I get some version of this question in almost every first conversation: "Do we actually need an AI strategy, or are we fine for now?"
It's a fair question. It's also one that most of the content out there won't answer honestly, because most of it is written by people selling you a tool, a platform, or a transformation program. I'm not selling any of those. I'm a fractional AI executive — I get brought in to tell executives the truth about where they stand, then help them do something useful about it. So let me give you the honest answer.
Sometimes, no. You don't need one yet. And I'll tell you exactly how to know which camp you're in.
The two failure modes I actually see
In practice, executives get into trouble in one of two ways — and they're opposite problems.
Failure mode one: doing nothing on purpose. You've watched the hype cycle for two years, you've seen enough breathless LinkedIn posts to be permanently skeptical, and you've decided AI is mostly noise for a business like yours. Sometimes you're right. Often you're not, and the gap between you and a competitor who quietly automated their ops six months ago is now wider than you think.
Failure mode two: doing something for the sake of it. Someone on your leadership team pushed for a "generative AI initiative," you bought some tool licenses, maybe stood up a pilot, and eighteen months later you have a line item in the budget and not much else to show for it. This is the far more common failure mode, and it's the one nobody wants to admit to.
Neither of these is a strategy. One is avoidance dressed up as prudence. The other is activity dressed up as progress.
What "needing an AI strategy" actually means
Here's the reframe I use with every client: the question isn't "do we need AI?" The question is "is there a specific, measurable place in our business where a decision, a process, or a piece of work is currently expensive, slow, or error-prone in a way that AI-shaped technology plausibly fixes?"
If the honest answer is no — if your bottlenecks are sales execution, hiring, or cash flow, and AI touches none of those in any meaningful way right now — then you genuinely don't need an AI strategy this quarter. You need to keep an eye on the space and revisit it in six months. That's a legitimate, defensible position, and I'll tell a client that directly if it's true.
If the honest answer is yes — if you can point to a function (customer support tickets, contract review, code review, content production, forecasting, whatever it is) where the constraint is real and AI-shaped tools are already solving it for companies like yours — then you don't need a "strategy" in the grand, 40-slide-deck sense. You need a plan for that one thing, sized to your business, with an owner and a way to know if it worked.
Most executives skip straight past this diagnostic step. They either dismiss AI wholesale or commission a broad "AI strategy" before anyone has named the actual bottleneck it's supposed to solve. Both are guesses dressed up as decisions.
The tell that you're in "activity, not strategy" territory
A few signs I look for when I'm doing this sanity check with a leadership team:
- You can name the AI tools you've bought, but not the business metric they were supposed to move.
- Nobody in the room can tell you who owns the outcome — IT owns "the tool," but no one owns "the result."
- The pilot from a year ago is still called a pilot.
- Your team is using consumer AI tools individually and inconsistently, with no shared view of what's working.
- The word "strategy" shows up in a slide deck, but not in a single sentence that names a number you're trying to move.
If two or more of those are true, you don't need more AI. You need someone to sit down with your leadership team for a few hours, ask uncomfortable questions, and help you decide — with actual conviction — whether to double down, redirect, or stop.
What I'd tell you if we talked
I'm not going to pretend every business needs a fractional AI executive on retainer. Plenty of businesses I talk to walk away from that first conversation with a clear "not yet, and here's what to watch for" — and that's a good outcome, not a missed sale.
But if you've got that nagging feeling that something in your operation should be faster, cheaper, or better than it currently is, and you suspect AI is part of the answer but you don't trust your own read on it (or your team's), that's exactly the conversation worth having. Not a pitch. A sanity check, with someone who has no reason to tell you "yes, buy more AI" if the honest answer is "no, not yet."
That's the whole premise of what I do: come in as the person in the room who isn't selling the tool, isn't defending last year's pilot, and isn't afraid to tell you the boring truth if that's what's true.
If you want that conversation, book 30 minutes with me and bring your honest, unfiltered version of where things stand. We'll figure out together which camp you're actually in.