Guide for HVAC owners
What HVAC leadsactually cost
Every benchmark you will find quotes one number for a market that has at least six. Here are the ranges we see by channel, split repair against replacement, and the reason the table below is not the number you should be managing to.
Across the HVAC accounts we run in Southern California, repair leads typically cost between $35 and $120 and replacement leads between $60 and $260, depending on channel and market.
The spread within those ranges is driven more by job type, market and season than by which channel you buy, which is why a single benchmark figure is close to useless for planning.
Contractors ask what a lead should cost because they want to know whether they are being taken advantage of. It is a reasonable question with an unhelpful answer: the same lead is a bargain in one business and a disaster in another, depending on what happens after it arrives.
So this page gives you the table, then explains why the number underneath it, cost per acquired customer, is the one that decides whether your marketing works.
Cost per lead by channel
These are ranges from accounts we manage, not industry benchmark reports. Replacement leads cost more everywhere, on every channel, without exception.
Typical cost per lead, by channel and job type
-
Google Search
Repair$55 to 120Replacement$110 to 260 -
Local Services Ads
Repair$45 to 95Replacement$80 to 180 -
Paid social and video
Repair$70 to 160Replacement$60 to 140 -
Shared lead marketplaces
Repair$35 to 90Replacement$70 to 190 -
Organic and Google Business Profile
Repair$10 to 35Replacement$15 to 50 -
Your own database
Repair$5 to 20Replacement$8 to 30
Ranges from accounts under management, refreshed twice a year. The two amber rows are channels you own rather than rent, which is why they are an order of magnitude cheaper and also why they are capped by the size of your list.
Three things in that table are worth more attention than the headline numbers.
Marketplace leads look cheap and are not. A shared lead is sold to several contractors at once, so the price has to be read against a contact and close rate that is a fraction of an exclusive lead's. Cheap per lead, expensive per customer.
Paid social is the only row where replacement is cheaper than repair. Nobody browses social with a broken system, so it serves urgent repair badly, but it reaches the homeowner with an ageing system months before they search. That is the cheapest replacement demand most contractors can buy.
Owned channels are ten times cheaper and strictly limited. Your database will not produce more leads than it contains, which makes it the first thing to work and the last thing to rely on.
The number that actually decides it
Cost per lead is an input. Cost per acquired customer is the outcome, and the distance between them is made of two things you control: how many leads you reach, and how many you close.
What 100 paid leads become
-
100 leads bought$9,000 spent
-
68 reached$132 each
-
41 quoted or booked$220 each
-
12 closed$750 each
A $90 lead becomes a $750 customer. Every contractor's version of this chart looks different, and almost none of them have drawn it.
That chart is why two contractors buying identical leads at identical prices can have completely different businesses. The one who answers the phone and follows up is not paying less per lead. They are paying less per customer, which is the only figure the bank cares about.
Answer rate is a cost lever, not a service issue
The first step in the chain is the cheapest to fix and the most commonly ignored. If you answer a share of the calls you have already paid for, the ones you missed did not cost you nothing. They cost full price and produced no revenue, so every answered call is subsidising them.
Run your own volume through the missed-call calculator and the number that comes out is not a customer service metric. It is your real cost per lead, corrected for the ones that never got picked up.
Your ceiling is not the benchmark
None of the ranges above tell you what you can afford. That comes from your ticket, close rate and gross margin, and it differs between repair and replacement by enough that blending them misleads you in both directions.
Work it backwards with the max cost per lead calculator, twice: once on a repair invoice and once on a replacement. If your maximum sits above the ranges in the table, you have room to bid harder than your competitors. If it sits below, more traffic will not save you and the problem is upstream in pricing, close rate or job mix.
If maintenance plans are part of your mix, run the calculation a third time on plan lifetime value rather than first-job margin. The shoulder season guide has the plan economics, and the number it produces usually justifies bidding you had written off.
Why your market changes the answer
Geography moves these numbers more than most contractors expect, and not always in the direction that seems obvious.
