You put fifty dollars a day behind a lead form, and the leads come in. Forty names in three weeks, under twenty dollars apiece. Two months later you have booked one job, and it was somebody's cousin. The ads did what ads do — they produced leads. Everything that happened after that is where the money went.
This is the most common way paid advertising disappoints a service business, and it almost never gets diagnosed correctly. Ads do not create a sales system. They push volume through the one you already have. If the offer is vague, the service area is wrong, or nobody answers for six hours, advertising does not fix any of that — it just finds out faster and sends you a bill for the discovery.
Here is what to have in place before you spend, why cheap leads are cheap, and how to tell within a month whether it is working.
The cheap lead is cheap for a reason
Facebook leads really are inexpensive next to search, and the gap is wider than most people expect. WordStream's 2026 Facebook Ads benchmarks, drawn from 452 US lead-objective campaigns running between April 2025 and June 2026, put the average cost per lead at $27.39, with 8.54% of people who clicked completing the form.
Now hold that against search. LocalIQ's 2025 home services benchmarks, built from 3,211 US search campaigns between April 2024 and March 2025, put the average home services cost per lead at $90.92 — and roofing and gutters at $228.15.
That spread is not a bargain sitting on the table. It is the price of intent. Somebody typing "furnace not turning on" into Google has a problem right now and is looking for whoever fixes it. Somebody who saw your ad between two videos did not wake up looking for you; the offer was just good enough to tap. Both can become jobs. They do not become jobs the same way, and a Facebook lead handled like a search lead — one callback, a voicemail, a shrug — will convince you the whole channel is a scam.
Five things to fix before you spend
1. Decide what the ad is actually asking for
"Get a free quote" is not an offer, it is a form label. It asks a stranger to start a sales process and tells them nothing about what they are agreeing to. The offers that work name the job, the area, and what happens next: Panel upgrades in Linn County — send two photos of your breaker box and we will get you a ballpark the same day.
Specificity does two jobs at once. It lifts response from the people who have that exact problem, and it quietly turns away everyone else — which is the half most people forget to want. Only promise what you will actually do, though. A same-day ballpark you deliver on Thursday costs more trust than the ad bought attention.
2. Draw the service area before you draw the audience
The default radius around your shop address is a guess, and it is usually wrong in both directions at once. It pulls in towns you would not drive to for a two-hour job, and it misses the one on the far side of the county where you already have six happy customers. Set the geography from the jobs you want and the drive you will still make in February. Exclusions matter as much as inclusions.
3. Qualify on the form — a little
Meta's instant forms are frictionless by design, which is both the point and the problem. The name and number are prefilled, so tapping through takes four seconds — including for people who barely registered what they tapped. Two or three questions fix most of that: the job, the timeframe, and something that shows whether they are in a position to buy, like the property address or whether they own it.
Understand the trade you are making. Every question you add lowers volume and raises cost per lead. That is fine. You are not trying to buy names cheaply, you are trying to buy conversations that go somewhere. Do not turn the form into an application, though — the rest of the qualifying belongs in the first call.
4. Know exactly where the lead lands
Leads sitting unread in Meta's Lead Center is a real and ordinary failure, and it is completely silent. Before you turn on spend, submit the form yourself and follow it: does it reach a phone, an inbox, a CRM? Does anything acknowledge the person automatically? How long before a human could realistically see it?
This is the same problem we wrote about in why leads go cold in the first hour, except advertising makes it expensive instead of merely disappointing. A slow reply to a referral costs you a job. A slow reply to a paid lead costs you a job you already paid for.
5. Pick the number you will judge it by
Cost per lead is the easiest number to improve and the least useful one to improve. You can halve it by next week — ask for less, target more loosely — and book fewer jobs than you do now. The number that matters is cost per booked job, and that means three unglamorous things have to exist before the first ad runs: a source recorded on every lead, a pipeline stage that actually moves, and a reason written down when one is lost.
Without those, in ninety days you will be arguing about the ads from memory. Memory always blames the ads.
When ads are the right move — and when they are not
Paid ads make sense when you have capacity you cannot fill, an offer you can state in one sentence, and somebody who answers. They are a poor first move if you are already booked six weeks out, if you are closing only a small share of the leads you get today, or if the phone genuinely cannot be answered during the workday and nothing automatic covers for it. In that last case the money does more good spent on the intake first. The ads will work better later and cost the same.
One more thing worth being straight about: ads rent attention. The day you stop spending, the leads stop that afternoon. SEO, referrals, and your existing customer list compound instead — slower to start, and they do not switch off. Most healthy businesses run both, and a business that runs only ads is one budget cut away from silence.
What we would do first
If you are starting from zero, do not start with the ad. Start with one offer, one town, a form you have submitted yourself, and a budget small enough that you can leave it alone long enough to learn something — a couple dozen leads, not five. Then judge it on jobs booked, not on the cost per name.
If you are already spending and cannot tell whether it is working, the first place to look is not the campaign. It is what happens in the twenty minutes after somebody hits submit.
Want a second opinion before you spend, or a look at ads that are running now and not adding up? Tell us what you are running and what you are trying to book — call (319) 727-8500, or use the form below and we will take a look.
Sources: WordStream, Facebook Ads Benchmarks 2026 · LocaliQ, 2025 Search Ad Benchmarks for Home Services
