Google Local Services Ads for HVAC: The 2026 Economics

What counts as a charged lead

Google publishes five conditions that make a lead valid and therefore billable:

1. You receive a text message or email from the customer (United States and Canada only).
2. You receive a voicemail, or the customer uses an automated system to meaningfully engage with your business — providing service details, requesting a callback, scheduling an appointment, or similar.
3. You answer a phone call and speak with the customer.
4. You receive a missed call with no voicemail and you return it by text, email or call, either speaking with the customer or leaving a voicemail yourself.
5. You receive a booking request (United States and Canada only).

Condition four is the one that catches people: a missed call alone is not billable, but a missed call you chase becomes billable. Google's How leads work documentation sets all five out, and adds that message leads are "typically priced lower than the corresponding phone lead price," with message pricing driven by the customer's estimated likelihood to book — including whether that customer has already contacted other Local Services advertisers.

Google sets the lead price and does not publish it. The documentation states only that "lead prices may vary depending on your location, the job type, the type of lead, or your bidding mode." No dollar figure for HVAC appears anywhere in Google's Local Services help, so every number you have been quoted came from someone else.

What you control is budget and bid mode. You set an average weekly budget based on the leads you want per week; Google can overspend it in a given week but caps the month at that weekly figure multiplied by the average number of weeks in a month. Once you hit the cap: "After you've reached your monthly max, your ad won't appear for the remainder of the month (unless you adjust your budget)." Billing runs through Google Ads, and any suspension there suspends the Local Services Ads with it.

Bidding offers three published modes: Maximize leads (automated, Google's recommendation, with a suggested minimum budget of 10 leads per week), Target cost per lead, and Max per lead (manual). A bid "is the most you'd pay for a lead" — you may pay less, and Google warns the model takes about two weeks to settle.

The auction that orders the ads weighs bid, likelihood of producing a lead, responsiveness, and profile quality. One clause in that list is worth more than most agency advice: "Higher quality profiles, including those with images, may rank higher and may also pay lower costs per lead." Reviews, response time, images and completed verification checks are not only ranking inputs — Google says they can move your price. On missed calls it is equally direct: "Missed calls may negatively affect your responsiveness."

The August 2026 move into Performance Max

Google is folding Local Services Ads into Google Ads as a Performance Max campaign type with pay-per-lead goals, and HVAC is in the first wave: "August 2026: The first phase of the migration begins for select home and storefront service advertisers in the United States (including plumbing, HVAC, electrical, appliance repair, house cleaning, lawn care, roofing, pest control, and moving)." Service-area businesses without a physical storefront — which describes a large share of HVAC shops — move in late 2026; non-US accounts in 2027.

The pay-per-lead model, the Search-and-Maps-only placement and the keywordless targeting all survive. Four things do not:

• Weekly budgets become daily budgets — your weekly figure divided by 7, with monthly spend capped at that daily figure times 30.4.
• Manual bidding is deprecated: "Manual bidding (such as setting a maximum cost-per-lead) is no longer supported in Google Ads."
• Historical performance reports do not carry over. Export them before your migration date; afterwards the old dashboard redirects and those reports are gone. Past leads, messages and call recordings do transfer.
• Better Business Bureau callouts are gone, replaced by a requirement to select at least six other structured callouts.

Verification carries over intact, badge included, and Google states insurance and licence reverification are no longer required. Administrators get an email 14 days ahead and a reminder 7 days later.

The consumer-facing promise that HVAC marketing content still sells is also gone. Google states it "will be discontinuing the Money Back Guarantee associated with the Google Guarantee badge," with reimbursement available only for services booked through Local Services Ads before December 7, 2025, and only if the request arrives within 30 days of the initial service completion date. If a page tells you the Google Guarantee protects your customers up to a dollar amount, that page is out of date.

HVAC screening is heavier than most trades, and Google says why. Its screening requirements page classifies HVAC as an urgent category "because consumers often call HVAC professionals for time-sensitive, urgent needs, such as a heater that is broken in freezing temperatures," and concludes: "For this reason, HVAC professionals undergo an additional level of screening, which includes service professional background checks." House cleaning and appliance repair, by contrast, list their background checks as "Select users only."

