- Published HVAC acquisition costs describe the herd. None of them can set your ceiling.
- Your ceiling starts with the gross margin dollars a job leaves you, multiplied by the jobs a customer brings you.
- Your acquisition allowance is a share of that number, and it is a decision about payback rather than a benchmark to match.
- Take the bought-lead share and your close rate off the allowance before you name a price per lead.
- Use the published averages to check your answer, never to start it.
Quick answers: What is a good HVAC customer acquisition cost? · Why is my CAC lower than the benchmarks? · How much can I pay for an HVAC lead? · Why do some shops report far higher costs?
Key numbers An all-in acquisition cost of $296 to $350 per new customer across 816 tracked contractors, published by PipelineOn on 5 June 2026. Ad-spend-only self-reports of $80 to $150 per customer from that same contractor population, according to PipelineOn. A $350 per-customer figure still circulating in 2026 that traces to a 2019 Decision Analyst study, as published by JB Warranties in September 2024. Two-thirds of US households use central AC or a central heat pump, reported by the EIA from its 2020 survey of 18,496 households. An average residential central air conditioner life of 18 years, used by the US Department of Energy in federal acquisition guidance updated December 2024.
Every page ranking for this term quotes a cost, then pivots to advice about lowering it. None convert it into the number a lead buyer needs. There is a deeper problem underneath that one. The cost they quote describes what other shops spend, and other shops have never been what sets your ceiling. What a lead is worth to you is a function of the job it might produce, the margin on that job, and how often you win. We ran that same method for solar in Solar Customer Acquisition Cost: Max Payable Per Lead. This is the HVAC version.
Five published HVAC acquisition costs, sorted by what each one counts
Read the population column before the dollar column.
| Figure | Who published it | Population it describes | What it counts | As of |
|---|---|---|---|---|
| $296 to $350 per customer | PipelineOn. It tags this range to a SearchLight 816-contractor sample in its summary, but ties that same sample to the $104 cost per lead in its body. | 816 HVAC and plumbing contractors, $14.9 million in tracked spend | All-in: ad spend plus CSR payroll, agency retainers, software and review incentives | June 2026 |
| $80 to $150 per customer | PipelineOn, describing what owners self-report | Population not stated by the publisher | Ad spend only, overhead excluded | June 2026 |
| $250 to $350 per customer | ZyraTalk, citing JB Warranties | The US residential HVAC industry in aggregate, not any one company | A top-down division of total industry acquisition spend by the pool of potential customers | A 2019 study, republished 2024 |
| Rises of 40%, 31% and 36% year over year | Jon Torrey, The Data-Driven Trades | A blind sample of contractors running paid search | Change in PPC-only cost per acquired customer, January to March | March 2023 |
| $1,500 per estimate call | One contractor on Reddit, as reported by SmartAC | A single four-person shop, self-reported | Cost per replacement estimate call, not per customer | 2026 anecdote |
Sorted this way, the disagreement mostly evaporates. The top two rows come from the same publisher and differ only by definition, but only the first states the population it describes. The rest measure different quantities. There is no average to take here, and taking one is what the whole results page does.
Now notice what none of the five rows contains. Not one of them knows what your jobs are worth. Every figure on that table describes other people’s spending, and spending is an outcome, not a limit.
Ad spend only is not a cheaper CAC, it is a smaller definition
Owners underreport acquisition cost by 40 to 60 percent because they count only ad spend and ignore CSR payroll, agency retainers, software and review incentives, according to PipelineOn. So a shop quoting itself a figure near the low end of the $80 to $150 self-report band published by PipelineOn is not contradicting the $296 all-in benchmark from that same source. It is leaving the office out. Bid off your own ad-spend-only number and you have already spent the overhead without counting it.
Watch the denominator move between a source and its retelling
You can check this one yourself. A blended figure of $104 appears as a cost per lead on Google Ads in PipelineOn’s own write-up. The same $104 appears as a cost per paying customer on branded search in SmartAC’s retelling of it. Open the page a figure came from before you benchmark against it.
