Short answer: a good AI strategy workshop gets your leadership team to agree, in one or two days, which AI use cases matter most to the business, how AI will be governed and who owns the first 90 days. It should end with a ranked use-case portfolio, a one-page governance framework and a 90-day roadmap with named owners - not a list of ideas or a vendor shortlist.
Most leadership teams already know AI matters. The problem is that they can't agree what to do about it. Departments run pilots nobody else knows about, vendors push their platforms, and the board keeps asking for 'the AI strategy'. This guide explains how to run a workshop that turns that noise into decisions, and what to look for if you bring in a facilitator.
Why AI strategy stalls in leadership teams
AI strategy rarely stalls because leaders lack information. It stalls because of alignment problems that look like knowledge problems:
- Shadow experiments: different teams are trying different tools with no shared view of what exists.
- Technology-first thinking: conversations start with 'should we use a chatbot?' rather than 'which business outcome matters most?'
- No owner: AI sits between technology, operations, data and legal, so everyone has an opinion and nobody has the decision.
- Governance as an afterthought: risk, data and ethics questions surface late and stop momentum.
- Vendor influence: the use cases being discussed are the ones a supplier happens to sell.
A workshop that just generates more ideas makes these problems worse. A workshop built around decisions fixes them.
What an AI strategy workshop should produce
Before you book anything, agree what you need in writing at the end. For an executive team, that should be:
| Output | What it contains |
|---|---|
| AI landscape map | Everything the organisation is already doing with AI, mapped against strategic priorities, with the gaps made visible |
| Use-case portfolio | Every use case expressed as a business outcome and scored for impact and feasibility |
| Top 3-5 priorities | The use cases the team commits to first, with the reasons recorded |
| Governance framework | One page covering risk appetite, data, ethics, approvals and escalation |
| 90-day roadmap | Named owners, success measures, resources and decision checkpoints |
| Decision log | Every choice made, with its rationale, owner and deadline |
A four-phase structure that works
The most reliable AI strategy workshops follow four time-boxed phases, each producing a written output before the next begins.
1. Map the landscape
Surface everything already happening: official projects, shadow experiments and live vendor conversations. Map them against your strategic priorities. Teams are often surprised by how much is already going on, and how little of it connects to what matters most.
2. Discover use cases
Generate ideas silently and individually first, so the most senior or most enthusiastic person doesn't anchor the room. Then build on each other's ideas together. The key rule: every use case must be written as a business outcome, such as 'reduce customer onboarding time by 40%', not as a technology, such as 'implement a chatbot'.
3. Evaluate and prioritise
Score every use case against the same criteria. A practical set is:
- strategic alignment
- feasibility of implementation
- data readiness
- risk profile
- expected return
The senior decision-maker then selects the top three to five, and the reasons are written down.
4. Govern, own and plan
Agree who approves AI deployments, how data and ethics are managed and how issues are escalated. Assign a named owner to each priority and build a 90-day plan with decision checkpoints. A 90-day plan is more useful than a 12-month strategy, because AI changes too quickly for long plans to survive.
Who should be in the room
- The decider: the chief executive or most senior decision-maker, present for the whole session, with final authority at each decision point.
- The leadership team: usually 5-14 leaders across strategy, operations, technology, finance and people. AI decisions affect all of them, so all of them need to own the result.
- A neutral facilitator: someone with no product to sell, who runs the process and keeps the focus on business outcomes.
Leave vendors out. Their input is useful later, when you are choosing how to deliver a use case, not when you are deciding which use cases matter.
One day or two?
- One day works for teams with a clear AI ambition who need to align and prioritise quickly. All four phases fit into a single, intensive day for 6-12 people.
- Two days gives more room for governance design, reflection overnight and a board-ready summary. It suits larger teams or organisations where AI decisions carry significant risk.
How to choose an AI strategy workshop provider
Providers fall into three broad groups: technology vendors and AI consultancies, large management consultancies, and independent specialist facilitators. Whichever you consider, check five things:
- Are they vendor-neutral? If the provider sells or implements a platform, its recommendations will lean towards that platform.
- Do they start from outcomes? Ask how use cases are framed. 'Business outcome first' is the right answer.
- Is governance built in? Governance designed during the workshop is far easier than governance bolted on after a problem.
- Will we leave with owners? A roadmap without named owners and dates is a wish list.
- What happens after? Look for a follow-up to check that priorities and owners are holding.
How Echelon runs AI strategy workshops
Echelon’s AI Design Sprint follows the four-phase structure above, in a one-day or two-day format, for UK and international leadership teams. There are no vendors in the room, every use case is framed as a business outcome, governance is designed on the day and every priority leaves with a named owner and a 90-day plan. The AI Design Sprint is priced from £6,000 to £9,000 - see our pricing for details, including founding client rates.
Echelon was founded by Dr Andrew Greenland, a medical doctor who has spent years running large, high-stakes working sessions where expert groups must produce work that stands up to scrutiny. You can read more about the experience behind the practice.
Frequently Asked Questions
What is an AI strategy workshop?
An AI strategy workshop is a facilitated session in which a leadership team agrees which AI use cases to pursue, how AI will be governed and who owns delivery. A good one ends with a prioritised portfolio, a governance framework and a 90-day roadmap with named owners.
What is an AI design sprint?
An AI design sprint applies the pace and structure of a design sprint to AI strategy. Over one or two days, a leadership team maps current AI activity, generates and scores use cases, and agrees governance and owners, so it leaves with decisions rather than a list of ideas.
How much does an AI strategy workshop cost in the UK?
Costs vary widely depending on the provider and length. Echelon’s AI Design Sprint costs £6,000 to £9,000 for the whole team, including preparation, the session, written outputs and a 30-day follow-up call.
Who should attend an AI strategy workshop?
The most senior decision-maker plus 5-14 leaders from across the business, including strategy, operations, technology, finance and people. Vendors should not attend the prioritisation stage.
Do we need technical AI knowledge to take part?
No. The workshop focuses on business outcomes, priorities and governance. Leaders leave with a shared vocabulary and a clearer ability to judge AI proposals, but technical expertise is not required to take part.