When a homeowner's AC dies in a Texas July, they increasingly ask an AI who to call — and the AI names two or three HVAC companies. It picks based on review volume and recency, Google Business Profile completeness (especially emergency hours and service-area cities), consistent directory listings, HVACBusiness schema markup, and website pages that state prices and response times in plain text. HVAC has two distinct AI battles — the emergency call and the replacement purchase — and they're won with different signals.
The Two HVAC Searches AI Now Answers
HVAC is unusual among the trades because customers come in two completely different modes, and AI answers both.
Mode 1: The emergency ("AC blowing warm air, who can come today")
Panic mode. The homeowner asks Google or an assistant who can come now, and the AI's answer favors companies whose data screams responsiveness: emergency or 24/7 hours listed on the Business Profile, recent reviews mentioning same-day service, and the specific city named in the service area. Whoever the AI names gets the call within minutes. There is no comparison shopping at 102°F.
Mode 2: The replacement ("how much does a new AC system cost, who should install it")
Research mode — a $8,000–$15,000 considered purchase. These buyers have long conversations with AI before calling anyone: What size system do I need? Is a heat pump worth it in North Texas? What's a fair price for a 4-ton, 16-SEER install? What do these new refrigerant rules mean for my quote? The companies whose content answers those questions get woven into the AI's responses across the whole conversation — so by the time the buyer requests quotes, one company has been their adviser for a week. Guess who's on the shortlist.
Most HVAC marketing only fights battle 1. The companies quietly winning right now publish honest, plain-language answers for battle 2 — and the AI rewards them in both, because educational content builds the authority that emergency recommendations draw on.
What the AI Checks Before Naming an HVAC Company
| Signal | What wins for HVAC specifically |
|---|---|
| Google Business Profile | "HVAC contractor" primary category; services itemized (AC repair, installation, heating, maintenance plans); every service-area city listed by name; emergency/after-hours availability shown; photos of real techs and real installs, added monthly |
| Reviews | Volume and recency above all; reviews that mention speed ("came same day"), specific cities, and specific jobs ("replaced our condenser") — these phrases get quoted by AI verbatim |
| Citations | Identical NAP on Google, Yelp, Angi, BBB, Nextdoor, and Apple Maps; license number displayed consistently (TACLA/TACLB in Texas) — license verification is a trust signal machines can check |
| Schema markup | HVACBusiness type with serviceArea and openingHours; Service entries for repair/install/maintenance with price ranges; FAQPage on common questions; AggregateRating |
| Website content | Price ranges in plain text; a page per major service; question-phrased headings ("How much does AC repair cost in [city]?"); response-time promise stated near the top |
The Pricing Page Question (Just Publish Ranges)
The single most common objection we hear from HVAC owners: "I can't publish prices — every job is different." But the replacement buyer is asking the AI for numbers right now, and the AI is answering — from whichever sites publish them. If that's not you, the buyer's price anchor is being set by a competitor or a national lead-gen site.
You don't need a menu; you need honest ranges with the variables named: "A full system replacement in the Dallas–Fort Worth area typically runs $X–$Y depending on tonnage, SEER rating, and ductwork condition." That sentence is extractable, it earns the citation, and it pre-qualifies your calls. The companies that publish ranges aren't giving away an edge — they're taking the adviser seat at the start of a five-figure decision.
The 30-Day HVAC Fix List
- Week 1 — Profile & baseline: Complete every GBP field, list all service-area cities, set emergency hours, add 10+ real photos. Record your baseline: ask ChatGPT, Gemini, and Perplexity "best HVAC company in [your cities]" and screenshot the answers.
- Week 2 — Reviews engine: Set up a same-day review ask for every completed job (text with direct link). Brief techs to mention it at the door. Respond to every existing review, including old ones.
- Week 3 — Citations & schema: Fix NAP across the top 15 directories. Deploy HVACBusiness, Service, FAQPage, and AggregateRating schema on the site.
- Week 4 — Content: Rewrite the homepage to state services, cities, response time, and license up top. Publish a pricing-ranges page and one FAQ page answering the ten questions your office hears most. Re-run the baseline questions and compare.
Timing note for Texas: AI engines re-crawl on their own schedule, so signals built in spring decide who gets named during the summer emergency rush. The best month to do this work is the month before you need it.
Frequently Asked Questions
How do homeowners use AI to find HVAC companies?
Two modes: emergency searches, where AI names companies signaling fast response and strong recent reviews, and replacement research, where buyers hold multi-question conversations about sizing, efficiency, and cost before calling. Responsiveness signals win the first; clear pricing and educational content win the second.
Why does my competitor show up in ChatGPT and I don't?
Almost always: more or fresher reviews, a more complete Business Profile with cities listed, cleaner directory listings, schema markup, or plainer website text. AI recommends the company whose data it can verify — not necessarily the better contractor.
Should my HVAC website list prices?
Yes — honest ranges with the variables named. AI answers "how much does a new AC cost" from sites that publish numbers; "call for a quote" gets skipped. Ranges also pre-qualify your calls.
What schema markup should an HVAC company use?
HVACBusiness with service area, hours, and emergency availability; Service entries with price ranges; FAQPage markup; and AggregateRating tied to reviews.
How fast can an HVAC company get visible in AI search?
Profile and website fixes can appear in Google AI Overviews within weeks; assistant recommendations build over one to three months as reviews accumulate. Build signals before peak season.
Want the 30-day fix list done for you?
The Visibility Ignition Sprint covers all five signals — profile, reviews engine, citations, schema, and content — in 30 days for a $599 flat fee, with a before-and-after report showing exactly how AI describes your company. Built for HVAC companies with no marketing team.
Start Your SprintMore: the complete local AI search playbook · the master guide to SEO, AEO & GEO