Most AI strategy engagements produce a beautiful deck and no decisions. We work the other way — decisions first, then the two pages of evidence behind each one.
A retained AI strategy is a stack of decisions. Some are obvious: does the operator want to be on the frontier or two steps back. Most are not: which of the workflows the company already runs are AI-shaped, and which are not — no matter how loud the hype is around them. Where the leverage actually is on the P&L, and where AI would only paper over a process problem. What the operator is willing to spend twelve months getting wrong to eventually get right, and what they aren't.
Almost every executive team we sit with inherited these decisions rather than made them. A vendor showed a demo. A board member sent a link. A competitor announced something. The team now has an AI program that fits the noise, not the business.
Our engagement flips that. We build the roadmap you actually need, out loud, on paper, with you in the room. Where the honest answer is that the highest-return AI project you have this year is a better report and a cleaner data pipeline — not a chatbot — we say so, and we say why.
What we produce
- Strategic thesis. A short written document that answers the questions most consultants never bother to ask: what does AI change about how the business makes money, what changes about how it loses money, and which of those matter more this year. Everything downstream flows from this.
- Prioritized initiative map. A ranked list of the AI investments actually worth making in the next twelve months, with expected value, honest risk, and the two or three things that could kill each one. We rank them; you decide.
- Anti-list. What we recommend you not do — usually the more useful half of the memo. Named projects that look promising in a demo but that we would advise against, with the reason each one won't survive contact with your operation.
- Board briefing. A ninety-minute working session with the board or executive team to walk the thesis, take challenge, and leave with a set of decisions the CEO can act on the following Monday.
What we don't do
We don't stand up your platform. We are not writing the pipelines or configuring the vector store, and we are not the ones on the call at 3 a.m. when the eval regressed. We advise on the architecture of the program, we help you retain the right builders, and we audit their work at the intervals we set together. Anyone who tells you they can be both the strategist and the operator is trying to sell you an engagement that outlasts its usefulness.
We also don't dress up transformation theatre. If the honest answer is that the operator doesn't need an AI strategy this year — that a decent BI dashboard, a rewritten SOP, and a two-page written policy will cover ninety percent of what's likely — we say so. We have talked more executive teams out of AI programs they didn't need than into ones they did.
Roadmaps we've helped design have ranged from a single functional pilot inside one line of business, to an enterprise-wide portfolio spanning product, ops, and shared services. The thesis scales; the deliverables do not. A hundred-page transformation blueprint is a way to bury a bad idea in paper. We keep them short on purpose.
Engagement
Strategy engagements are typically project-shaped — a four-to-six-week design phase followed by an optional light retainer for quarterly reviews as the market and the technology shift. Fee structures are fixed-fee for the initial phase; retainers cover a fixed number of hours and a direct line for judgment calls between quarters.
Introductions happen by referral. If you were pointed here by someone we work with, mention their name when you write. If not, tell us plainly what brought you and what you're trying to decide. We reply within two business days.