How a 60-year-old, five-star, neighbor-recommended HVAC and appliance company was nearly invisible to the AI assistants now answering "who should I call?" — and the diagnosis that explained why.
Segment: Local service business (home services) · Service: AI visibility diagnosis (SEO / AEO / GEO) · Status: Findings delivered; fixes in client's hands
Ask anyone in St Charles County, Missouri who to call when the furnace quits or the refrigerator stops cooling, and a name comes up again and again. On Nextdoor, neighbors type it out from memory — the company, the town, the phone number — and vouch for it without being asked. On Angi, it carries a perfect five-star rating. The business has been doing the work since 1961, it's family- and female-owned, it's an American Standard and Mitsubishi dealer, and the owners answer their own phones.
By every human measure, Quality Appliance Service is exactly the kind of business a recommendation engine should surface first.
It mostly doesn't. And the reason has nothing to do with the quality of the work.
The paradox
The way people find a repair company is changing underneath everyone's feet. "Google it" is quietly becoming "ask the assistant" — who's the best appliance repair near me, typed into ChatGPT, or spoken to Gemini, or read off the AI summary at the top of a search. Those answers aren't built from a lifetime of neighborhood goodwill. They're built from structured data, consistent listings, and the handful of sources the engines trust — and a sixty-year reputation that lives in people's heads doesn't automatically travel into any of them.
That's the gap we were asked to look at. Not "is this a good company" — the reviews settle that — but "when the machine is asked to recommend one, can it confidently name this one?"
What the diagnosis found
We audited how Quality Appliance Service appears to the search and AI layer the way an engine would read it, not the way a loyal customer would. The picture was consistent: a strong real-world reputation sitting on top of a digital identity the engines can't quite resolve.
The business had no structured "this is who we are" data on its site — none of the machine-readable name, address, hours, service-area, and service-type information that Maps and AI assistants lean on to ground a local business. Its listings were fragmented: duplicate profiles on one major directory, and an address that read as one town in some places and a neighboring one elsewhere. The brand name and the website's domain didn't match, giving the engines a weaker thread to follow from "the company people search for" to "the company's website." The social links in the site's own header led nowhere. And small signals of neglect — a copyright year frozen two years in the past, an "about" page still counting to a milestone the company passed years ago — quietly told both visitors and freshness-sensitive engines that nobody was minding the store, even though the people behind it never stopped showing up.
None of these is dramatic on its own. Together, they explain how a company this trusted can be this hard for an assistant to name with confidence.
The hidden asset was just as clear: those organic Nextdoor recommendations. Real neighbors, recommending by name, unprompted — the exact kind of community signal the newer engines weigh heavily. The reputation isn't missing. It's just not yet wired into the places the machines read.
What we know, and what we don't
We hold ourselves to a simple rule: a diagnosis is only as good as its honesty about its own limits. So every finding is filed under what we could actually confirm.
- Verified — confirmed directly. No structured business data on the site. Dead social links in the header. A stale copyright year and an out-of-date company-age claim. Duplicate listings and an inconsistent address across directories. A brand-name-to-domain mismatch. A five-star Angi rating and genuine, organic Nextdoor recommendations.
- Directional — strongly supported, not laboratory-proven. That this entity fragmentation is why a business this well-reviewed struggles to surface in AI and Maps answers; that the community layer is its single strongest, most buildable asset.
- Pending — needs one access-gated check. Whether the Google Business Profile — the biggest single lever for a local company — is claimed and fully optimized. Whether the site permits the AI search crawlers. Whether leads are being tracked at all.
- Unconfirmed — can't be known from the outside. Exactly how each assistant (ChatGPT, Perplexity, Gemini) describes the business today, and where it lands against competitors, without a live test inside each one.
A glossier write-up would blur those lines. We'd rather a business owner know precisely which findings are bankable and which still need a five-minute check — because that's the difference between a report and a sales pitch.
Why this is the moment
Home services have barely been touched by AI search yet. That's not a reason to wait — it's the reason to move. The companies that ground their identity before the engines mature are the ones the engines will reach for once these questions become routine. Quality Appliance Service has the hardest part already done: six decades of trust and a town that recommends it by name. What's left is unglamorous and fast — claim and tighten the Google profile, add the structured data, consolidate the listings, fix the dates, point the community goodwill at the surfaces the machines read.
How we work
We diagnose; we don't take over. The findings above were delivered as a prioritized report the owner can read in plain language, paired with a step-by-step work order — a literal check-off list — that hands cleanly to whoever maintains the site. No retainer to run the company's marketing, no contract to rank for a keyword, no promises about a number we can't honestly guarantee. We identify what's wrong and what matters most, the client's team executes, and we follow up at 30, 60, and 90 days to confirm the fixes took hold.
For a business that earned its reputation the hard way, that's the point: make the machines see what the neighborhood has known for sixty years.
Published with the client's permission. Every finding is drawn from an actual diagnostic audit and presented findings-first, with no claim of ranking or revenue outcomes.