- The Inland Empire has the cheapest clicks in the state and some of the thinnest tickets, which means a lead price that looks excellent can still be unaffordable against a run of low-value work.
- The Coachella Valley has the best ratio in the build: cheap traffic against high tickets, because the market is small enough that larger agencies have never bothered to compete there.
- West Los Angeles is the opposite, with head-term clicks several times the state average and an entrenched field, where the only affordable route is a narrower job type nobody else is bidding on.
- The Bay Area carries the highest costs in the country, and the only sensibly priced traffic there is compliance-driven rather than climate-driven.
- In the San Gabriel Valley the gap between repair and replacement economics is at its widest, which makes running them as one campaign the most expensive mistake available.
- In Southeast LA County the arithmetic only works on plan lifetime value, because income caps the replacement ticket and repair alone will not carry an acquisition cost.
And why July is not the month to judge it
Cost per lead moves with the season as competitors ramp and pull back. We typically see peak-month costs running meaningfully above trough-month costs on the same terms in the same market. Judging your account on a July report compares your worst prices to nothing, and switching spend off in October surrenders the conversion data that made July work.
Common questions
What is a good cost per lead for HVAC?
There is no single figure, because the same price is good or bad depending on job type, market and what you do with the lead after it arrives. A replacement lead at $180 is excellent. A repair lead at $180 usually is not.
The useful version of the question is whether your cost per acquired customer sits comfortably below the gross profit on the jobs you win. If it does, your lead price is fine regardless of what a benchmark says.
Why do replacement leads cost so much more than repair leads?
Because everyone wants them and there are fewer of them. Replacement intent is a small fraction of search volume and every contractor in the market is bidding for it, while repair volume is larger and more evenly spread.
The premium is usually worth paying. A replacement job can be worth twenty or thirty times a diagnostic call, so a lead costing twice as much is still the better purchase provided your tracking can tell the two apart. Many accounts cannot, which is why they optimise toward the cheap one without meaning to.
Are shared leads from marketplaces worth buying?
Sometimes, as a supplement, and almost never as a foundation. The price per lead is genuinely lower, but the lead is sold to several contractors simultaneously, so contact and close rates fall hard and cost per acquired customer often ends up higher than exclusive channels.
If you use them, measure them on customers won rather than leads bought, and be honest about what speed of response they demand. The contractor who calls in ninety seconds wins those leads, and everyone else pays for them.
How much should an HVAC company spend per month?
Work up from capacity rather than down from a percentage of revenue. Decide how many additional jobs a month you can actually deliver, multiply by your cost per acquired customer, and that is the budget. Anything above it buys leads you cannot serve well, which damages the reviews that make the next month cheaper.
There is a practical floor. Below roughly the value of two system replacements a month in media, most accounts gather conversion data too slowly for automated bidding to work properly.
My cost per lead jumped in summer. Is something broken?
Probably not. Peak season brings every competitor into the same auction at once, so click prices rise and lead prices follow. It is normal and it is why the peak should be judged on revenue and cost per customer rather than on cost per lead.
What is worth checking is whether your close rate fell at the same time, because that usually means volume outran your ability to answer and quote, and that is fixable.
What should cost per acquired customer be?
Low enough that the gross profit on the job comfortably covers it with margin left, and low enough that the arithmetic still works on your worst month rather than your best.
For replacement work most healthy accounts sit at a small fraction of job value. For repair work the honest answer is that a single repair frequently cannot carry a paid acquisition cost at all, which is the argument for maintenance plans rather than an argument against advertising.
Next step
Let's find your
real number
Thirty minutes. We pull up your accounts, work out what you are actually paying per acquired customer rather than per lead, and show you where the gap between the two is going. Not ready for a call? Take the marketing scorecard first, twelve questions and no email required, or read how we run the trade on the HVAC marketing hub.
★★★★★ 4.9 across 312 reviews No obligation · Usually booked 1-2 days out