Nobody on Facebookwas looking for you.
That is the whole channel in one sentence, and it explains both numbers that matter. Meta leads cost about a third of what search leads cost — $34 average against $90.92 — and they close at about a third of the rate: roughly 12% against 40%. Run it like search and you will conclude Facebook is broken. Run it as demand creation and it does something search cannot.
Sources: AdAmigo.ai 2026 Meta cost per lead benchmarks (home services $34, construction $45 after a 9% rise); Elev8 Operations 2026 home services CPC and CTR data; BuiltRight Digital close rate comparison, cited via 2026 trade analysis; PPCChief 2026 for Google HVAC cost per click. These four numbers only make sense together. Cheap clicks and a weak close rate is not a contradiction, it is the definition of an interruption channel — and it is why cost per lead is the wrong number to judge Meta on.
Why do Facebook leads behave differently from Google leads?
Because of what the person was doing thirty seconds earlier. Someone searching "emergency plumber near me" has a problem and is looking for you. Someone scrolling a feed was looking at a friend's holiday photos. Both can become customers. They do not arrive in the same state, they do not close at the same rate, and treating them identically is the single most common reason contractors write Meta off.
You are creating demand, not capturing it
Search finds people who already decided they need work done. Meta reaches people who have been meaning to sort the fence for two years. That is a genuinely bigger pool and a genuinely colder one, and it needs a different offer and a different follow-up.
The offer matters more than the creative
A specific priced hook consistently outperforms "call us for a free estimate" by several times on cost per lead. On search the keyword does the qualifying; on Meta nothing qualifies the person except what you put in front of them, so the offer is the whole filter.
Your trade is unusually well suited to the format
Before and after photos outperform every other creative type for home services, and you produce them every single week without trying. Most industries advertising on Meta would pay a lot for content that inherently visual sitting on a phone in a van.
Creative expires, and you get billed for it
Meta charges more to keep showing the same thing to the same people. Holding frequency under 2.0 has been associated with cost per thousand impressions falling 15–25%, which is why the same three photos running since spring are quietly costing you a premium.
Small budgets cost more, not less
Campaigns need a certain volume of conversions to exit the learning phase. A budget too small to produce them leaves the campaign optimising badly forever. Underspending on Meta does not buy a cheap test, it buys an unreliable one.
Your customer list is the best targeting you have
One percent lookalike audiences built from past customers have been reported outperforming interest targeting by roughly three times on cost per lead efficiency. The list sitting in your invoicing software is worth more than any targeting option in the platform.
Is Facebook actually cheaper than Google for contractors?
Per lead, yes and by a lot. Per booked job it is much closer, and which one wins depends entirely on your close rate on cold leads. The fourth column is the honest comparison, and it is the one that stops contractors switching their whole budget to Meta on the strength of a cost per click.
Sources: AdAmigo.ai 2026 Meta benchmarks; BuiltRight Digital close rate comparison via 2026 trade analysis; LocaliQ home services benchmark across 3,211 campaigns; SearchLight Digital LSA benchmark, February 2026. Every cost per booked job above except the LSA row is derived by dividing cost per lead by close rate, so treat those three as arithmetic rather than measurement. Close rates in particular vary enormously by trade, offer and how fast you follow up.
Read it honestly and Meta is not the bargain the cost per click suggests — but it is not far off Google either, and it reaches an audience search physically cannot. The reason to run it is not that it is cheaper. It is that the person who has been meaning to replace that fence for two years is never going to type anything into Google. Combining both has been reported to cut cost per lead by 3–29% through audience reinforcement.
Is your Meta problem the auction, the creative, or the form?
Cost per lead is never one problem. It is three numbers multiplied together — what you pay for impressions, how often people click, and how often clickers complete the form. Enter yours from Ads Manager and this tells you which of the three is actually dragging, because the fix for each one is completely different and cutting your budget fixes none of them.
The Meta CPL Diagnostic
AFC · No. 019
Runs entirely in your browser. Nothing is stored or transmitted. The arithmetic is the standard identity — cost per lead equals CPM divided by (1,000 × click-through rate × form conversion rate) — so this is not a model, it is your own numbers rearranged. The benchmarks it compares you against are $48 CPM, 2.1% click-through and a 6.8% form rate, and those three are chosen to be internally consistent rather than individually quotable: a $2.30 home services click at a 2.1% click-through is a $48 CPM, and $2.30 against the published $34 average lead is a 6.8% form rate. Run them back through the identity and you get $34.00. If a set of benchmarks does not reconcile to the published cost per lead, one of the numbers is from a different dataset — which is worth knowing before you judge your account against something you read online. Your trade, market and season still move all three, and a 46% swing between the cheapest and dearest months of the year is normal.
How do you run contractor Facebook ads that produce jobs?
In this order, because the first two decide most of the outcome and cost nothing but attention. Almost every underperforming contractor account we take over is failing on the offer or the creative, not on the targeting settings people spend their time adjusting.
Make the offer specific and priced
"$59 tune-up including a filter" against "call us for a free estimate" is reported to move cost per lead by several times. On search the keyword qualifies the person; on Meta the offer is the only filter you have. A vague offer buys you volume you cannot close.
Shoot before and after, badly
Before-and-after pairs outperform every other creative format for home services, and video beats static with reported cost per lead 30–60% lower. Owner-on-camera vertical phone footage beats polished agency production — the roughness reads as real, which is the entire currency of the feed.
Refresh before frequency passes 2.0
Meta charges more to keep showing the same thing to the same people. Holding frequency under 2.0 has been associated with CPM falling 15–25%. Creative fatigue is cited as a main driver of the 2026 construction cost per lead rise, which means the fix is a calendar reminder rather than a budget increase.