What an HVAC contractor must clear, exactly as Google publishes it:

1. Three background checks — a business check, an owner check, and a service professional check.
2. Two insurance policies — general liability and professional liability.
3. Two licences — a business licence and an owner licence at state level, marked "where applicable by local law."
4. A public, verified Google Business Profile. Google states advertisers "must have a public and verified Google Business profile to advertise with Local Service Ads." Get the Business Profile itself right first — after the Performance Max migration, your business name, address and hours sync one-directionally from that profile into the campaign, and significant changes trigger a 24-to-48-hour review that can pause your ads.
5. Billing and budget set. Google's published pre-badge go-live list also requires acknowledging you hold all appropriate business licences, plus reviews subject to category requirements.

Timeline: Google states that "on average, this process takes 3–4 weeks to complete after submitting your documents," and recommends starting screening and verification "as soon as possible after you've created your Local Services Ads account." Until the badge lands, pre-badge ads can run — but Google places those listings "below providers who've completed all onboarding requirements and obtained the Google Badge".

SearchLight's is the only figure here with a published methodology, a stated month and downstream metrics attached. Its February 2026 benchmark covers 888 home-services contractors, 1,774 campaigns, $6.72M in LSA spend, 126,650 leads and $52.7M in tracked closed revenue; the HVAC row is $51 per lead at a 44.0% book rate, a $2,110 average ticket and a 9.55x closed return — the strongest return of any trade in that dataset. It remains one agency's client book, and February is heating season, which the page itself credits for both the conversion rates and the ticket size. Do not plan a July cooling campaign against it.

The eight-year comparison is instructive once the sample is read correctly. Landers attaches "typically between $25 and $35" to the "dozens" of HVAC accounts his own agency had set up, not to any one city — the sentence before the figure opens "We've set up dozens of Google Local Services ads for our HVAC clients" and ends on how regularly the agency disputed lead validity — and the article's own market list runs to 29 metros, which was the product's entire footprint that month. Against $51 in a now-nationwide product, the direction of travel is clear even though neither figure is a market average.

"By market" is where the evidence runs out. Nobody publishes metro-level LSA lead prices, Google included. SearchLight asserts a national average can "mask $30 CPLs in smaller markets and $90+ CPLs in competitive metros" but publishes no per-market table behind it. The nearest honest proxy is auction pressure on the click side. A 40-keyword Google Ads pull for the United States on 2026-08-17 returned average costs per click from $8.27 on "ac compressor replacement" to $68.74 on "emergency ac repair," and for that one keyword the reported top-of-page bid range alone spans $13.96 to $100.00 [VENDOR — DataForSEO]. Those are auction estimates shown to advertisers, not amounts any contractor was observed to pay, and they price clicks rather than leads. What they establish is dispersion: the spread inside a single keyword exceeds the spread between keywords, which is why a national lead price would tell you nothing about your county.

The arithmetic that matters is your own breakeven, not anyone's average. SearchLight works one publicly: at a $1,800 average ticket and a 25% margin, each job produces $450 of profit; at a 44% book rate and a 43% match rate — match rate being the share of leads attributed to a paying customer in the CRM — roughly 19% of leads become paying customers, so a contractor can afford about $85 per lead before the first job stops paying for itself. Those inputs are SearchLight's, not yours.

Two constraints hurt. Credits land "within 30 days" but "the original lead charge will still appear on your invoice," so your spend reporting and your net cost diverge. And Google removed two credit reasons outright: "Google no longer supports credits for 'job type not serviced' and 'geo not serviced' leads."

That second removal matters because the problem it used to cover was documented eight years ago by an agency running the channel at scale. Ben Landers, President and CEO of Blue Corona, listed it as a platform con in ACCA's HVAC Blog in 2018: "Google isn't very good at telling when people are calling about an area that the client has indicated they do not serve." In 2018 that lead could be disputed. In 2026 it cannot be credited.

