Google Ads Cost Per Lead in 2026: HVAC, Roofing, Moving, Cleaning & Garage Door

If you search for cost per lead benchmarks in home services, you will find dozens of articles with confident tables. Follow the citations and something uncomfortable emerges: most of them are quoting each other.

We traced the sources behind the commonly repeated 2026 figures. A large share are aggregator posts citing other aggregators. Several of the most-quoted numbers are modelled rather than measured. One of the better trade-level datasets still in circulation was collected in 2021.

So this article does two things. First, it presents the published benchmarks with their actual provenance attached - sample size and date where the publisher discloses it, and a clear note where they do not. Second, it shows 90 days of real spend from an account we manage, so you can see how far a live account sits from the published average.

Cost Per Lead = Advertising Spend divided by Number of Leads

Spend $2,000, generate 20 leads, and your cost per lead is $100.

The problem is that not every lead is a potential customer. Some come from outside your service area, some want services you do not offer, and some are wrong numbers. That is why the number worth managing is cost per qualified lead - advertising spend divided by leads that meet your own criteria for location, service need, and genuine intent.

In most home service accounts we audit, the gap between total leads and qualified leads is between 20 and 40 percent. Which means a reported cost per lead can understate the real figure by a third or more.

WordStream 2026 Google Ads Benchmarks - provenance: strong. The Home and Home Improvement category runs a 6.47 percent click-through rate, $8.33 cost per click, 8.05 percent conversion rate, and $90.92 cost per lead. Drawn from 13,474 US search campaigns between 1 April 2025 and 31 March 2026, minimum 52 active campaigns per subcategory, reported as medians rather than means. The most methodologically transparent dataset in this group, and also the broadest - the category compresses roofing, HVAC, garage door, and pool into one row.

LOCALiQ Home Services Benchmarks - provenance: real data, but dated. Air Conditioning Services at $8.94 cost per click and $55.15 cost per lead. Heating and Furnaces at $7.92 and $62.18. Roofing and Gutters at $9.03 and $101.49. Plumbing at $8.67 and $48.91. Pools and Spas at $2.94 and $29.08. From 4,595 North American client accounts, medians, minimum 10 active clients per subcategory - but collected between 1 May 2020 and 25 June 2021. Use these for the shape of the relationships between trades, not for current absolute values. Anyone presenting them as 2026 figures is misrepresenting them.

Foundry CRO, April 2026 - provenance: mixed. The only source we found that breaks out garage door and cleaning separately. House Cleaning at $47 blended, $28 non-branded, $22 Local Services Ads, 17.65 percent conversion. Garage Door at $95, $58, $48, 7 to 10 percent. Roofing at $124, $98, $88, 3 to 5 percent. Plumbing at $127, $78, $65. HVAC at $149, $89, $68, 8 to 12 percent. It compiles Searchlight Digital, Invoca, ServiceTitan, and Unbounce, but states no sample size or date range, carries no per-row attribution, and contains an internal inconsistency in its general contractor figures. Treat it as directional.

The branded versus non-branded split is its genuinely useful contribution. Blended cost per lead always flatters an account, because branded search converts cheaply and easily. If someone quotes you a cost per lead without saying whether it includes branded traffic, the number means very little.

MB Advertising, Moving and Storage - provenance: largely modelled. Median $8.50 cost per click, 8.0 percent conversion rate, $82.50 cost per lead, reflecting 2025-2026 auction conditions. The publisher discloses that of 127 US cities in the dataset, only six are directly observed from active campaign data. The remaining 121 are estimates calibrated against WordStream's per-metro distributions. Credit to them for disclosing it, because most publishers do not. But a benchmark that is 95 percent modelled is a projection, not a measurement.

Look at HVAC across these sources. Cost per lead is reported as $55.15 by LOCALiQ, as $104 by Searchlight Digital via secondary reporting, and as $149 blended by Foundry CRO.

Those are not rounding differences. They reflect different definitions of a lead, different mixes of branded and non-branded traffic, different geographies, different years, and in at least one case a large amount of modelling.

This is the honest state of published benchmarks in home services: useful for orientation, unreliable for planning.

Against all of that, here is an account we manage, in full.

Account: Ship Shape Moving, residential moving, San Francisco Bay Area. Window: 14 July 2026 to 11 October 2026. Channel: Google Search only.

Total spend of $26,698.25 produced 716 clicks and 148 conversions. That is a cost per lead of $180.39, an average cost per click of $37.29, a conversion rate of 20.7 percent, and a click-through rate of 3.48 percent.

Set against the published moving benchmark of $82.50 cost per lead at $8.50 cost per click, this account runs more than double the benchmark cost per lead and over four times the benchmark cost per click - while converting at 20.7 percent against a published 8.0 percent.

Neither set of numbers is wrong. They describe different things. The published figure is a national median built mostly from models. This is one real account in one of the most expensive local service auctions in the United States.

A mover in Wichita planning around $180 would badly over-budget. A mover in San Francisco planning around $82.50 would conclude after three weeks that Google Ads is broken.

The lesson is not which number to use. It is that your metro deviates from the national median by more than any published table will tell you.

Splitting the same account by market is where it earns its keep.

Berkeley: $8,185.94 spend, 54 leads, $151.59 cost per lead, $40.52 cost per click, 26.7 percent conversion rate.

San Francisco: $18,002.27 spend, 94 leads, $191.51 cost per lead, $35.51 cost per click, 18.5 percent conversion rate.

