System+Signal: You Don't Need More AI. You Need a Roadmap.

How to stop spending blindly on AI — and start with a plan that actually fits your firm.

You're starting to see it. Maybe it's your own budget, maybe it's a conversation with another firm owner over coffee — but the pattern is hard to miss. Subscriptions are stacking up. Every other week there's a new AI tool someone on the team signed up for. The invoices are small enough that none of them raise a flag on their own but pull them all together and the number is starting to get uncomfortable.

A ChatGPT subscription someone expensed in January. A Copilot license rolled out in March that half the team forgot about by April. A meeting transcription tool the office manager found, a separate one the paralegals prefer, and a third that came bundled with something else nobody remembers buying. Each one felt like a small, reasonable decision at the time. Collectively, they've become an expensive fog — and more people are starting to notice as professional service firms in Massachusetts are feeling the crunch.

The data backs up what you're already feeling. A recent audit of more than 100 small and mid-sized businesses found that 87% had significant waste in their AI tool spending, with a median annual waste of $18,000. The average company now runs 7.2 AI tools — up from 5.3 just two years ago — and 63% of those tools have overlapping functions. Writing tools that duplicate each other. Summarization features are built into three different platforms. Code assistants nobody in the office writes code with.

And perhaps the most telling number: only 12% of AI purchasing decisions at small and mid-sized firms come from any kind of formal needs assessment. The rest? Forty-two percent come from executive preference — someone saw a demo, got excited, bought a license. Twenty-eight percent are "keeping up with competitors."

That's not a strategy. That's a shopping spree with a vague sense of urgency. And people are starting to say so out loud.

The Spending Problem That's Getting Harder to Ignore for Massachusetts Businesses

Walk through any professional district between Barnstable and Boston — the law offices along Route 3A in Hingham, the accounting firms tucked behind Braintree's town center, the financial advisors up and down Hancock Street in Quincy — and the conversations are shifting. A year ago, the question was "should we be doing something with AI?" Now it's "how much are we spending on this, and is any of it working?"

It's a fair question. And most firms don't love the answer.

Thomson Reuters' 2026 report on AI in professional services found that while 40% of professionals now say their organization uses generative AI — nearly double from a year ago — only 18% say their organization tracks return on investment. Another 40% say they don't even know whether ROI is being measured. That means the majority of firms adopting AI are doing so with no success criteria at all.

The average mid-size company now spends roughly $4,200 per employee per year on AI-powered software, up 38% from 2024. For a 40-person firm, that's nearly $170,000 a year. And half of that spend, according to multiple analyses, is going to overlapping tools, unused seats, and plans that don't match actual usage.

This isn't an AI problem. It's a planning problem dressed in AI's clothes. And it's becoming visible enough that people are finally ready to do something about it.

The Real Question Isn't "What Tool?" — It's "What Hurts?"

There's a better way to think about this, and it starts with a question that has nothing to do with technology: What's the most painful, repetitive, low-judgment task in your office?

Not the most exciting AI use case. Not the one that sounds impressive at a networking event in Plymouth or a bar association lunch downtown. The one that quietly eats hours every week — the work your best people spend too much time on and resent just enough to mention at their annual review.

For a lot of the firms we work with — from Weymouth down through Duxbury, and up into Quincy and the city — the answers tend to cluster around a few familiar pain points. Document drafting and formatting. Meeting summaries that never get written, or get written badly. Client intake that involves the same twelve questions typed into three different systems. Email triage that buries the urgent under the routine.

These aren't glamorous problems. But they're the right starting point — because they're specific, measurable, and small enough to pilot without betting the firm on a technology nobody fully understands yet.

A Harvard Business Review analysis published this spring put it bluntly: most professional services organizations are failing at AI because they've misdiagnosed which problem they're trying to solve. They buy tools first and look for problems second. The firms that succeed — the ones seeing real returns — do it the other way around. They start with the pain, find the simplest tool that addresses it, and measure whether it actually helps before expanding.

What a Plan Actually Looks Like for Businesses in Southeastern Massachusetts

An AI strategy for a 30-person law firm in Norwell or a 50-person accounting practice off Route 18 in Weymouth doesn't need to look like a Fortune 500 roadmap. It doesn't need a committee, a consultant, or a six-month timeline. It needs five things:

An honest inventory. What AI tools are you already paying for? Who's using them? What are they using them for? You might be surprised — or alarmed — by the answers. Nearly 90% of small and mid-sized businesses have never conducted a formal usage audit of their AI tools.

A pain-first priority. Pick the one workflow that wastes the most time relative to its value. Not the one that's trendy. The one that hurts.

A single pilot. One tool, one team, one workflow, 30 days. That's it. You're not transforming the firm. You're testing a hypothesis.

A way to measure. Before you start, write down what "better" looks like. Fewer hours per task? Fewer errors? Faster turnaround? If you can't define it before you buy the tool, you won't be able to prove it after.

A governance baseline. Who's allowed to use which tools? What data is safe to put into them? What data isn't? If your team can't answer these questions consistently, you don't have a policy — you have a liability. This is especially critical for firms handling privileged or regulated information — which, around here, is most of them.

That's it. Five things. None of them require technical expertise. All of them require honesty about where you are and discipline about where you start.

The South Shore Firms That Win This Won't Be the Ones That Spent the Most

There's a version of the AI story that sounds like an arms race — spend more, move faster, adopt everything. It's the version that sells conference tickets and software licenses. But it's not the version that's playing out in practice.

The firms that are getting this right — the ones in Scituate and Braintree, in Rockland and along the waterfront in Quincy — aren't the ones with the biggest AI budgets. They're the ones that asked better questions before they spent anything just like you would with any other capital expenditure or hiring decision.

They mapped the work before they bought the tool. They measured the baseline before they launched the pilot. They wrote down what "success" means before the first invoice came due.

In communities built on practical, client-centered professional work — where relationships still drive referrals and a handshake still means something — that approach isn't just sensible. It's the only one that makes sense.

You don't need more AI. You need a map. And the best time to draw one is before you spend another dollar. Join us this month at www.systemsupport.com/summer and across all of our platforms as we map out an AI road trip to help your business.