How to run an AI strategy workshop that isn't a waste of time

Every company wants an AI strategy. Most approach it with a town hall and a consultant deck. The town hall creates anxiety. The deck creates confusion. Neither produces decisions.

The real problem with AI workshops

Most AI workshops fail because they ask the wrong question. 'How do we use AI?' is too broad. It generates a list of use cases nobody owns and a to-do list nobody acts on. The useful question is narrower: 'What is the most expensive problem we have where AI could plausibly help in the next twelve months?'

Who should be in the room

AI strategy workshops work when they combine three types of people: someone with enough technical literacy to filter fantasy from feasibility; someone who understands the operations well enough to identify real pain points; and someone with the authority to allocate budget and headcount. Without all three, you'll either dream too big or decide too narrowly.

A structure that works

• Morning: identify and rank your top five operational pain points that have data at their core.
• Midday: for each, assess what AI could realistically do, what you'd need to start, and what the cost of the status quo is.
• Afternoon: select one initiative to prototype. Not five. One.

The output that matters

A good AI workshop ends with a one-page initiative brief: what problem you're solving, how you'll know if you've solved it, who owns it, and what you're doing in the next 30 days. If you leave with a slide deck instead, the workshop hasn't finished.

“AI won't tell you which problems are worth solving. That still requires people who understand the business talking honestly with each other.”

The technology conversation is the easy part. Deciding what's important enough to change — that's exactly the conversation a well-facilitated workshop is designed to have.