36,000 HVAC contractors have no website: how invisible operators lose jobs and the four-page minimum that fixes it

I run the dataset behind Full Stack HVAC. We currently track 123,000+ HVAC Google Business Profiles in the United States and have run over 1.1 million Lighthouse audits across the contractor sites that do exist. Two findings from that dataset are the starting point for this post: about 30% of GBPs have no website URL at all, and the sites that do exist are thin, a median of 4 pages when we crawl them, with 99.7% under six.¹

The first number is the easier story to tell. The second is the one that compounds.

The 36,000 invisible operators

A residential HVAC contractor with no website in 2026 is not necessarily a poor technician. Many of the operators in that no-website segment came up through referral, run a tight book, and never built the digital scaffolding because the truck and the network kept the schedule full. They are in the no-website segment by default, not by decision.

The problem is that the referral channel is shrinking against the search channel. A homeowner whose furnace dies at midnight does not call the neighbor; they search on their phone. A Google Business Profile with no website attached costs the operator three things that compound daily.

1. The emergency call. The midnight search ends at the first contractor with a tap-to-call number and a clean profile. No website is one of the heuristics homeowners use to filter, even informally.
2. The AI surface. ChatGPT, Perplexity, and Google's AI Overviews build recommendations from the indexable web. A business with no website cannot land in those answer sets at all. (For the mechanics of how LLM recommendations are sourced, see the ChatGPT post.)
3. The trust check. Referred homeowners still google the name before they book. A contractor with no findable site in 2026 reads to a 2026 homeowner the way a contractor with no phone number read to a 2006 homeowner.

None of this argues that the operator's work is bad. It argues that the digital surface is the gate, and the gate is closed.

The thin-site gap is the gap that compounds

The page-count number is the one that shapes the long-run trajectory. Page count is the surface area Google has to match a contractor to a query, and the typical independent has very little of it: four pages of "Home / About / Services / Contact" cannot rank against a national or franchise operation that publishes a separate page for each city, each service line, each financing option, each seasonal landing page, and each FAQ cluster. The big players treat the website as a content engine; the independent treats it as a brochure.

Peter Troast of Energy Circle has written publicly about exactly this gap: the best-trained, best-equipped independent operators routinely have the worst websites in their markets, because their attention is on the work and the franchise marketing arm is on the search results.³ The mismatch is not new, but the consolidation pressure in 2025 and 2026 makes it more expensive every quarter that an independent runs without addressing it.

Having a website is not the same as having a working website

The Lighthouse data is the second half of the picture. Most contractor sites are fast on a phone (the median mobile site paints in under a second), so the headline is not that everyone is slow. It is that a stubborn slow tier exists, and it is brutal:¹

Google's published Core Web Vitals thresholds are unambiguous: an LCP under 2.5 seconds is "good", 2.5 to 4.0 is "needs improvement", and anything above 4.0 is "poor".² The 11% of sites in the "poor" tier average about 9.5 seconds on a phone, more than double the threshold. If your site is in that tier, it stalls on mobile, where the customer is, even when it looks fine on the desktop the operator checks.

Slow contractor sites are a separate post in this series; see the mobile-speed analysis for the per-component fix list. The point for this post is that 36,000 contractors with no website at all are joined by a much larger group whose site is technically present but functionally absent on mobile.

The minimum viable HVAC website is four pages

The four-page configuration that clears the decision-stage checks for residential HVAC, in priority order:

1. Home. Visible phone number in the header that taps to call. Service area cited in plain text (not an image). A recent-reviews widget that surfaces dates, not just star averages. Hero loads in under 2.5 seconds on a phone.
2. About / Team. Real photos of the operators in the trucks. Real licence numbers, certifications, and insurance information. The audience that books on quality reads this page before any other.
3. Services. Individual sections for each Google Business Profile category the operator claims. "AC repair", "heating installation", and "indoor air quality" each get their own treatment, not a bulleted list under one page.
4. Contact. Tap-to-call. Embedded service-area map. Business hours. Financing language if applicable. A short form for non-emergency inquiries.

Specialty pages (per-city service stubs, equipment-specific landing pages, blog posts on the operator's actual installs) compound on top of these four. They do not replace them.

The cost question

The contractor-website market has two visible price tiers in 2026:

• Template-based builders at the $19 to $199 / month range (HVAC-specific templates from Marketing 360, Service Direct, and the in-house builders inside ServiceTitan, Housecall Pro, and Jobber).
• Custom agency builds at the $2,000 to $17,500 fixed-cost-plus-retainer range (Energy Circle, Scorpion, Footbridge Media, and similar trade-focused agencies).

For an operator at the small-independent end of the dataset (one to ten trucks, the typical FSH-tracked independent), the template route delivers most of the conversion value at roughly 1% of the agency capital cost. Custom is the right answer when the operator is competing for branded queries at a high volume and the SEO arms race is the strategy. The two are not in competition; they are sequenced.

What to do this week

1. Check your own GBP. Confirm that the website URL is present and that it points to a working page on the operator's primary domain (not a Facebook page or a directory listing).
2. Run a Lighthouse audit on your site. Chrome DevTools, Lighthouse tab, mobile profile, page load. Note the LCP number. If it is above 4.0 seconds, the site is in Google's "poor" tier.²
3. Count your pages. A Google site:yourdomain.com search returns the indexed count. Four pages is the independent baseline in our data; building past a dozen, with real per-city and per-service pages, is what starts to compete on long-tail queries.
4. Pick a tier and budget it. $19 to $199 / month for a template build inside the dispatch-software stack; $2,000+ for an agency build when the operator is past the SMB tier. Either is better than the no-website status quo.

The point of the dataset behind this post is not to shame the 36,000 contractors without sites. It is to surface the size of the segment and to make the case that the four-page minimum is achievable cheaply enough that no operator should still be sitting outside it in 2026.

For the dispatch software that bundles a basic site into the monthly stack, see the dispatch-software comparison. For the Google Business Profile work that brings homeowners to the site in the first place, see the GBP completeness analysis.