There is a moment every HVAC owner on Angi or Thumbtack knows. Your phone buzzes with a new lead notification, you call within ninety seconds, and the homeowner says the same sentence you heard yesterday: “Oh, I’ve already had three companies call me.” You did everything right. You paid for the lead, you answered fast, and you still lost the job to whoever dialed eleven seconds sooner.
That is not bad luck. That is the product working exactly as designed.
I run Sprout Sage Solutions, a marketing agency I started in 2020 in Chandigarh, and home-service companies make up a large share of the 450+ websites my team and I have shipped. I lead every account personally, supported by a team of 17, and the HVAC owners who come to me almost always arrive with the same wound: three or four years of lead-platform invoices and no marketing asset of any kind to show for the money.
The math the lead platforms hope you never run
Start with a single shared lead. A furnace-repair inquiry in a mid-size metro typically sells for $50 to $100 on the big marketplaces. Call it $80. That same inquiry is sold to as many as four contractors at once, which means the platform collects up to $320 for one homeowner’s ten minutes of form-filling.
Now look at it from your side of the table. Four companies are racing to the same phone number, so your realistic close rate on a shared lead sits somewhere between 10% and 20%, and that is if your office answers within minutes. Take the middle: at a 15% close rate, an $80 shared lead costs you about $533 per booked job. Not per lead. Per job you actually win.
Compare that with an exclusive channel. Google’s own Local Services Ads documentation prices HVAC leads in roughly the $25 to $85 range depending on metro and job type, and those leads call you directly, one contractor, no race. Owners I work with close 40% to 50% of answered LSA calls. A $60 exclusive lead at a 45% close rate is $133 per booked job. Same homeowner, same furnace, one quarter of the acquisition cost.
The gap gets wider from there. A lead that arrives through your own website costs whatever you spent building and ranking that site, divided across every call it produces, and that divisor grows every month. The marketplace lead costs $80 every single time, forever, and the price moves in one direction. One owner I talked to last winter had his per-lead price raised three times in two years while his territory got two new competitors added to the same rotation.
Symptoms you can verify tonight
You do not need me for a diagnosis. Pull up your accounts after dinner and check these five things.
Count leads against booked jobs for the last 90 days. Export or screenshot every lead you were charged for, then match them against your job board. Most owners doing this for the first time find a close rate under 20% and have never once calculated their true cost per booked job. Do the division. That number is your real price.
Search your own company name plus your city. If the top results are your Angi, Thumbtack, and Yelp profiles instead of a website you own, the platforms have captured your brand. A homeowner who was referred to you by a neighbor lands on a page listing your competitors below your reviews.
Check what happens when you pause. Think back to any month you turned the lead spigot off, or turned it down. If the phone went quiet within a week, you do not have a marketing engine. You have a subscription to someone else’s.
Read your dispute history. Tally the leads you flagged as bogus: wrong service area, landlord shopping for the cheapest bid, number that never picks up. Then tally how many credits you actually won. The gap between those two numbers is a quiet tax on top of the sticker price.
Look at your speed-to-call routine. If you have trained your dispatcher to drop everything and dial within sixty seconds, you have built an internal process around a defect in someone else’s product. Exclusive channels do not require panic dialing.
Three or more of these ringing true means the rest of this page is about you.
Why the treadmill never slows down
Here is the structural problem, and it has nothing to do with whether the platforms are honest. Their business model requires that you never graduate.
Every dollar you send a lead marketplace buys a perishable good. The lead exists, it is consumed, and the meter resets to zero. Nothing compounds. Your close rate does not improve the platform’s price. Your five-star reputation mostly improves the platform’s page, which ranks for the searches your own site should own. After five years and $150,000 in lead fees, a typical owner holds exactly what he held on day one: an account login.
Meanwhile the platform is spending your money on the thing you should be building. Those fees fund the ads and the search rankings that intercept homeowners before they ever see an actual contractor’s site. You are financing your own middleman.
