Answer Engine Optimization for Trades — What AEO Actually Means for a Contractor

Answer engine optimization for trades is the work of getting your shop named inside an AI-written answer — the paragraph ChatGPT, Perplexity, or Google’s AI Overview produces when somebody asks a question. Not the blue links underneath it. The answer itself. That’s the whole idea. Everything else in this post is detail.

We’re two working builders out of Rigby, Idaho. We started paying attention to this when homeowners in Idaho Falls and Ammon quit saying “I googled you” and started saying “I asked the AI thing and your name came up.” Same customer, same job, different front door. Getting through that door is a different task than ranking, so it got a different name: AEO.

This is the plain-English version. What an answer engine is. Why getting cited matters more than getting ranked. What a page has to look like to get quoted. Where schema fits. What to quit doing. We have separate posts on each tactic — this one is the map that ties them together.

TL;DR — A search engine hands a homeowner a list of links. An answer engine reads the list for them and writes a paragraph, naming a handful of sources. Answer engine optimization for trades means structuring your pages so yours is one of the sources it names. The pages that get quoted answer the question in the first two sentences, use question-shaped headings, state plain facts a machine can lift without guessing, and name real towns. Schema helps a machine confirm what you already said in plain text — it does not fix a page that says nothing. In a thin market like Eastern Idaho, almost nobody has done this yet, which makes it the cheapest advantage on the board.

What an answer engine is, and how it differs from a search engine

A search engine hands you a list. You type “concrete driveway Idaho Falls,” you get links, and you pick one. Sorting them out is your problem.

An answer engine reads the list for you and writes the reply. ChatGPT, Perplexity, and Google’s AI Overviews each do a version of this. They pull from web pages, summarize what they found, and name some of the sources they leaned on. The homeowner reads a paragraph instead of scanning ten headlines.

The practical difference is simple. On a search engine, position ten still gets seen by somebody who scrolls. In an answer, there is no position ten. There is the paragraph, and there are three or four sources hanging off it. You are in it or you are invisible.

Google publishes its own guidance on how sites show up in AI features, and it’s worth reading straight from the source instead of from an agency deck: AI features and your website. Anthropic publishes documentation for its own models and tools at anthropic.com. Where a vendor hasn’t said how something works, we don’t pretend to know.

Why citation beats ranking

Ranking is a position. Citation is a mention. Not the same trade.

Think about how a homeowner in Shelley actually uses this. She asks a chat tool who does stamped concrete near her and what it runs per square foot. She gets back a paragraph with two or three company names in it. She doesn’t open ten tabs. She calls one of the names.

Being named in that paragraph does something a page-one ranking never did: it comes with an implied recommendation. The machine put you in the answer. That reads like a referral, not like an ad.

The flip side is that citation is winner-take-few. A list has room for ten. An answer has room for three. The bar is higher, but the payoff per slot is bigger. That’s the trade AEO is making.

What answer engine optimization for trades actually changes on your website

Almost nothing about the plumbing. You still need a real WordPress site, real service pages, a claimed Google Business Profile, and reviews. AEO doesn’t replace any of that. If your Google Business Profile isn’t ranking, fix that first. AEO on top of a broken foundation is a wasted weekend.

What changes is how the words on the page are arranged. A search engine can reward a page that buries the answer three scrolls down, because the human will scroll. An answer engine is pulling a sentence or two. If your good sentence is in paragraph nine, underneath a hero video and a company history, it may as well not exist.

So the work is mostly rewriting, not rebuilding. Same facts, different order.

The page structure that gets quoted

Four things, in order of how much they matter.

Answer in the first two sentences. Whatever the page is about, say the answer immediately. A page titled “How much does a concrete driveway cost in Idaho Falls” should open with a range and what moves it. Not with “At Acme Concrete, we pride ourselves…”
Question-shaped H2s. Write headings the way a homeowner types. “How long does a stamped patio take to cure?” beats “Curing.” Each heading is a chance to match a real question, and the paragraph under it is the candidate answer.
Plain-text facts. Anything a machine has to infer from an image, a PDF, or a slider is a fact you didn’t state. Service area, price ranges, licensing, lead times, brands you install, whether you do warranty work — put them in text. A photo of your license doesn’t say your license number.
Named towns. Say Rigby. Say Ucon. Say Iona, Shelley, Blackfoot. “The greater Eastern Idaho area” is not a place. When somebody asks a chat tool for a framer in Rexburg, the pages that say Rexburg in ordinary sentences are the ones in play.

We wrote a longer, hands-on version of this: how to write a contractor page that AI actually cites. That post is the how. This one is the why.

