Ask a contractor what Angi costs and he will tell you the price of a lead. Somewhere between $80 and $180 for HVAC, depending on the metro and the season, and higher during a heat wave when everybody wants the same thing.
That is not what it costs. That is the entry fee.
The number that actually matters
Angi does not sell you a customer. It sells the same form fill to three to five contractors at once, sometimes as many as eight, and every one of you is billed for it. You are not buying a job. You are buying a place in a race that started before your phone rang.
So the honest measure is not cost per lead. It is cost per booked job, and industry reporting puts shared HVAC leads at roughly $542 per job actually won — against something closer to $190 for a Google Local Services ad, and effectively nothing for a call that came from your own listing.
Here is how it gets there.
Work it through with your own numbers
Take the numbers off your own statement rather than mine. Three lines:
- What you pay per lead. Off last month’s invoice, not the sales call.
- How many of those leads you actually booked. Reported close rates on shared leads run 8–12% — that is roughly one job for every nine or ten leads. If yours is better, use yours.
- Divide. Lead cost ÷ close rate = what one booked job cost you.
At $65 a lead and a 12% close, that is about $540 a job. At $85 and 8%, it is over $1,000. Both are real numbers that real HVAC companies are paying right now.
Do it before you read the rest. The figure is usually two to three times what people expect, because nobody counts the leads they lost.
Why the close rate is so low, and why it is not your fault
You are not bad at sales. You are fourth in a queue.
The lead is blasted to several shops simultaneously, so the homeowner takes the first call and stops answering. Response time decides most of it, which is why Angi’s whole model rewards whoever is sitting by the phone rather than whoever does better work. A contractor on a roof at 2pm loses to one in an office, every time, regardless of who would have done the better job.
Then there is the quality problem — leads outside your service area, leads for work you do not do, leads that filled the form twice. You pay for those too.
The part that decides whether it works at all
Here is the thing that gets missed. Whether $542 is a disaster or a bargain depends entirely on which job you get, and you do not choose.
- A service call is a few hundred dollars. Paying $542 to book a $350 repair means the van left the yard at a loss.
- A system replacement is five figures. Paying $542 to book an $8,000 install is a 7% acquisition cost, which is fine — better than fine.
The lead does not tell you which one it is. So Angi’s economics only work if enough of your leads turn into replacements, and nothing in the model gives you any control over that mix. You are buying a lottery ticket priced as though it always wins.
That is the real question to ask yourself: across last year, what share of your Angi jobs were replacements rather than repairs? If you do not know, that number is the whole answer and it is sitting in your invoicing.
When Angi is genuinely the right call
It would be dishonest to pretend the answer is always no.
If you are brand new, have no website worth the name, no reviews and no ranking, then buying leads may be the only way to get the phone to ring at all. Cash flow now beats a better cost per job in eight months, because there is no eight months if you cannot make payroll in March.
Angi is a reasonable bridge. It is a bad destination. The problem is not that contractors start there — it is that they are still there four years later, paying rising prices to be introduced to customers who were already looking for exactly what they sell.
What the other side of the arithmetic looks like
The alternative is not free and it is not instant. Being straight about both:
A rebuilt site with a page per service and a properly finished Google listing is a one-off cost in the low thousands. Say it produces two calls a month — a deliberately modest assumption. Those callers found you, they are not talking to four of your competitors, and they close far better than a shared lead does. Over a first year that lands somewhere in the low hundreds per booked job, and here is the part that matters: in year two the same asset costs you nothing per job. Angi’s price goes up.
The honest catch is time. Listing work moves in weeks, pages take two to four months to settle. Angi rings today. That is a real trade-off and anyone telling you otherwise is selling something — which is why, if you are currently dependent on paid leads, the sane move is to keep buying them while you build the thing that replaces them, and taper rather than quit.
Where the calls come from instead
Three places, in the order they pay off:
- The map pack. Three listings sit above every organic result on a phone, they come from your Google Business Profile rather than your website, and being fourth means being invisible. How Google picks those three is worth understanding properly, because most profiles are half-finished and that is the cheapest ground you will ever gain.
- A page for each thing you actually do. Not one Services page trying to rank for AC repair, furnace replacement and duct cleaning at once — which is why it ranks for none of them.
- Being the name the assistants give. A growing share of homeowners now ask ChatGPT who to call rather than typing into a search box, and it answers with two or three names. What decides those names is mostly the same work, plus one or two things almost nobody has done yet.
The one-line version
Angi sells you an introduction to a customer who was already searching for your service, at a price that only makes sense if the job turns out to be a big one, and it is shared with your competitors either way.
Getting found directly costs more up front, takes a few months, and then keeps working. Whether that trade is worth making is a maths question, not a marketing one — and you now have the arithmetic to answer it with your own figures.
Work out where you currently stand: the free website scorecard takes two minutes, or see what SEO for HVAC contractors actually involves and what it costs — both on the page, without a phone call.