How much does a missed HVAC lead really cost?

No single dollar figure is honest here, and you should be skeptical of anyone who gives you one. What's real and well-documented is the mechanism: contact odds drop sharply the slower you call back, and in our own audit of 352 HVAC-business websites, only 27.2% make any explicit promise about how fast they'll respond at all.

A claim you'll see repeated everywhere is "78% of customers buy from whoever responds first." While building our own research report, we went looking for the original source of that exact number. We couldn't find one: no traceable study, no dataset, nothing to cite. We don't use it, and it's worth treating skeptically anywhere else you see it.

Where does the real evidence come from?

Call data tells a similar story. Invoca's 2026 home-services benchmark found only 52% of callers spoke with a person, and 55% of businesses didn't ask the caller to book the job. Neither figure suggests the work wasn't there. The pattern is a response problem, not a demand problem.

Layer our own HVAC-specific finding on top of that: just 27.2% of the HVAC businesses we crawled make any kind of explicit response-time promise on their site, AI-confirmed, not just a keyword match, the highest rate of the four trades in this batch, but still under a third. Most aren't even claiming to be fast, which means the ones that actually are fast are competing against a low bar.

So what's the real number, honestly?

So instead of a made-up industry average, the honest answer is this: the cost depends entirely on your own lead volume, close rate, and average ticket. Those are numbers only you actually know. The free Lead Loss Calculator below runs the real math using your own figures instead of guessing at someone else's.

To show the actual math rather than just asserting it exists, here's a worked, purely illustrative example, not a claim about any real business: an HVAC company getting 45 web leads a month during a hot stretch, where 50% never get a real reply (a typical small-business range, not a SeenOnMain-measured figure), a 40% close rate on the ones they do reach, and a $500 average job. That's roughly 23 leads a month going unanswered. Even recovering just a third of those, about 8 jobs, at a 40% close rate is roughly 3 additional jobs a month, or around $1,500. The exact numbers change entirely based on your own volume and ticket size. That's the whole reason a calculator using your real inputs beats a borrowed industry average.

Does a bigger ad budget fix this?

One caveat is worth knowing before assuming a bigger ad budget fixes this. In our full report, businesses confirmed to be running paid search ads scored slightly lower on average (32.1) than non-advertisers (35.5) on our Digital Lead Readiness Score. They also made an explicit response-time promise slightly less often (20.8% vs. 24.3%). Paying to generate a lead doesn't reliably come with better machinery to catch it once it arrives. The cost of a slow response applies just as much, if not more, to a paid lead as an organic one.

How heavily is response time weighted in the score?

Response-time signals are worth 20 of the report's 100 points in our scoring methodology. Ten go to making an explicit promise, and ten go to actually having an automated channel behind it, like chat or an AI assistant. HVAC's 27.2% response-promise rate is the highest of the five trades in this specific batch, but still means fewer than three in ten HVAC businesses are even claiming to be fast, let alone proving it.

The report's own recommendation for operators is direct on exactly this: "State a specific response-time commitment if you can meet one. Only 23.1% of sites made any explicit promise. The research this report opens with ties fast callback directly to higher contact and conversion odds. A vague 'contact us' does not do this job." Running your own numbers through the calculator below is the fastest way to see whether that specific fix would be worth the effort for your business. Do this before you spend a dollar changing anything else.