Strategy
Where should we actually use AI?
Most AI strategies are vendor lists and capability maps. They do not tell a COO which workflow to change first, what to ignore, or how the result will show up in cost, revenue, or capacity.
2/3
of organizations remain in experiment or pilot mode with AI, even as 88% report using it somewhere.
McKinsey, The State of AI in 20251
How I see it
AI strategy that names the work, not the stack
A useful AI strategy is a set of bets, not a vision statement. Each bet names a workflow, an owner, a baseline, a path, and a reason the economics could matter.
The first job is subtraction. Many ideas are real problems that are not AI problems. Some are too rare to compound. Some have no system of record. Some are political. A strategy that cannot say no will fund a portfolio of pilots that never reach production.
The second job is sequencing. Improve the work you already have, redesign the workflow, or put an agent on one outcome. Those are different investments. Mixing them into one roadmap hides the cost and the risk.
The third job is constraint. Talent, data access, controls, and the willingness to change process decide what is possible this quarter. Strategy that ignores those constraints becomes a slide that nobody can execute.
How the engagement runs
What you actually buy.
A service is a sequence with an artifact at each step. It is not a transformation program you purchase as a bundle.
Name three bets
Each bet has a workflow, an owner, a baseline, and a reason the economics could matter.
A ranked list a COO can fund this quarter.
A capability map or a vendor landscape.
Write the no-list
Record which ideas are real problems that are not AI problems, too rare, or political.
Popular ideas can be stopped in writing.
A roadmap where everything is still a yes.
Sequence one workflow that can ship
The first bet must reach production without a platform program.
A contained path with an owner and a kill criterion.
A transformation sequence that needs a new operating model first.
Do not buy this for
These engagements fail for predictable reasons.
Starting with platforms and models
The stack is a later decision. If the first artifact is an architecture diagram, the team has skipped the economic question.
Publishing a roadmap of use cases
A list of twenty ideas is not a strategy. Three ranked workflows with owners and baselines will move the company further.
Confusing strategy with transformation
You do not need a new operating model to get a first win. You need one process that is expensive enough to change.
Before we start
Questions that decide whether to engage.
Can you name the three workflows that would change the operating numbers if they got faster or cheaper?
If the list is long or vague, the strategy is not finished.Does each candidate have an owner who can change the process?
Without an owner, the work will stall after the workshop.Do you know which ideas are not AI problems?
If everything is an AI opportunity, nothing is prioritized.Is the first bet small enough to reach production?
A first bet that requires a platform program will not ship.Will you stop work that fails the economic test?
A strategy without kill criteria becomes a standing committee.
How we size it
Strategy filtermaterial annual opportunity × owner who can change the process × path the organization can operate
If any term is missing, the idea stays on the list. It does not become a funded bet.
This is the wrong engagement if
Do not start this engagement yet.
- Leadership wants a vision deck more than a ranked set of bets.
- There is no access to the people who run the work.
- The company has already chosen a platform and wants strategy to justify it.
- No one will accept a recommendation to do nothing on a popular idea.
A useful next step
Bring one expensive workflow. Get guided into production.
We guide the implementation, go deep on the technical path, and stay hands-on through operations — or tell you when a simpler answer is better.
Discuss an AI opportunity