Many businesses I talk to are feeling the pressure to “do something with AI.” Competitors are talking about it. Someone’s seen a demo, and it looked impressive. Here’s what I tell them as Mowery & Schoenfeld’s Chief Technology Officer: Slow down. The problem is almost never the one you think it is.
The board wants action. That’s not the same as an AI adoption strategy.
The pressure is real. Directors are reading the same articles as everyone else. Competitors are announcing artificial intelligence initiatives in earnings calls. LinkedIn posts are flying left, right, and center about how some new startup is redefining the way your industry operates. And so, the question shows up in the room: What’s our firmwide AI strategy?
That pressure isn’t wrong. The landscape is genuinely shifting. But this pressure is a terrible project brief that doesn’t tell you what problem needs to be solved.
The leaders who handle this moment well push back — not on the idea of AI, but on the framing. They say, “Tell me what we are trying to solve for, and we’ll figure out the right path.” The ones who handle it poorly buy a tool, announce the initiative, and spend the next year trying to show ROI on something that was never tied to a real business need.
Here’s the practical reality: The technology is moving faster than any organization can fully evaluate it. What was the right answer two months ago has already shifted. Vendors are rebranding existing features as AI. New tools appear every week. Press releases promise more capabilities are on the horizon. Pricing is harder to assess as token costs are changing constantly.
Businesses need to be careful, because the cost of making the wrong bet is the implementation time, the change management, the team distraction, and the opportunity cost of not doing something that would have actually moved the needle.
Leaders who jump directly from “We should implement AI” to vendor selection are navigating without a map. The real conclusion almost always ends up being: “We didn’t know what we were solving for.”
Your pain point is specific to you, and the AI landscape is evolving every day. Those two facts, together, make a strong argument for doing your internal diagnostic work first. The clearer you are on what you’re actually trying to fix, the easier it becomes to evaluate options — and to tune out the ones that don’t fit, regardless of how much noise they’re generating.
Identify Pain Points Before Building a Firmwide AI Strategy
Most of the problems leaders bring to the table are operational, whether it’s margin erosion they can’t fully explain, reporting cycles that take too long, decisions that depend on information sitting across systems that don’t talk to each other, or job costs that look fine until the invoice lands.
These are real problems that cost real money. They’re not, in most cases, problems that a chatbot or generative AI tool solves. They’re problems of process, data architecture, and visibility, and solving them starts with being honest about what’s broken.
It’s essential to figure out the specific moment where the pain occurs. Not the general category — the exact decision, the exact person, the exact place where something goes wrong or takes too long. That specificity matters because it determines what kind of solution helps.
When you understand the pain point clearly, the path to the right tool gets a lot shorter.
Before any technology decision, answer these questions:
- What is the exact decision we’re trying to make faster or better?
- Where does the data we need for that decision currently live — and do we trust it?
- Is this a visibility problem? Would clean, timely data alone change behavior?
- Would AI meaningfully improve the solution, or would a well-built non-AI tool solve the problem just as effectively?
- If yes, are the AI features already in our core systems being used?
- Is there someone in this organization connecting business strategy to our current technology state?
A valuable next step is getting the right stakeholders in the same room. Sometimes it’s almost more important to get everyone aligned than to immediately jump to solutions. When people realize they’re marching toward the same goal, the path forward tends to get a lot clearer.
Improve Visibility Before Investing in AI
Before you can automate a decision, you must be able to make it clearly yourself — backed by aligned business strategy. That’s a data and alignment problem, not an AI problem.
There’s a category of solution that gets undervalued because it doesn’t carry the right label. A rules-based workflow that flags exceptions before they become problems. A dashboard that consolidates the five reports your ops team runs manually. An integration that moves data between systems without a human in the middle.
None of these are “AI” in the technical sense, yet all of them produce outcomes that leaders have been told they need AI to achieve. The right tool is whatever solves the problem at the highest return on investment and lowest complexity. Sometimes that’s a language model, but more often it’s better architecture, cleaner data underneath, and maybe AI used to build a tool that stitches it all together.
When you understand the problem clearly, you stop evaluating tools based on what they are and start evaluating them based on whether they solve what you’ve identified. That shift alone cuts through most of the noise.
Look at Your Current Software
Your existing software is probably doing more than you think.
Most enterprise software vendors — ERP, CRM, financial platforms — have spent the last two years shipping AI-adjacent capabilities into their existing products, including forecasting modules, anomaly detection, automated categorization, and workflow triggers. Most business executives I talk to aren’t using them.
The reason is rarely that the features don’t exist, but rather that no one has taken the time to understand what the tool already does, ensure it has access to clean data, and map it to what the business needs. That’s a governance gap, and it’s a cheap one to close relative to buying something new.
Before spending money on a new platform, take a full pass through what you already have. Read the release notes from the past 18 months. Better yet, have your approved AI tool do it for you. You’ll find things that surprise you.
Why an AI Strategy Consultant Can Help Connect Business and Tech
The gap I see most consistently is in translation. There’s rarely someone in a mid-market organization whose job is to look at the business strategy and ask: Does our current technology posture support this? Are we heading in the same direction?
In large enterprises, that’s a CTO or VP of technology. In the mid-market, it usually falls to whoever is closest to the systems, which typically means IT, the COO, or the CFO. And because those people have other pressing responsibilities keeping the lights on, technology strategy defaults to reactive problem-solving rather than forward planning. The result is technology that accumulates rather than evolves. Each tool solved a real problem when it was added. Over time, the landscape gets harder to manage, data fragments, and investment decisions get made in isolation.
That coherence doesn’t require a full-time executive. It requires someone with enough context across both the business and the technology to ask the right questions consistently, and the standing to act on what they find.
The leaders who get the most out of AI aren’t the ones who move the fastest. They’re the ones who understood their own operations clearly enough to know where intelligent automation would actually make a difference and where it wouldn’t. That clarity is available to any organization willing to do the diagnostic work before reaching for a solution. Working with an AI strategy consultant is a good place to start.
Solve the problem first. The right technology tends to become obvious once you know what you’re solving for.