Oakland (new campaign): $510.04 spend, 0 leads, $72.86 cost per click.

Read the first two carefully, because the intuitive conclusion is wrong.

Berkeley costs more per click than San Francisco - $40.52 against $35.51. Standard optimisation instinct says move budget toward the cheaper clicks.

That instinct would lose money. Berkeley converts at 26.7 percent against San Francisco's 18.5 percent, producing a cost per lead roughly 21 percent lower. The more expensive clicks are the better buy, by a wide margin.

This is the most common expensive habit in home service accounts: managing to cost per click because it updates daily and feels controllable, when cost per lead is what pays the bills. Cheap traffic that does not convert is the most expensive traffic you can buy.

The mechanism is usually intent concentration. San Francisco produced 15,103 impressions to Berkeley's 5,305 - far more volume, and with it far more noise. A tighter market has fewer mismatched searchers and better geographic alignment with the service area.

No published benchmark would have surfaced this. It only appears when you segment your own account.

The Oakland campaign is the part most case studies delete.

It has spent $510.04, generated 162 impressions and 7 clicks, and produced zero conversions, at a cost per click of $72.86 - roughly double the account average.

Seven clicks is not a result. It is not enough data to conclude anything. But it does show something real: a new campaign in a competitive metro starts expensive and stays expensive while automated bidding gathers conversion data it does not yet have.

Budget a learning window in money and patience. Killing a campaign at $510 and zero conversions is usually premature. Killing it at $5,000 and zero conversions is usually overdue.

Before comparing yourself to any benchmark, work out your own ceiling.

Maximum CPL = Target acquisition cost x Lead-to-customer close rate

Worked example. Average revenue per completed job of $800, gross profit before advertising of $400, maximum acceptable acquisition cost of $200, and a lead-to-customer close rate of 25 percent.

Maximum target CPL = $200 x 0.25 = $50

Two companies in the same city and the same trade can have correct answers that differ by three times, because job mix and close rates differ more between businesses than cost per click differs between markets.

HVAC. Separate emergency repair from system replacement. A repair converts fast; a replacement involves multiple estimates over weeks. Track completed service visits and closed installations separately, because a blended number hides which campaign is working.

Roofing. Measure the full funnel: lead, inspection, estimate, signed contract, completed project. With a 3 to 5 percent published conversion rate - the lowest of any trade above - roofing tolerates an expensive lead only if the close rate and project margin justify it. Storm-driven demand distorts everything seasonally.

Moving. A quote request is not a booked move. Customers routinely get three estimates. Track quote-to-booking conversion, average booked value, and cancellation rate. Local, long-distance, and commercial moves should not share a campaign - their economics are not comparable.

Cleaning. The only trade here where lifetime value dominates the maths. A biweekly recurring customer is worth many multiples of a one-time deep clean, which is why the published cost per lead is the lowest in the group at $47 blended. Separate one-time from recurring campaigns and judge them against different ceilings.

Garage door. Emergency repair and new installation are different businesses sharing a truck. Emergency converts on the phone within minutes; installation is a considered purchase. Separate them, and set a minimum call duration on conversion tracking or your lead count will include people asking about remote controls.

Cleaning companies not yet ready for paid traffic should start with the free channels ranked in our guide to getting cleaning clients.

1. Review the search terms report weekly and add negatives. This is the single highest-return recurring task in any home service account.

2. Match landing pages to intent. A spring repair ad should land on a spring repair page, not a homepage.

3. Set a minimum call duration on call conversions, usually 60 seconds, so hangups stop counting as leads.

4. Import offline conversions. Feed booked jobs back into Google Ads so bidding optimises toward revenue rather than form fills.

5. Segment by geography and cut the areas that produce inquiries you cannot profitably serve. Use drive time, not radius.

6. Separate branded from non-branded reporting, so your real acquisition cost is visible.

7. Fix call answering before increasing budget. More leads into an unanswered phone is the most expensive mistake available.

What is a good cost per lead for Google Ads?

One that lets you acquire customers profitably and consistently. It depends on job value, gross margin, and close rate far more than on your industry. Calculate your maximum profitable cost per lead before comparing yourself to any published figure.

Why is my Google Ads cost per lead increasing?

Common causes are new competitors in the auction, a stale negative keyword list, seasonal demand shifts, declining landing page conversion, or tracking changes that altered what counts as a conversion. Check the search terms report first.

Are published cost per lead benchmarks reliable?

Only as orientation. As this article shows, HVAC cost per lead is reported as $55, $104, and $149 across three sources, one widely cited moving dataset is 95 percent modelled rather than measured, and a commonly quoted trade-level dataset was collected in 2021. Use them to sanity-check, never to budget.

Should I optimise for cost per lead or cost per booked job?

Cost per booked job, wherever you can track it. The Berkeley and San Francisco comparison above shows why - the market with the more expensive clicks delivered the cheaper leads, and only segmented data revealed it.

How much should a home service company spend on Google Ads?

Enough to produce sufficient conversion data for the bidding to work, without exceeding capacity to serve the jobs. In an expensive metro that floor is higher than most owners expect, which is why underfunded campaigns in competitive markets so often look like channel failure.

Does a lower cost per click mean a lower cost per lead?

No, and assuming it does is the most common expensive error in this category. Berkeley in the account above had a higher cost per click than San Francisco and a 21 percent lower cost per lead, because it converted at a far better rate.