Your website says you serve 43 towns. You drive to 6.

There's a page on almost every trades website in America. It's called "Areas We Serve" or "Service Areas," and it lists every town within an hour of the shop. Forty-three of them, say, each with its own page: "Plumber in Riverside," "Plumber in Cedar Grove," and forty-one more.

Behind that page: one van, one owner, and six towns he'll actually drive to.

What follows is a worked example. The shop is made up, standing in for a page you'll find on trades sites everywhere. No customer story here. But if you own a trades website, some version of this page is probably on it, and what our onboarding does with it is the point of this post.

The 43 towns are there on purpose

The page isn't a lie, exactly. It's local SEO done by the book. The standard playbook for a service business is a page per town you want to rank in, because Google won't surface you in a town your site never mentions. Search Engine Land's guide to service area pages walks through the tactic, and also names the failure mode: brands publishing "thousands of thin pages for every possible town."

So at some point an SEO, or a web guy, or a nephew asked the owner "what areas do you cover?", the owner said "oh, all around here," and 43 town pages got generated. They did their job. They ranked. Nobody ever asked them to be true.

That's the gap: marketing reach and operational truth are two different maps, and almost every trades site carries both. The website's map was drawn for Google. The real map gets drawn by gas prices, traffic, and how much the owner hates the bridge at rush hour. The service area is where the two maps drift furthest apart.

A machine reads that page differently than a person does

A homeowner in town #38 who lands on the page still calls and asks, "do you really come out here?" People discount marketing copy without thinking about it.

Software doesn't. Put a chatbot on that site, ground it on the pages as written, and it will tell the homeowner in town #38 "yes, we serve your area" with full confidence, because the site says so 43 times. Now the owner has a lead 50 minutes away priced like it's around the corner. Or an angry customer who was promised a visit nobody's making.

An agent that answers from the marketing version of your business will make promises the truck can't keep. That's the trap in bolting a bot onto a website and calling it a front desk.

What the crawl does with the 43 towns

ServiceFront onboarding starts from a URL. The crawl reads the site and drafts a profile: services, hours, and yes, the claimed service area, all 43 towns.

Here's the part that matters: the crawl doesn't know the truth either. No crawl does. Only the owner knows which six towns are real. What the crawl can do is notice that the claim looks implausible: a 43-town list next to one shop address, with everything else on the site pointing at a much smaller radius. Those don't fit together. So instead of importing the marketing map as fact, onboarding flags it and asks the owner directly: your site lists 43 towns. Where do you actually go?

The flag is the honest move. Silently trusting the page would be easier, and wrong.

Two minutes of cutting

The owner answers with taps, not paragraphs. Six towns: yes, always. A ring around them: we'll go, but there's a $40 trip fee, and the agent should say so up front. The other thirty-odd: drop them.

That's the verify step for the service area, and it changes what the agent is allowed to claim. The confirmed boundary becomes the fact the agent checks against, and the trip-fee note travels with it. ServiceFront's agent has one rule it never breaks: it only asserts what the owner confirmed. Anything unconfirmed it treats as unknown and says "let me check" instead of guessing.

The trip-fee answer beats the false yes

Same homeowner, town #29, asking "do you serve my area?"

Grounded on the raw website: "Yes, we proudly serve your town!" A promise, and a false one.

Grounded on the verified profile: "We can get out there. You're outside our usual area, so there's a $40 trip fee. Want me to have Dave give you a call to set it up?"

The second answer reads worse in a marketing meeting and better everywhere else. The homeowner knows the terms before anyone drives anywhere. The owner gets a lead who already said yes to the trip fee, delivered to his inbox with the address and the problem written down. The jobs that would've meant a 50-minute drive for a $150 service call filter themselves out. And the false yes costs a wasted trip at best, a one-star review at worst.

Keep the 43-town page. It's still doing its Google job, and cutting it would throw away rankings you paid for. Just don't let it be the thing that answers your customers.

See what your own site claims

If you're curious what a crawl makes of your website, paste your URL into the box on servicefront.ai and ask it what your customers ask. If your service-area page is bigger than your service area, you'll meet the same question we'd ask you: where do you actually go?