Google also publishes ten examples of what will not be credited, and the list is longer than most contractors expect. Nine of the ten reach HVAC; the tenth is law-vertical-only, covering advertisers who opted into general lawyer leads. Leads are not credited when:

1. A valid lead arrives outside your business hours.
2. A customer asked for advice on completing a project related to a service you offer.
3. A customer cancelled a booking.
4. A customer was researching potential projects or prices.
5. A customer did not respond to your return call or message.
6. You listed a general service type but do not perform a specific type within it.
7. You generally serve an area or service but are temporarily unable or unwilling to.
8. You gave a quote and promised a call back.
9. You referred the customer elsewhere.

Automatic credits and booking leads are United States and Canada only, and credits are unavailable in health care verticals, for tax specialists and for advertisers in EMEA. Google concedes the system is imperfect: "you may still notice some poor quality leads processed via auto credits within the first 24 hours."

Your two remaining levers are the Lead Feedback Survey — "Rate this lead" in the Leads section — and your targeting. Google's advice carries its own cost: "If you're receiving unwanted leads, revisit your service areas, job types, and categories." One sentence later — a line telling you to refine the settings to attract your ideal customers — comes the warning: "Remember that any changes you make to these settings could reduce the number of leads you receive." Tightening to stop junk leads also tightens the tap.

SearchLight's own LSA page shows how stale the public record is. It supplies the best HVAC figure on this page, and the mechanism section beside that figure still describes disputing unqualified leads and cites an industry estimate of 6–7% of spend returned as credits — a process Google retired.

On price, the same agency measured LSA and Google Ads weeks apart, which makes it the least confounded comparison available — though the months differ, and that matters in a seasonal trade.

Source: SearchLight's LSA and Google Ads benchmarks, 888 and 816 contractors respectively. The LSA book rate, cost per customer and average ticket in that column are blended home-services figures; HVAC's own row is a 44.0% book rate and a $2,110 ticket. Read the table honestly and LSA wins one comparison and loses another. Against non-branded search it delivers cheaper leads and a higher book rate, at lower average tickets, which SearchLight attributes to LSA's "heavier weighting toward service and repair calls." Against branded search it loses on every metric printed — $34 per lead against $51/$53, a 55.3% book rate against 43.9%, $104 per paying customer against $233 — and a branded query is one that already contains your company name, so that column measures the cost of collecting demand your name has already created. The dollar-by-dollar comparison of search against paid clicks sits on its own page, and fifteen HVAC lead channels ranked by published cost puts LSA next to the rest.

Rented placement is what the benchmark cannot price. Google states that after the monthly max "your ad won't appear for the remainder of the month." Targeting is keywordless, so you cannot choose which searches you appear for. And the auction is explicitly not pay-to-win: ranking "optimizes for the overall health and diversity of the Local Services Ads ecosystem, and provides sufficient opportunity for all providers," which is the mechanism behind the rotation contractors complain about. You rent a position in that rotation, priced by a rate card you cannot see, inside a product that changed its badge, its credit rules and its host platform inside a year.

A defensible sequence for a shop starting from nothing:

1. Verify the Google Business Profile first. It is a hard prerequisite for LSA and it feeds the campaign after migration.
2. Start LSA screening immediately — three background checks and two licence checks take 3–4 weeks on average, and this is the long pole.
3. Answer the phone. Responsiveness sits in Google's published auction factors, and profile quality may lower your cost per lead.
4. Export your LSA reports before your migration date. They do not survive the move.
5. Build the assets in parallel, because steps one to four all stop the day the card declines.

Step five is our work; running Local Services Ads is not. Beyond setup consulting, we do not manage LSA campaigns, and this page is not a pitch for them — it exists because contractors ask where the channel fits before they ask about search. The organic side of the same budget is the question of what you own versus what you rent, and how we replace rented placement with pipeline you keep is the programme that answers it. Nobody, this agency included, can promise a search position or an AI citation; what can be scoped is the audit, the build and the measurement.