Your next customer might not Google you.

Last week I watched my niece figure out where to take her car. She didn't open Google. She didn't look at Yelp. She didn't call her dad, which is the thing I kept waiting for her to do. She opened ChatGPT on her phone, typed a sentence about a weird noise her brakes were making, and asked it where she should go. It gave her two shops. She picked one. She went.

She's 24. She is not an anomaly. She is the customer every shop owner I know is going to meet more and more of, and the moment I started paying attention to this pattern I couldn't unsee it. Younger adults. Tech-comfortable older adults. Anyone who's gotten tired of sifting through ten blue links and four ads to find a decent recommendation.

What's shifting isn't where people go first. It's what they consider an answer. A Google search returns a list — the customer has to do the work. An AI search returns a sentence — the customer doesn't have to do anything. Once you've had one conversation where the AI told you to try a specific shop and the specific shop turned out to be exactly right, you stop scrolling. You trust the sentence.

For the shop on the other end of that sentence, this is either the biggest opportunity in local search in a decade, or a quiet disaster playing out in slow motion. Which one it is depends on a handful of things most owners aren't thinking about yet.

What the AI is actually doing.

Here's the part nobody explains in plain language. When someone asks an AI where to take their car, the AI is not pulling the answer from some magic internal database of "best shops in America." It's doing something much simpler and much more knowable: it's reading the live internet, on the spot, and deciding.

Specifically, in about one second, it:

1. Reads the question and figures out what the customer actually needs — not just the literal words, but the intent. "Brakes grinding" means brake inspection, probably urgent, probably under $500 if it's just pads.
2. Searches the live web for shops in the relevant area that match.
3. Reads the websites of those shops and decides which ones to name.

That last step is where everything happens. The AI is not running a popularity contest. It's asking: does this place actually do this? Is the information specific enough to cite? Does the rest of the internet agree this shop exists and operates the way its website says? Would I stake my recommendation on this?

If your website reads like you care about what you do — if it's specific, current, and consistent with everything else that says your name online — the AI can recommend you with confidence. If it doesn't, the AI moves on.

Where most shops are invisible.

I've looked at a lot of shop websites over the past year. The specific problems are boring and consistent.

Service pages that say nothing.

"Our skilled technicians provide top-quality brake service to keep you and your family safe on the road. Contact us today!" This sentence exists on maybe a thousand shop websites. An AI reads it and learns nothing it can use. No prices, no parts, no warranty terms, no specifics about what actually happens when you bring your car in. Nothing to stake a recommendation on. The AI moves to the next shop.

Information that doesn't match across the internet.

Your website says you open at 7 AM. Your Google listing says 8. Your Facebook page says you close at 5 but also 6, depending which post you read. An AI sees all this and makes a reasonable decision: this shop isn't paying attention to its own information, so I won't stake my answer on it. Consistency is a trust signal, and it's one of the easiest to control.

Websites that look abandoned.

A copyright date in the footer from four years ago. A blog with one post from 2021. A staff page with people who haven't worked there since before COVID. The AI reads all of this and concludes what a customer would conclude: this place might not be a going concern anymore. Don't send people there.

The AI is not asking "are you fancy?" It's asking, "are you real?"

AI vs. AI slop.

There's another wrinkle worth knowing about. Ironically, the same technology that's powering all of this is also being used to churn out floods of generic, hollow, algorithm-bait content. "AI slop" — the stuff that exists on the internet not to help anyone but to fill pages and rank for keywords. Most small business websites in 2026 either are slop or have been slop-adjacent for years.

The AIs your customers are using are getting better at spotting slop and ignoring it. They reward the opposite — specific, honest, useful, human-sounding writing. Which means the shops that win aren't necessarily the ones that hire the best SEO agency. They're the ones that sound like a real shop run by real people who know what they're doing.

This is, to be clear, a weird moment. The AI systems rewarding human-sounding content are themselves AI. But the mechanics work out the way they work out. Specificity wins. Genericity loses. "2-year / 24,000 mile warranty on most brake work, using Wagner OEX pads, same brand the factory puts on new Civics" beats "top-quality brake service" every time.

What to do about it, realistically.

Here's the uncomfortable part. Nobody is coming to do this for you. There is no corporate initiative, no franchise group task force, no SEO vendor with a magic subscription that solves this. The shops that show up in AI answers five years from now are going to be the shops that took it seriously right now.

If I were advising a shop on where to start, it would be these four things, in this order:

1. Read your website out loud like an AI would. Is every service page specific enough to be useful to someone who doesn't already know you? Or does it all sound like it could apply to any shop in America? If it reads like a template, it is one. Rewrite it.
2. Fix every place your information lives. Google Business Profile, Facebook, Yelp, Bing, Apple Maps, your own website. Phone, hours, address, services. Make them all agree with each other. Then keep them agreeing.
3. Handle the unsexy fundamentals. Google Business Profile filled out completely. Bing Webmaster Tools registered. Schema markup on your service pages. Steady reviews flowing in every month. None of this is glamorous. All of it still matters — a lot.
4. Test it yourself. Open ChatGPT, Claude, Perplexity. Ask each one: "Best auto repair shop in [your town] for [your specialty]." Read the answer carefully. That's what your customer is seeing. Fix whatever you see.

Why I care about this.

I spent thirteen years in the Navy learning that operations succeed or fail based on whether the people running them pay attention to the small, boring things. Then I spent eight years in oil and gas learning the same lesson in a different uniform. I came home to auto service because the pattern doesn't change. The shops that take care of the small details are the shops customers come back to. The systems that reward that behavior are the systems that survive.

AI search is the newest system. It rewards the same things every other serious system rewards: clarity, consistency, specificity, honesty. If you run a shop that already operates that way, none of this is actually a lift. You just have to translate what you're already doing into something a machine can read.

If you don't — well, the customer asking ChatGPT about the weird noise her brakes are making isn't going to wait for you to figure it out.