
I have read more strategy documents in the last eighteen months than in the previous five years combined — because every engagement I take on begins with a review of what the business has already built. And in almost every one of those documents, there is a section called something like "Our AI Strategy" or "AI-Powered Growth".
In almost every case, it describes a set of tools the business plans to use — or is already using. ChatGPT for content. Midjourney for images. A CRM with AI features. An analytics platform that surfaces predictions.
That is not an AI strategy. That is a list of software subscriptions.
AI is an accelerant. It makes things that were possible faster, cheaper, and more scalable. The question that matters strategically is not "are we using AI?" but "what are we accelerating — and is that the right thing to accelerate?"
If you have a differentiated value proposition, AI can help you deliver it at a scale and speed that was previously impossible. If you do not have a differentiated value proposition, AI will help you deliver an undifferentiated one faster — which is not a competitive advantage. It is competitive noise.
AI creates genuine strategic advantage in three ways.
Personalisation at scale. The ability to deliver a genuinely personalised experience to every customer — not a segmented experience, but an individual one — at a cost structure that was previously only available to the largest enterprises.
Predictive capability. The ability to anticipate customer needs, operational bottlenecks, and market shifts before they become visible — and to act on that anticipation rather than reacting after the fact.
Operational leverage. The ability to deliver more value with the same resources — not by working harder but by designing the operation so that intelligent automation handles the repeatable and humans handle what only humans can do.
Before adding AI to your strategy, answer this: what is the core strategic bet we are making about where we can win, and how does AI help us win there more decisively?
If you cannot answer that question, you are not building an AI strategy. You are collecting tools.