What I Tell Every New Client in Their First Meeting

In the last 30 months, I've worked with 40+ professional services firms on AI adoption. Big firms and small. Consulting, law, engineering, architecture, accounting. Tech-forward and tech-skeptical.

They come in with different questions, different problems, different appetites for risk. But the first meeting follows a familiar arc. I tell them the same things. Here they are, for you.

1. "AI Isn't About the Tools"

Every firm wants to talk about which platforms to adopt. Claude or ChatGPT. This vendor or that vendor. They want a tool roadmap.

I tell them: tools don't matter yet. Not because tools aren't important—they are. But because the tool decision happens later, after you understand your actual problems.

The firms that bought the fanciest AI tools first? Most of them aren't using them. The firms that started with "what's broken about how we work?" and then found tools to fix it? Those are thriving.

So we don't start with tools. We start with problems.

2. "Find the People, Not the Processes"

Every firm has processes that suck. But not every sucky process is worth fixing with AI. Some aren't painful enough. Some are about to be automated by your vendor's next release. Some only affect 2 people.

I look for the opposite: problems that are painful, frequent, and affect a significant portion of your team. Better yet, find the person (or people) who feel that pain most acutely. They will be your early adopters, your cheerleaders, your "proof of concept."

Start there. Get them a win fast. They'll pull others along.

3. "Measure Before You Deploy"

This is where most firms fail. They deploy an AI tool and hope it works. Three months later: "Are we seeing value?" No one knows.

Before you deploy, define success measurably:

• If it's a time-saving tool: how much time are we saving per transaction, per person, per year?
• If it's a quality tool: how do we measure quality? What's the baseline? What's the target?
• If it's a throughput tool: how much more work can we do with the same resources?

Measure the baseline now. Deploy. Measure again at 30, 60, and 90 days. If you don't see the expected improvement, diagnose why immediately. Don't wait for the next quarter.

The fastest learning comes from rapid measurement and adjustment.

4. "Adoption Isn't Guaranteed; Plan for Resistance"

People resist change, especially when they're comfortable with how things work. Your partners might see AI as a threat to their billing model. Your staff might worry about obsolescence. Your clients might not trust automated work.

Pretending this away doesn't make it go away. It makes it worse. Instead:

• Be explicit about what AI means for different people. (This is what I call the "culture conversation" from earlier.)
• Involve resisters early. Let them shape how the tool gets used.
• Start small. Prove it works. Scale gradually.
• Make adoption part of performance reviews and compensation. If you want people to use AI, incentivize it.

Resistance properly handled becomes buy-in.

5. "Start With Data Hygiene, Not AI"

Most professional services firms have terrible data practices. Time entries are sloppy. Project data is outdated. Financial records are messy. CRM data is half-filled.

You can't build AI systems on bad data. So before you deploy AI, get your data house in order:

• Define what data you need. Be specific.
• Audit what you actually have. What's missing? What's wrong?
• Fix it. Real fix. This takes time and discipline.

It's boring. It's not as exciting as "AI transformation." But it's the prerequisite. Firms that skip this step fail at AI. Every time.

6. "You Need Someone Accountable for This"

AI adoption doesn't happen if it's everyone's job. It becomes no one's job.

Assign ownership. Not to IT, not to operations, not to a committee. To one person (or a small team) who has explicit authority and accountability for AI strategy and rollout. They report to leadership. Their compensation is tied to outcomes. They have budget.

This person doesn't have to be the CTO. They have to be someone who understands your business, respects your people, and has the credibility to make decisions others will actually follow.

7. "Celebrate Small Wins, Learn From Small Failures"

Big transformations don't happen in one move. They happen in a series of small moves, each building on the previous one.

Your first AI project should be scoped small: one process, one team, clear success criteria. Get it done in 60-90 days. Celebrate it publicly. Then tackle the next problem.

When things don't work—and some won't—treat it as a learning opportunity, not a failure. "We tried X, it didn't work because Y, here's what we learned." Then move forward.

Momentum compounds. Early wins create belief. Belief creates engagement. Engagement creates culture change.

8. "Your Competitive Advantage Is Your Judgment, Not Your Speed"

This is the thing I emphasize most. AI will make your firm faster. But speed is easily copied. Your competitor can buy the same tools and get fast too.

Your real advantage is judgment. What problems matter most to your clients? Which solutions actually work? When should you say no? Which risks are worth taking?

Use AI to do the fast, boring parts. Free your team to focus on judgment. The firms that get that right pull further ahead every quarter.

9. "This Takes 18-24 Months; Plan Accordingly"

Serious cultural and operational change doesn't happen in 90 days. It takes time to build new habits, to build trust in new systems, to see results compound.

Plan for 18-24 months of focused effort. Some results come faster. Some take longer. But if you're thinking about AI as a "let's do it this quarter" project, you'll fail. If you're thinking about it as a capability you're building over the next 1.5-2 years, you'll succeed.

Patience is underrated in AI strategy.

10. "We're At the Beginning, Not the End"

This is my closing thought in every first meeting. AI capability is advancing faster than deployment. The models get better every quarter. New use cases emerge constantly.

You're not trying to solve the AI problem once and then move on. You're building the capability to continuously adapt as AI capabilities expand.

That's what you're actually building. Not a project. A capability.

The End of the First Meeting

I finish with this: "Here's what we'll focus on. Problem identification, ruthless measurement, phased deployment, accountability, and celebrating small wins. We'll move fast where we can, slow where we should. And we'll learn continuously."

Every firm I say this to nods. Some of them follow it. Those ones thrive.