How Much Do HVAC Leads Cost? About Five Times the Sticker Price

An HVAC lead costs about $40 to $150 on the invoice, depending on where you buy it. Meta home-improvement leads run about $41 to $43. Google Local Services Ads run about $51 to $60. Non-branded Google Search runs about $149. Angi and Thumbtack do not publish a set price: third parties estimate Angi at roughly $45 to $100, and Thumbtack pros set their own. That is the sticker. The cost is a different number. Measured all the way to a paying customer, it runs about four to five times the sticker, and that is before you count your office's time or look at replacement jobs alone.

Two prices for every lead

A lead price is what the vendor charges when a homeowner raises a hand. A cost per paying customer is what you spent by the time one of those homeowners paid you. The gap between the two is the whole subject of this article, and it is wider than most owners think.

Here is the number that made me want to write it. SearchLight Digital pulled Google Ads data from 816 HVAC and plumbing contractors for January 2026, about $14.9 million in spend and 143,000 leads. Non-branded search leads, the ones where the homeowner typed "AC replacement near me" and not your company name, cost $149 each. Only 37.6% of them booked an appointment. By the time one of them became a paying customer, the contractor had spent $804 in ad spend. Same lead, more than five times the sticker.

That shape is not unique to Google Search. The same vendor's Local Services Ads data, 409 HVAC accounts in February 2026, shows $51 a lead and about $233 per paying customer across the trades it tracks. Data Driven Trades, a second attribution vendor with its own contractor sample, puts HVAC Local Services Ads at $215 to $276 per paying customer in early 2025. Two vendors, two channels, one pattern: the paying customer costs four to five times the lead.

And those multiples are the low version, for two reasons. First, they count ad spend only. The hours your office spends calling, chasing and rebooking are not in them, and neither are the drives to empty houses. Second, they count every paying customer the same, a repair call and a full system replacement alike. Replacements are the rarer job, so landing a replacement customer costs more than the blended figure. If replacements are the work you want, read these numbers as the floor, not the price.

If you searched "cost per HVAC appointment" and landed here, the answer is the same. An appointment is one stop on the pipeline below. Its cost is the lead price divided by everything that leaks before the booking.

The lead price is the down payment. The cost per paying customer is the price.

The cost pipeline, stage by stage

Every lead that becomes a paying customer passes five stages. At each one, some leads fall out, and the ones that survive carry the cost of the ones that did not. Here are the two Google channels with the best public data, run through the pipeline. The per-lead prices, the book rates and the per-customer figures are measured. The "per booked" lines are my arithmetic from those figures, so read them as the shape, not a quote.

Read the two money columns top to bottom. The price grows at every stage. The invoice never changes.

Stage 1: the lead is billed

This is the stage every vendor is good at, and it is the only stage a lead vendor is paid for. Whatever the record turns into afterward is your problem by design. A lead is a phone number, not a homeowner at the kitchen table.

Stage 2: somebody answers

Invoca's 2026 home-services report, as summarized by PPC Land, says only 34% of inbound HVAC calls are answered by a person. Techs are in the field, the office is on another line, and the homeowner is one tap from the next company. That leak used to be free on Local Services Ads. Since October 1, 2026, Google bills a business-hours missed call as a valid lead when the caller stays on the line more than 20 seconds. An unanswered phone is no longer a lost lead. It is a paid one.

Stage 3: the booking

On Local Services Ads, 44% of HVAC leads book. Divide $51 by 44% and the booked appointment costs about $116. On non-branded Search, 37.6% book, so $149 becomes about $396 per booked appointment. Nothing shady happened. The leads that did not book still had to be paid for, and the ones that did are carrying them.

The hours are the part nobody invoices. Your office manager made those calls on payroll you were already paying. It does not feel like lead cost. It is.

Stage 4: the shown estimate

I could not find a public dataset I trust for the show rate on HVAC in-home estimates, so I will give you ours and label it. On our book, about 20 to 25% of booked estimates cancel after booking. That is my estimate from watching the calendars, not an audit. Under bought leads, that cancellation is yours: the vendor billed at stage 1 and your advisor drove out. Under our model, an eligible no-show is credited, not kept.