Schema’s role — and where it stops

Schema is structured data. It’s a block of code on your page that restates your facts in a format machines were built to read: name, address, service area, hours, services, questions and answers. The vocabulary is public and free at schema.org, and Google documents what it looks for in its local business structured data guidance.

Here’s the honest framing. Schema doesn’t make you a better answer. It makes you an easier one to confirm. When a machine reads “we serve Bonneville, Jefferson, and Madison counties” in your paragraph and then finds the same list in your structured data, that’s two confirmations of the same claim. Ambiguity is what gets you skipped.

What schema will not do is rescue a page with nothing to say. Marking up an empty page just tells the machine, in a well-formatted way, that the page is empty. Write the plain text first, add the markup second. Templates: LocalBusiness schema and FAQ schema.

Why AEO is a bigger opening in Eastern Idaho than in Phoenix

This is the part that should get your attention if you run a shop in Bonneville, Jefferson, Madison, or Bingham County.

An answer engine can only quote what somebody wrote down. In a big metro, forty roofers have already published a priced-out, question-headed, town-named page for every service they offer. The competition for those three citation slots is a knife fight.

Out here it isn’t. Pull up the trade sites in Idaho Falls, Rexburg, Ammon, Ucon, and Shelley and read them the way a machine would. Most are five pages. Most describe the service area as “Eastern Idaho and surrounding areas.” Most never state a price range, a lead time, or a warranty term anywhere in text. There is very little for an answer engine to grab, which means the first shop in a town that writes it all down plainly has almost nobody to beat.

That’s the whole opportunity. Not a clever hack. Just being the only business in Ucon that answered the question in writing. Bryce did this on his own site, SwagerBuilds, after a cleanup — rewrote the pages to answer questions directly, named every town he works in, and went from nothing to top-20 for forty-plus Eastern Idaho keywords in sixty days. That’s our own site, our own numbers, and the only case study we use.

The window closes eventually. Right now it’s the cheapest advantage available to a two-truck operation.

What to stop doing

A short list, because the subtractions matter more.

Stop opening pages with your company story. Put it on the about page. The service page opens with the answer.
Stop hiding facts in images. Price sheets, service-area maps, badges, certifications. If it’s only in the picture, it isn’t on the page.
Stop writing town pages that are the same page with the town swapped. Thin duplicated pages are a liability, not a strategy. Details here: the contractor service-area page template.
Stop chasing word count. A tight 700-word page that answers six questions gets quoted more than a padded 2,500-word page that answers one.
Stop paying for “AI visibility” that’s just a report. A monthly PDF showing whether you appeared in ChatGPT is a measurement, not a change. Ask what got rewritten.

How the pieces fit together

AEO is the umbrella. Under it sit the individual surfaces, and each has its own quirks. Perplexity leans hard on citing sources, so Perplexity for contractors is mostly about being quotable. ChatGPT behaves differently depending on whether it’s browsing, which we covered in how to rank in ChatGPT as a local contractor. Google’s AI Overviews sit on top of regular search results, which makes classic SEO the prerequisite — see the AI Overviews field guide.

Different surfaces, one job underneath: write pages that state real answers in plain text, confirm them with structured data, and name the towns you actually drive to. Do that and you’re working on all three at once. That’s why we don’t sell “ChatGPT optimization” as a separate line item. It’s the same page work.

FAQ

What is answer engine optimization for trades?
It’s structuring a contractor’s website so AI answer tools like ChatGPT, Perplexity, and Google’s AI Overviews name that contractor inside the answer they write, instead of just listing the site as one link among many.

Is AEO different from SEO, or just a new name for it?
It’s a layer on top, not a replacement. AEO needs the same foundation — a real site, a claimed Google Business Profile, reviews, and pages that rank. What it adds is a specific way of arranging the words so a machine can lift a clean answer out of them.

Do I need schema markup to get cited?
Not strictly, but it helps. Schema restates your facts in a format machines read without guessing, which reduces ambiguity. It won’t rescue a page that never stated the fact in plain text to begin with.

How do I know if an AI is citing my business?
Ask it. Open ChatGPT or Perplexity, type the question a homeowner would type — your trade plus your town — and read what comes back. Do it monthly and write down who got named.

How long does this take to show up?
Longer than a Google ranking change and less predictably. These tools pull from indexed pages, so the page has to be crawled and indexed first, and then it has to actually be the best available answer.

Is this worth doing for a two-person shop in a small town?
Out here, more than anywhere. In Rigby or Ucon or Iona, most competitors haven’t written a single plainly-answered page. Being the only business in town that wrote it down is the whole play, and it costs a few evenings.

If you want your pages rewritten so machines and homeowners can both get a straight answer, we do that. Start here →