Go tight, and build off your customer list
A 10–15 mile radius beats a 40 mile blanket reliably. Wide audiences look cheaper on CPM and bleed money on clicks that never convert. One percent lookalikes from past customers have been reported outperforming interest targeting by about three times, so upload the list before you touch anything else.
Three qualifying questions, no more
Zip code, project type, timeline. Adding them has been reported lifting close rate from 8% to 12–15%, which is a 50% effective improvement in cost per lead without touching the ad. Add a fourth and completion falls faster than quality rises.
Treat them as colder, and reach them faster
A Meta lead was doing something else a minute ago, so the window is shorter, not longer. Contacting within five minutes materially improves close rates and it matters more here than on search. Everything on the lead generation page applies double.
If you can't fund the floor, this isn't your channel yet.
Meta campaigns need enough conversions to exit the learning phase, and the commonly cited floor is $50–$65 a day. Below that the algorithm never gets the signal it needs, cost per lead goes up rather than down, and you spend three months concluding Facebook doesn't work for your trade. If your total marketing budget is under that floor, put it into Local Services Ads instead and come back to this when there's room. That's the smaller sale for us and it's the right answer.
Get a free Meta auditWhat is included in contractor Facebook ads management?
Offer construction, creative you can actually sustain, tight targeting off your own data, and the refresh discipline that keeps cost per lead from drifting up. Plus the honest reporting that measures Meta on cost per booked job rather than on a flattering cost per click.
Offer & creative strategy
Specific priced hooks rather than free estimates, before-and-after pairs from your own finished work, and a shooting routine simple enough that it survives a busy month. Video where it earns its place, given reported cost per lead 30–60% below static.
See what works DisciplineCreative refresh & frequency control
A 10–14 day rotation with frequency held under 2.0, so you stop paying the fatigue premium that is driving construction cost per lead up 9% year on year. Unglamorous, and the difference between a campaign that improves and one that decays.
Run the diagnostic Your own dataLookalikes & radius targeting
Customer list uploaded, 1% lookalikes built from it, and a 10–15 mile radius rather than a metro blanket. The list in your invoicing software outperforms every interest category Meta offers by a reported factor of three.
See the playbook The long middleRetargeting
Where Meta is unarguably the best tool a contractor has. Someone who read your cost page in March should still be seeing your finished work in July, when the money lands. No other channel accompanies a months-long decision this cheaply.
Contractor SEO Where it is wonLead form & speed to lead
Three qualifying questions, instant alerts, and follow-up built for a colder lead. Qualifying questions alone have been reported lifting close rate from 8% to 12–15%, which beats most changes you could make to the ad itself.
Lead generation systems Alongside, not insteadPaired with paid search
Running Meta and search together has been reported cutting cost per lead by 3–29% through audience reinforcement. Search captures the people already looking; Meta creates the ones who were not. Neither replaces the other.
Google Ads & LSAWhat can a contractor fix on Meta this week for free?
Four things, and the first two will move more than any targeting change. If you do only these and never call us, this page has done its job.
Check your frequency number
It is in Ads Manager under the columns menu. Over 2.0 means you are paying a fatigue premium right now, and the fix is new creative rather than more budget.
Photograph the next job from the same angle twice
Once before, once after. That pair outperforms anything you could buy, and you were standing there anyway with a phone in your pocket.
Upload your customer list
Export from your invoicing software and build a 1% lookalike. It is the single most valuable targeting asset you own and it costs nothing but an export.
Put a real number in your offer
Replace "free estimate" with something specific and priced. It is the largest reported lever on cost per lead in the whole channel, and it is a copy change.
Where can a contractor verify Meta advertising numbers?
Meta's own documentation for how the platform works, and independent benchmark datasets for what things cost. None of these pay us to be listed, and benchmark methodology varies enough that reading two is genuinely better than trusting one.
Meta Business Help Centre
How the learning phase works, what frequency means, and how lead forms are built. The learning phase documentation in particular explains why small budgets underperform.
LocaliQ Facebook Benchmarks
Cost per click and cost per lead across thousands of US advertisers, split by industry. One of the two datasets worth reading side by side.
AdAmigo 2026 Meta CPL Benchmarks
The source of the $34 home services and $45 construction figures on this page, including the year-on-year movement and what is driving it.
2026 Facebook Ads Benchmarks
Construction and home improvement specific click-through and cost per click trends, useful for checking whether your account is drifting or the whole auction is.
SearchLight LSA Benchmark
The comparison point for the table on this page. Worth knowing what the cheapest exclusive channel costs before committing budget to the coldest one.
Eye To Ad Media
Our parent company, running paid social for local businesses since 2012. Listed with the interest declared.
Contractor Facebook ads questions
How much do Facebook ads cost for contractors in 2026?
Do Facebook ads actually work for contractors?
Why is my Facebook cost per lead so high?
What kind of Facebook ads work best for contractors?
How should contractors target Facebook ads?
How often should contractors change their Facebook ad creative?
How much should a contractor spend on Facebook ads per month?
Should contractors run Facebook ads or Google Ads?
More plain-English definitions in the contractor marketing glossary.
We'll tell you which of the three numbers is broken
We look at your CPM, click-through, form conversion and frequency, then tell you whether you have an auction problem, a creative problem or a funnel problem — and what the fix costs. Usually it is the offer, and usually the fix is free.
About a day to turn around. You keep the findings whether you hire us or not. If your budget is below the level where Meta can work properly, we will say so and point you at the channel that fits it instead.
Prefer to talk? 1-800-481-8638 — a person answers. Or email info@eyetoad.com.
Request your free Meta audit
No cost, no contract, no pitch deck.
Reach the ones who were never going to search.
One call. We'll diagnose the three numbers, tell you which is costing you, and say plainly if Meta isn't the right channel for your budget yet.