I audited an HVAC company in the American Midwest this spring that had spent, by its own bookkeeping, about $4,100 a month across two lead platforms. Roughly 460 paid leads a year, 74 booked jobs, and a cost per booked job north of $660 once disputed leads were netted out. The owner’s website had not been touched since 2019 and drew fewer than 100 visits a month. He was not lazy. He was busy, which is precisely the condition the treadmill depends on.
The exit is not negotiating a better per-lead rate. The exit is owning the asset that produces the calls.
What owning your pipeline actually takes
Be clear-eyed about this part, because it is slower than buying leads and anyone who tells you otherwise is selling something.
An owned HVAC pipeline has four working parts. A website built to convert, meaning service pages for every job type and city you serve, load times under three seconds, and a phone number that follows the visitor down the page. A Google Business Profile with a steady stream of fresh reviews, because the map pack decides most emergency searches. Local Services Ads as your paid layer, since they are pay-per-lead, Google-screened, and close to exclusive. And call tracking on all of it, so you know your cost per booked job instead of guessing.
None of that is exotic. What makes it hard is sequencing and persistence: reviews accumulate weekly or not at all, service pages rank after months of consistent work, and the whole system needs someone watching it while you run crews.
Timeline honesty: LSAs can produce calls in the first month. The map pack usually starts moving in two to four months. Organic rankings for “furnace repair + your city” searches are typically a six-to-twelve-month build. The compounding is real, but it is back-loaded, which is exactly why owners quit at month three and crawl back to the marketplaces.
This is where a retainer earns its keep, and I will say this once and plainly. The point of paying an agency $800 or $1,500 a month is not the deliverables list. It is that the money builds equity you keep instead of renting calls you lose. My Starter tier runs $800 a month and Core runs $1,500, with no contracts, and the pricing is published. The Starter tier is performance-gated: if your first 30 days produce zero new leads, month one is free. You pay once the leads start. Run that against $4,100 a month in lead fees that vanish on consumption and the comparison does its own arguing.
If you want a second pair of eyes on your numbers first, book a free consultation or call me directly at +91 97297 12388. Bring your last 90 days of platform invoices. Twenty minutes of division is usually enough.
FAQ
Should I cancel Angi and Thumbtack immediately?
No, and I tell every owner the same thing: do not rip out a working revenue source before its replacement produces. Keep the shared leads flowing while your own pipeline comes online, then throttle them down channel by channel as your cost per booked job on owned channels beats theirs. Most owners can cut marketplace spend meaningfully within six months and to near zero within a year. A cold-turkey cancellation in month one just creates a revenue hole and panic.
Are Local Services Ads just another lead treadmill?
Partly, yes. You are still paying per lead, and if you stop, the calls stop. The differences are real, though: the lead is close to exclusive rather than sold four ways, disputes for bad leads are handled inside Google’s system, and the Google Guaranteed badge lifts your close rate. Treat LSAs as the paid layer of an owned system, not the whole system. Your website and reviews are what eventually shrink your dependence on any pay-per-lead channel.
My website already exists. Why isn’t it producing calls?
Existing and producing are different jobs. Most contractor sites I review are a five-page brochure: one generic services page, no city pages, no tracking, and a contact form nobody answers on weekends. Producing requires a page for each service-plus-city search a homeowner actually types, proof elements like reviews and license numbers near every call button, and a load speed that survives a 4G connection in a parking lot. The audit takes about an hour. The rebuild is usually the first two months of any engagement I run.
Worth reading next
Send me your website. I’ll tell you the 3 things costing you leads.
Not a template report. I open your site, your Google Business Profile and your search results myself, then reply with the three specific things I’d fix first, in order. It is free, it takes me about twenty minutes, and there is no obligation after it. If the fixes are things you can do yourself, I’ll say that too.
I read every one of these myself. No autoresponder sequence, and I don’t pass your details to anyone.