What a job is worth sets your ceiling, not what the industry spends
A shop selling replacement systems and a shop selling service calls are not in the same business. An industry average hands both of them the same starting line, which is how a number that describes everybody ends up describing nobody. Work it the other way instead. Take the gross margin dollars a job leaves you, multiply by the jobs a customer brings you over the relationship, and you have what a customer is worth. Your acquisition allowance is a share of that. Choosing the share is a payback decision, and it is yours.
Then count more than one job. A customer who buys a system will not buy another one soon, because the US Department of Energy works from an average residential central air conditioner life of 18 years in federal acquisition guidance updated December 2024. Maintenance and repair fill that gap. They are where a replacement customer pays you a second and a third time, and a shop that counts only the install undervalues its own customer and then underbids for the lead that produced it.
The trade press has been making this argument the whole time. Chris Hunter, principal industry advisor at ServiceTitan, told ACHR News that “a $30 lead that never books is expensive” while “a $150 lead that turns into a replacement, a membership, and five-star reviews is profitable.” In the same report, Reputation Igniter owner Chris Lollini said cost per lead “is often misleading in HVAC” and that cost per booked job “tells the real story.” Both men are valuing the job behind the lead rather than the lead. That is the right scoreboard for running the business and the wrong one for the buy, because booked job is not a denominator anyone sells at. The ladder below is how you carry their point back to the unit on the invoice.
Two haircuts, and the second one can double-count
The first correction always applies, and skipping it leaves your ceiling too high, which is the direction that costs you money. The second one depends on how you measured your close rate. Apply that one at the wrong moment and your ceiling comes out too low, which costs you the lead instead.
The bought-lead share. An acquisition allowance covers every channel and all the overhead behind them. Referrals, organic and your own paid search all sit inside it. Treating the whole figure as a lead budget assumes every dollar you spend acquiring customers goes to a lead seller. No shop works that way, so the overstatement is large and it always runs in the direction that makes you overpay. Multiply by the share of your acquisition budget that actually buys leads, and do it first.
The addressable share, but only if your close rate has not already absorbed it. A raw, general-population lead is not automatically a replacement candidate. Two-thirds of US households use central AC or a central heat pump as their main cooling equipment, reported by the EIA from a 2020 survey of 18,496 households. Now apply this article’s own rule to that figure. Its population is all US households, not your lead pool, so it is a national proxy standing in for a mix only your lead source can report.
Sequencing decides whether that second haircut is real at all. If you measured your close rate on a raw, unscreened pool, the people with no central system are already inside that rate, dragging it down. Cut for them again and you have charged yourself twice for the same problem. Take the haircut only when your close rate came from a screened pool. Then ask your source how it screens on equipment type before you trust any particular number. A pre-screened conversation does not carry the haircut at all, which is most of why one costs more, and we take that apart in HVAC Live Transfer Leads: Pricing the Screening.
From customer value to maximum payable per lead and per call
The rates below are illustrative placeholders. Substitute your own, because yours are the only ones that make the answer true. Two shops walk the same ladder here, one selling replacement systems and one selling service calls, so you can watch the same method land in two different places.
| Step | Replacement shop (illustrative) | Service shop (illustrative) | Where the input comes from |
|---|---|---|---|
| 0. Gross margin dollars per job | A $9,000 system at a 40 percent gross margin leaves $3,600 | A $450 service call at a 55 percent gross margin leaves about $248 | Your own job costing, after equipment, labor, permits and subs |
| 1. Jobs per customer, so what a customer is worth | The install plus six maintenance visits at about $80 margin each gives roughly $4,080 | Three calls over the relationship gives about $744 | Your own retention, against an 18-year equipment life used by the US Department of Energy |
| 2. Acquisition allowance, a share of that | Allowing 10 percent of customer value gives about $408 | Allowing the same 10 percent gives about $74 | Your payback policy, not anyone’s benchmark |
| 3. Apply your bought-lead share | A 40 percent bought-lead share leaves about $163 | The same 40 percent share leaves about $30 | Your own budget split |
| 4. Apply your close rate | A 20 percent lead-to-customer close rate gives about $33 per lead | The same 20 percent close rate gives about $6 per lead | Your own close rate |
| 5. Addressable haircut, only on a screened close rate | At the two-thirds central-AC share reported by the EIA, about $22 | On the same haircut, about $4 | EIA RECS 2020, a national household proxy, not a measure of your pool |
| 6. Price a call instead of a lead | A screened call skips step 5 and converts better, so its ceiling sits above the lead ceiling, not below it | Same direction, from a much smaller base | Your own contact and book rates |
Three things fall out of that ladder.
