22 minute read
Home service owners have been burned by ads. A freelancer or an agency launched a campaign. Leads came in for three weeks. Then cost per lead doubled, the inbox filled with tire-kickers, and someone declared "Facebook doesn't work for us."
Sometimes the targeting was wrong. More often the structure was wrong. One ad, one audience, one landing page, and a CSR who called back tomorrow.
Paid media for HVAC, plumbing, roofing, and the rest of the trades is not a creative contest. It is a production system. Creative dies. Markets saturate. Offers age. The companies that keep buying jobs are the ones who planned for that from week one.
The one-ad problem
A single concept — even a strong one — has a half-life. The people who were going to click, click. Frequency climbs. Comments get weirder. The algorithm has fewer new people who look like yesterday's converters. You raise budget to hold volume and you train the platform to hunt cheaper, worse traffic.
The fix is not "make a new image every Friday" with no system. The fix is a matrix: the same core offer expressed across where they live, what they need, and what they respond to.
A 3-ton changeout in a zip code you actually service is a different ad than an emergency no-cool at 9 p.m. A roof replacement after a hail week is a different ad than a maintenance membership in February. Age and home-ownership signals matter. So does the promise: book an inspection, book an estimate, talk to a human tonight.
We do not copy anyone's trademarked framework. We do insist on spread. Ten thin angles you can measure beat one hero creative you are emotionally attached to.
Prove the offer before you scale the spend
We launch with a test budget across a small set of proven angles for your trade and your service area. Keep the winners. Cut the losers early. Scale is a privilege the numbers earn.
That requires honesty about geography. If you do not service the west side, do not let the algorithm "find more customers" there. Home services are not national ecommerce. Every click outside the service polygon is waste plus a bad review waiting to happen when someone still calls.
It also requires honesty about capacity. Ads that book more estimates than your sales team can run will trash show rate and Google reviews. We would rather throttle than light your reputation on fire for a record lead month.
The click is not the job
The second failure mode is the funnel. The ad worked. The homeowner submitted a form. Then nothing for four hours. They called the next company on the map pack.
So paid ads in our world include:
- Landing pages that match the offer in the ad — not a homepage with a generic "contact us."
- CRM capture with source and fields your dispatchers actually use.
- Nurture the same day: SMS, email, optional voice, missed-call textback.
- Creative production on a cadence, not a panic when CPA spikes.
Campaign approval still lives in Ads. You see the targeting, the spend, and the creative before anything goes live. We do not "just turn it on."
Website lead optimization sits here too. If organic and LSA traffic hits a form that leaks, paid is not the only leak — it is just the leak you can see on a credit-card statement.
What "good" looks like without fake dashboards
We will not put someone else's $89,000 month on this page and imply it is yours. Booked revenue depends on ticket size, close rate, weather, and whether your estimators can sell.
What we will define up front:
- Cost per qualified lead (not cost per form fill from a renter outside the county).
- Speed-to-first-response.
- Estimate set rate and show rate.
- Close rate and revenue from the campaigns — in your CRM, not a screenshot from an ad account.
If those numbers are not in the CRM today, the first work is instrumentation. Otherwise you are optimizing a fiction.
Creative is a factory, not a muse
Home service creative that lasts is specific: the real truck, the real problem, the real neighborhood, the real constraint ("same-week slots," "financing on qualifying systems," "we service these ZIPs"). Stock smiles and floating logos die faster.
We produce variants against the offer and the landing page, then iterate. Video and static both have a job. UGC-style can work; so can a straight talk from the owner if they can say the sentence in one breath. We do not need a cinematic brand film to book a water-heater swap.
Fatigue is expected. The calendar should already have the next set in motion before the current set dies.
Compliance and claims
Trades have rules. Licensing, financing disclaimers, "up to" language, emergency claims, insurance work — we will not let an ad promise what your techs cannot legally or operationally do. That is not us being precious. That is how you avoid a bad month with the platform and the state board.
How ads sit with the other systems
If the phone is not answered, ads make the miss more expensive. If reviews are thin, ads lose the comparison. If the CRM is full of dead estimates, you are buying new people while ignoring paid-for ones. Coaching matters because a lead that reaches a CSR who cannot set an estimate is still a wasted click.
The order we prefer: stop the leaks (phone, follow-up, list), make the profile look like a company people can hire, then scale the matrix. You can start ads earlier if you accept that part of the budget is tuition. We will say so.
Signal still helps when you do not know which offer to run. Forge is how the landing, CRM, and agent pieces get deployed with evidence gates. Run is how the account does not rot after the first winning month.
The landing page for all six systems is home services. To walk your last 90 days of spend, book a discovery call.