Strategy decks are inert. They sit in inboxes. They get politely received in meetings. They occasionally influence a brief. But they don't do anything. They describe what could happen. They recommend what should happen. And then they wait for someone else to make it happen.
For fifteen years, that was my job. Observe the market. Synthesise the insight. Write the recommendation. Hand it to someone who could actually build. The thinking was the valuable part. The making was someone else's problem.
AI changed that equation completely.
The chain collapsed
The old process had a long chain from insight to real thing. Brief, concept, pitch, approval, resource allocation, sprint planning, development, QA, launch. Each link in the chain added time, cost, and the opportunity for the original idea to be diluted into something unrecognisable. By the time a strategy became a product, it had been through so many filters that the sharpness was gone.
That chain has collapsed. The distance between "I think this should exist" and "it exists" is now measured in hours, not months. A strategist with AI tools can observe a gap in the market on Monday and have a working prototype live by Wednesday. Not a mockup. Not a wireframe. A functioning product that real people can use and respond to.
AI didn't replace me. It promoted me.
This is the part that took me a while to understand. When I first started building with AI, I thought I was learning a new skill. What I was actually doing was removing the bottleneck between my thinking and the world. I wasn't becoming a developer. I was becoming a strategist who could prove their ideas instead of just presenting them.
The insight is still the hard part. Knowing what to build, who it's for, why it matters, what makes it different. That comes from years of pattern recognition across brands, categories, and cultures. AI can't do that. What AI can do is take that insight and turn it into something tangible before the moment passes.
The old process was observe, hypothesise, recommend, wait. The new process is observe, prototype, test, iterate. The feedback loop collapses from months to days. You learn faster. You're wrong faster. You course-correct faster. And the work gets better because it's tested against reality, not against a client's imagination of what reality might be.
Learn to build, not to code
There's an important distinction here. I'm not suggesting every strategist should learn Python or JavaScript. Coding is syntax. Building is decision-making. Building is knowing what to prioritise, what to cut, what the user actually needs versus what sounds impressive in a brief. Those are strategy skills. They always were.
The most hireable person in 2026 isn't the one with the best deck. It's the one with a portfolio of shipped work. Things that exist. Things people use. Things that prove the thinking was right (or reveal that it wasn't, which is equally valuable). Nobody reads CVs any more. They Google you. And when they do, having twenty live products says more than twenty years of PowerPoint.
Builders are replacing recommenders
This isn't a temporary trend. The structural advantage of being someone who can think and make is permanent. Every company would rather hire someone who brings a working prototype to the meeting than someone who brings a forty-page deck describing what the prototype could look like.
The strategy industry spent decades arguing that thinking is more valuable than doing. That the idea is worth more than the execution. That's a comfortable position when execution is expensive and slow. When execution is fast and cheap, the argument collapses. Thinking is still essential. But thinking without the ability to act on it is a luxury the market will no longer pay for.
The strategist who builds isn't less of a strategist. They're more of one. Because their strategies get tested, refined, and proven in the real world instead of dying quietly in a shared drive.
I made the full case in The Thinker and the Maker, a 30-slide presentation on what happens when a strategist learns to build.
