A growing share of your future customers will never see a page of search results.
They ask a question — "who's a good roofer near Everett?" "best family dentist in Spokane?" — and an AI assistant answers with two or three names, a sentence about each, and a link. ChatGPT does this. Claude does this. Perplexity does this. Google now does it directly at the top of search with AI Overviews.
If your business is one of the names, that is a customer conversation you did nothing to earn in the moment — the work happened months earlier. If it is not, you were eliminated before you knew the customer existed.
Answer engine optimization — AEO — is the work of becoming one of the names. Here is how it actually functions, without the hype.
Language models do not keep a secret directory of good plumbers. When someone asks for a recommendation, the assistant typically does some combination of three things:
Draws on training data — what the model absorbed about businesses from the public web
Searches live sources — reviews, directories, articles, business profiles — and synthesizes what it finds
Reads structured data — the machine-readable descriptions of businesses that well-built websites provide
Notice what those three have in common: they are all downstream of how clearly and consistently your business is described across the web. An AI engine cannot recommend a business it cannot confidently understand. Ambiguity that a human customer would shrug off — inconsistent service descriptions, unclear service areas, thin review history — makes a machine skip you for a competitor it understands better.
This is the most important thing to understand before spending money: answer engines lean on largely the same signals search engines do. Reviews, accurate listings, crawlable sites, authoritative mentions, clear content. If someone pitches you AEO as a magic replacement for all that boring foundational work, keep your hand on your wallet.
What AEO adds on top of a solid local SEO foundation is machine-focused work that traditional SEO rarely bothers with:
Structured data engineering — schema markup that states exactly what you do, where, for whom, at what price range, with what credentials
Content structured for extraction — pages organized as clear questions and direct answers, because assistants lift answers, not vibes
Entity consistency — your business described identically everywhere an AI might read about it
Presence on cited sources — the directories, review platforms, and local publications AI engines actually pull from when they answer
Technical access — making sure your site is readable by AI crawlers at all, including conventions like llms.txt
Almost nobody in the trades and local services is doing this yet. That is the entire opportunity.
When an AI engine answers "best electrician in Tacoma," it is choosing among businesses whose digital presence was built for an older game. The first electrician in a market whose reviews, structured data, and content are genuinely machine-legible tends to become the default answer — and language models, like search engines before them, are conservative about changing established answers.
We watched this exact dynamic twenty years ago with early SEO. The businesses that moved before their competitors held those positions for a decade. The window looks similar; the difference is that this time it is measurable from the start.
The fair objection to AEO is that it sounds unfalsifiable. It is not, if the work is done honestly. Measurement looks like this:
Define the questions your customers actually ask — dozens of variants, by service and by city
Query the major engines on a recurring schedule and record whether you are mentioned, how you are described, and who else appears
Track which sources the engines cite, and target presence on those sources
Watch referral traffic arriving from AI platforms in your analytics
Report all of it monthly, next to the number that matters: leads
That baseline-then-track loop is the core of our AEO service. The first report is often uncomfortable — most businesses discover the AI engines either do not know them or describe them wrong — and that is precisely the point of starting now.
AEO retainers at Ravenna run $750 to $2,500 per month, frequently bundled with local SEO since the work overlaps heavily and every improvement does double duty. It fits businesses that already have the fundamentals roughly in place and want to be early to the next channel — and it fits businesses in competitive metros like Seattle and Tacoma where the map pack fight is expensive and an additional channel changes the math.
If your Google Business Profile is empty and you have eleven reviews, start with local SEO first. We will tell you the same thing in the audit, for free.
Same discipline, different acronym — generative engine optimization, answer engine optimization, AI search optimization. The industry has not settled on a name. The work is what matters: making your business legible and credible to systems that generate answers.
Yes, and the share grows every quarter — driven less by chatbots than by Google putting AI-generated answers at the top of ordinary searches. Your customers do not need to change behavior at all to be receiving AI recommendations today.
No, and be suspicious of anyone who guarantees it — the engines change constantly and nobody controls them. What honest AEO does is systematically improve every input the engines are known to use, and measure whether mentions improve. In markets where few competitors are trying, the odds are very good.
For AEO, yes. Structured data is the difference between an AI inferring what your business is and being told outright. It is also developer work — hand-written, validated, page-specific — which is why an agency that builds websites professionally has a real advantage doing it.
Every established marketing channel was once an obvious opportunity that most businesses ignored until it was expensive. AI recommendations are at that stage right now, and for local services the field is nearly empty.
Start with a free AI visibility audit: we query the major engines with the questions your customers ask and show you exactly what comes back — who gets recommended, how you are described, and what it would take to change both. Request one here.
Ravenna is a Seattle-based team that designs and ships web platforms, mobile apps, and Laravel & Statamic builds for companies that need them done right.
Talk to Us