Stage 5: the paying customer

One Local Services Ads dataset reports that nearly half of booked appointments became paying customers. Take the $116 booked appointment, divide by 50%, and you get about $232. SearchLight measured $233. My arithmetic and their measurement land on the same number from two directions, which is why I trust the shape of this pipeline even where a single stage is fuzzy.

Every stage you own is a stage where the lead price grows and the invoice does not.

When cheap leads are the right buy

I sell appointments, so you know my bias. Here is the honest part anyway. Bought leads can be the right call, especially for repair and service work, where the blended per-customer numbers above fit the job you actually sell. If your shop fits one of these, buying leads is a good decision:

• You have a staffed desk that answers in minutes, including the 4 PM July rush. The 34% answer rate is an average, and you are not average.
• Your work is mostly repair and service, where the ticket is small and speed matters more than selling skill.
• You are new, with few reviews, and you need volume to build a reputation before you can be picky.
• You want to set your own price per lead and pause whenever you like. Pay-per-call vendors like eLocal and Service Direct let you set the price and sign no term contract.

Marketplaces are a different animal. Angi's own help pages say a homeowner's request goes to multiple pros; Angi does not publish how many, and it does not publish an HVAC lead price. The $45 to $100 figures you see online come from agencies, not from Angi. What you are buying there is a head start that several other shops got at the same moment. For a shop that can out-dial its neighbors, that head start can be cheap. For a shop where the office manager is also the dispatcher, it usually is not.

Cheap leads are a good buy for the shop that can answer the phone. They are an expensive buy for everyone else.

What a shown appointment costs

Now the unit we sell, so you can compare it fairly. Under our pay-per-appointment model, you pay one flat fee per shown, qualified appointment, with the ad spend included. Our setters own stages 2 and 3: the calling, the qualifying, the booking onto your calendar. An eligible no-show is credited, not kept. The price per unit is higher than a lead price because the unit sits at stage 4 of the pipeline, not stage 1. Compare it there.

The stage we do not own is stage 5. Our clients close about 38.5% of shown appointments on average, contractor-reported, with a range of roughly 35 to 50%. I hold that number loosely. Closing is your team's work, not ours, and it moves with your market and with how your price sits against the other companies the homeowner is talking to.

So here is the comparison I would run, as an example with your own inputs. Take whatever per-appointment price you are quoted, from us or anyone else. Divide it by your close rate on shown estimates. At 38.5%, divide by 0.385. That is your cost per sold job under appointments. Then take last month's lead bill, add the office hours spent dialing, and divide by jobs sold. That is your cost per sold job under leads. Put both next to the ticket. ServiceTitan's 2025 data puts the average HVAC installation invoice at $17,040 nationally. The Real Math piece runs this comparison end to end on the same ten estimates if you want the full worksheet.

Compare a shown-appointment price to your cost per paying customer, never to a lead price.

Run this on your own numbers

Seven lines. Use last month, or the last 90 days if your volume is small.

1. Lead bill and lead count. Divide the bill by the count. That is your sticker price, the one every vendor quotes.
2. Leads you actually reached. Divide the bill by that number. The price just went up, and nothing was billed.
3. Appointments booked. Divide the bill by booked. This is your real cost per HVAC appointment, before anyone drives anywhere.
4. Appointments shown. Divide the bill by shown. Every empty driveway is in this number.
5. Jobs sold. Divide the bill by sold. This is your cost per paying customer, the number the attribution vendors measure.
6. Office hours on the phone. Count the hours your people spent dialing and chasing. Price them at what you pay, add it to the bill, and run line 5 again.
7. Any per-appointment quote, divided by your close rate. Set it beside line 6.

The public datasets land at four to five times the sticker price by line 5. If your number is under that, your desk is beating the market, and leads suit you. If it is over, the leak is in stages 2 through 4, and the real question is whether you want to fix the desk or pay someone to own those stages for you. Either answer is fine. Buying more leads without knowing which one you are is the only wrong move.

If you run the seven lines and want a second set of eyes, book a 15-minute growth call and bring the sheet. We will mark up where your money leaves the pipeline. If the answer is "keep buying leads and hire a second person for the phone," I will tell you that.