The most common reason an AI strategy fails to get board backing has nothing to do with the quality of the strategy, and it fails because it is framed as a technology initiative rather than a business case.
Boards do not fund technology for its own sake, and they fund initiatives that protect competitive position, improve margins, manage risk or unlock growth. If your AI strategy does not speak that language, clearly, concisely and with credible numbers, it will be deprioritised in favour of investments that do.
What boards actually want to know
Having sat in board meetings across insurance, financial services, legal and higher education, as CTO and as an advisor, we find the questions boards ask about AI remarkably consistent.
The first question is the commercial case, and boards want the specific, quantified impact on this business rather than the theoretical value of AI in general. They want to see the revenue or margin uplift, the cost reduction and the payback period, alongside the sensitivity analysis. They need to understand what happens to that case if the assumptions are wrong by 20 percent or 30 percent.
The second question is what the business is risking, and boards in regulated sectors are acutely aware that AI introduces new categories of risk. The list runs to model risk, data risk, regulatory risk and reputational risk, and your strategy needs to address these directly rather than dismissing them with a governance paragraph at the end.
The third question is why the moment is now. The board has heard about AI before, and if they did not fund it last time, you need to articulate what has changed. The change might sit in the technology, in the competitive picture or in the regulatory environment, and it is what makes this the right moment.
The fourth question is what happens if the business does nothing, and it is often the most powerful of them. If competitors are adopting AI and you are not, the gap compounds, and if regulatory expectations around AI governance are tightening, waiting makes compliance harder rather than easier.
How to structure an AI strategy for board consumption
Start with the business context, not the technology, opening with where the business is today, where it needs to be and what is preventing it from getting there. AI should enter the conversation as a means to close specific capability gaps rather than as an end in itself.
Use a capability framework, mapping your business capabilities and scoring them against AI opportunity. This gives the board a visual, structured view of where AI fits and, crucially, where it does not, and boards trust structured analysis more than enthusiasm.
Sequence for early value, because boards will not approve a two-year programme without evidence that it works. Design your roadmap to deliver a measurable win within the first 8-12 weeks, building confidence, proving the approach and securing ongoing investment.
Be explicit about governance, making it a headline rather than burying it in an appendix. Show the board that you have thought about ownership, accountability, risk management and regulatory compliance, and in regulated sectors this is often the difference between approval and rejection.
Price it properly, including total cost of ownership (not just the technology): the people, the process change, the integration work and the ongoing operational cost. Boards distrust budgets that look too optimistic, and they are right to.
The common mistakes
Leading with technology is the first mistake, and if your first slide is about GPT-4, Claude or a vendor platform, you have already lost the room. The fix is to lead with the business problem.
Promising too much is the second mistake, and a promise that "AI will transform our business" is not a strategy. "AI will reduce claims processing time by 40 percent and improve fraud detection accuracy by 25 percent in the first six months" is a strategy.
Ignoring the operating model is the third mistake, and AI changes how people work, so your strategy needs to address role changes, process redesign and change management. If it does not, the board will ask and you will not have a good answer, and AI Operating Model Design sets out why this matters.
Missing competitive context is the fourth mistake, and your board wants to know what your competitors and peers are doing. If you cannot articulate the competitive picture around AI in your sector, your strategy lacks urgency.
For the foundational thinking behind capability-led AI strategy, see Why AI Strategy Must Lead Technology.
If you need help building an AI strategy that your board will back, Breathe is designed to produce exactly that, a board-ready strategy with a costed, sequenced roadmap, delivered in 5-10 working days.