I’m a little obsessed with strategy.
Lately, I’ve been getting the same questions week after week: What does AI transformation actually look like in practice? How does it fit into the way a company operates?
That got me sketching out a simple way to think about it: company strategy, goals and measures, and day-to-day operations.
A few disconnects stand out.
AI can become the strategy before the company is clear on what it wants to achieve. Goals can be set without a clear starting point. Teams can be busy automating work without knowing how it connects to the strategy or how it will be measured. And strategy and activity can look aligned while nobody has defined what success means.
Several of these can be happening at once.
Imagine a company says its goal is to improve the customer experience. It gives teams AI tools and measures how much time they save. But if the handoffs, decisions and delays that frustrate customers don’t change, is the experience actually any better?
In McKinsey’s 2026 survey, nearly three-quarters of respondents it classified as AI high performers said their organizations had fundamentally redesigned workflows, compared with a quarter of other respondents. That doesn’t prove redesign caused their results, but it reinforces a point I keep coming back to: adopting a tool and changing how work gets done are not the same thing.
For an AI initiative, I’d want a clear line of sight across all three: what the company is trying to achieve, how we’ll know it’s working, and what will actually change in the way people work.
The starting point can vary. An operational experiment might help shape the strategy. And sometimes AI won’t be the best answer.
The question I keep coming back to is:
Where’s the disconnect between what the company wants from AI, what it measures, and what it’s actually changing?
Source: McKinsey, The state of AI in 2026: On the road to ROI (August 25, 2026).