The two shops finish about five times apart on identical method and identical placeholder rates. Nothing in the published benchmarks would have told you that, because both shops sit inside the same industry average.
The allowance line is the tell. The replacement shop can allow more per customer than the low end of the all-in range published by PipelineOn, while the service shop cannot get near it. One starting line could never have been right for both, and the direction of the error is invisible until you run the margin.
The ceiling also stays sensitive to close rate, so a shop booking 30 percent of what it buys can outbid one booking half that on identical economics. That is the input worth measuring properly before you argue about anybody’s benchmark.
Published per-lead market ranges sit in our HVAC lead pricing guide. Compare them to the ceiling you just built, not to anybody’s average.
What these numbers cannot tell you
They cannot tell you what your own jobs are worth, and margin dollars per job move the answer further than any benchmark here. Your close rate is the next biggest input, and they cannot tell you that either.
They also split into two kinds, and only one is actionable for a lead buyer. All-in and ad-spend-only figures are contractor-operations numbers, describing what a whole business spends including payroll you cannot bid with. The per-lead and per-call figures are the buy-side numbers. The single-shop estimate-call figure reported by SmartAC is neither. Quote it as an anecdote, not a benchmark.
The ladder also assumes every lead you buy is a real, reachable person, and duplicate, aged and bot-filled leads cut the effective close rate that the whole ceiling hangs on.
Last, date every number before you benchmark against it. The oldest figure on page one in 2026 traces to a 2019 study, as published by JB Warranties, and it divided an industry total rather than measuring any single company. Weighing a retainer against buying leads direct is worked through in HVAC Marketing Agency Retainer vs Pay Per Lead, and the sellers are listed in HVAC Lead Generation Companies.
Frequently Asked Questions
What is a good HVAC customer acquisition cost?
There is no single good number, because the published figures count different things. The nearest current all-in benchmark is $296 to $350 per new customer across 816 contractors, published by PipelineOn in June 2026. Compare yourself to it only if you are also counting payroll, retainers and software.
Why is my CAC lower than the benchmarks?
Usually because you are counting ad spend and nothing else. Owners underreport by 40 to 60 percent by excluding CSR payroll, agency retainers, software and review incentives, according to PipelineOn, which is most of the gap between the self-reported band and the all-in range in that same report.
How much can I pay for an HVAC lead?
Start from what a job is worth to you, not from a published average. Take the gross margin dollars a job leaves after equipment, labor, permits and subs, multiply by the jobs a customer brings you over the relationship, and allow a share of that as your acquisition budget. Then multiply by the share of that budget you spend with lead sellers, and by your close rate. Check the answer against the all-in range published by PipelineOn. If the two are far apart, one of them is wrong, and it is worth finding out which.
Why do some shops report far higher costs?
Because they are quoting a different denominator. A widely shared cost per replacement estimate call comes from a single four-person shop posting on Reddit, as reported by SmartAC. That is a cost per estimate call at one business, not a cost per customer across an industry.
Cite this data
Elevarus assembled the sorted comparison above on 13 August 2026 and rebuilt the ladder beneath it the same day. Every underlying figure belongs to the publisher named in its row and links to the page we read it on. Cite it as Elevarus, “HVAC Customer Acquisition Cost: Max Payable Per Lead,” with a link to this page. We publish market ranges only, never our own rate card.
Bring us your margin and your close rate
If you know what a job leaves you after costs and what share of your leads you close, we can tell you whether a lead price pencils at your rates. Bring those two numbers and we will walk the ladder with you.
