They're all confused. That's your business.

I had a call last week with a guy who runs a 14-person landscaping company outside Raleigh. Sharp guy. Profitable. Been at it 22 years.

He said: "I know I need to do something with AI. I just don't know what."

I've heard that sentence probably 200 times in the last 18 months. From a dentist. A property manager. A commercial cleaning operation. A CPA with 3 locations.

Always the same sentence.

Here's what I noticed.

Every single one of them had already done research. They'd watched YouTube videos. Read articles. Signed up for some tools. Got overwhelmed. Stopped.

The problem they have is real. They know AI can probably help them. They can't figure out which AI, which workflow, which problem to start with, or whether what they've seen is even real or just hype.

So they do nothing.

That paralysis is not a character flaw. It's what happens when a busy operator gets hit with 400 options and no one they trust to help them sort it out.

The skill that pays in 2026

I've built businesses across two decades. Real estate in the early 2000s. Lending. Lost everything in 2008 and had to rebuild from scratch.

What I learned from that rebuild is that the most expensive thing in any business is confusion.

People will pay real money to get rid of confusion. That's what consultants do. That's what attorneys do. That's what a good doctor does. They charge for clarity, not for labor.

AI is the new confusion.

Every business owner in America knows they're supposed to do something with it. About 4% of them actually know where to start. The rest are paralyzed.

That's the market.

What brokering actually looks like

You get on a call with a business owner. You ask what's breaking down. Where are they losing time? Where are jobs slipping through the cracks? Where are they paying people to do things that should already be automated?

Then you go find 1-2 AI tools that actually solve that specific problem. You vet them. You figure out whether the setup is realistic for a company their size. You price it.

You come back and present a clear recommendation.

That's the offer. Clarity and vetting.

A landscaping company billing $2M a year will pay $3,000 to $8,000 for that. I've seen it. I've done it.

Why this works when everything else doesn't

I've watched people try to get into AI by learning to prompt better. By studying machine learning. By building their own automation stacks.

Some of them are sharp. Most of them are still studying.

The business owners don't care how the thing works. They care whether it fixes their problem. The broker sits in the middle. You understand just enough on both sides to translate.

That's a real job. It's always been a real job. Mortgage brokers, insurance brokers, freight brokers — they all do the same thing. They charge to match the need to the solution.

AI brokering is the same model. The need is exploding. The translators are almost nonexistent.

What's actually hard about it

I'll be straight with you.

The technical part is learnable in weeks. The business development is where most people stall.

You have to actually talk to business owners. You have to get comfortable saying "I can help you figure this out" before you feel like you know everything. You won't know everything. Nobody does. AI changes every 3 months.

The people making money at this are the ones who got started before they were ready and figured it out on the calls.

2008 took everything I'd built. When I started over, I didn't wait until I felt ready. I couldn't afford to.

You probably don't have a 2008 forcing your hand. But you might have a job you're tired of, or a ceiling you keep bumping up against, or just a quiet sense that there's an opportunity here and you're about to let it pass.

I've been on the wrong side of that feeling before.

One thing to do this week

Pick one local business you already know something about. A restaurant. A gym. A trade contractor. A medical office.

Ask yourself: where are they losing time? Where do things fall through the cracks?

Write down 2-3 answers. Then search for AI tools that address one of them.

You don't need to pitch anything. You don't need a deck. Just run the exercise and see what you find.

That's the whole model, compressed.

Nate

P.S. If you've got questions about starting this — how to price it, how to find clients, what tools to actually learn first — reply to this email. I read them. Or DM me "CLARITY" on LinkedIn.