evidence on the gap is sobering. Deloitte's State of AI in the Enterprise survey, which polled 3,235 business and IT leaders across twenty-four countries in late 2025, found that only 34% of organisations were using AI to deeply transform the business, while 37% were using it at surface level, with little or no change to how the work actually gets done. Same tools, mostly. Different thinking, rarely.
Pizza Hut stopped being a company that sells pizzas and became a technology company that happens to make pizzas, tracking your order and even your location so the pizza arrives fresh. That shift is not a tool purchase. It is a change in how the whole business understands what it is.
Ready to change the thinking?
The AI Strategy Session works on the human and cultural layer where AI value is actually won: ninety days, one plan.
Book your Strategy SessionIs "change or become irrelevant" really that stark?
Yes. In the 1960s the average company stayed on the S&P 500 (the index of America's 500 largest listed firms) for around 33 years. Today that tenure has roughly halved to about 15 years, and Innosight, the strategy firm that tracks the figure, forecasts it will keep falling. John Chambers, who ran the networking giant Cisco for two decades, put it bluntly: at least 40% of today's enterprises will not exist in a meaningful way within ten years if they fail to change the whole company around new technology. You cannot future-proof a business by standing still; the only durable advantage is a culture that keeps re-thinking.
You can't install new thinking with a new tool. AI transformation is a human transformation, or it does not happen at all.
What does re-engineering your thinking actually involve?
Three shifts, in order, culture first and technology last:
- Customer-centric leadership. Start with why the business exists and the value your customer actually wants, then ask how AI changes the way you deliver that value. Not "what can the tool do," but "how do we serve better now this exists."
- A strategic plan toward maturity. Replace the 18-month mega-project with short, iterative sprints: a quick win, measured, then the next one. Build toward a single, clear view of the customer rather than a pile of disconnected tools.
- A culture of accountability. Make the outcome explicit and measure against it, to learn rather than to punish. Most organisations use accountability as a stick; the ones that transform use it as a compass.
None of this concerns the technology. It is about the people inside the business thinking differently, which is the one upgrade no vendor can sell you. Install new software on broken thinking and you get an expensive disappointment. Upgrade the thinking first, and the technology finally has something worth running on. That is the human dimension of AI, and it is the only part that compounds.
Frequently asked questions
If we buy the best AI tools, won't transformation follow?
Where do we start if it's a "thinking" problem?
Can we future-proof the business?
About the author
British technology futurist, AI keynote speaker and advisor. Thirty years across enterprise technology and AI strategy, helping leaders navigate the future of work. The futurist who died.