Every home-service owner has heard both stories. One roofer swears Facebook ads built his whole business. The HVAC guy down the street spent $3,000, got a pile of “leads” that never answered the phone, and swore off it for good. Same platform, opposite results — and the difference isn’t luck. It’s whether the trade fits how Meta actually works.
Here’s the one idea that explains everything below. Google and Local Services Ads capture demand. Meta creates it. Get that right and you’ll know in two minutes whether Facebook and Instagram are worth a dollar of your budget.
The core difference: capturing demand vs. creating it
When someone’s water heater fails, they grab their phone and type “water heater repair near me.” That’s demand capture — the intent already exists, and Google Ads and Local Services Ads put you in front of it. You’re paying to be the answer to a question they’re already asking. Intent is high, and you can charge a premium to win the click.
Meta is the opposite. Nobody opens Instagram thinking “I’d love to see a roofing ad.” Facebook and Instagram ads interrupt — you target homeowners by location and profile, and you show them something that makes them stop scrolling and consider a service they weren’t actively shopping for. That’s demand generation. You’re not catching intent; you’re manufacturing it.
This single distinction decides almost everything about whether Meta pays off for your trade. (The third leg of “capture” — your Google Business Profile and reviews — isn’t paid at all, and for most trades it should be solid before you spend a dollar interrupting anyone.)
Which trades Meta actually fits
Meta works when the service is high-consideration, visual, planned, or seasonal — anything where a homeowner can be nudged from “someday” to “let’s get a quote.”
Strong fits:
- Remodeling, kitchen and bath, roofing, additions — big-ticket, planned purchases where a before-and-after photo does real selling.
- Exterior and recurring services — cleaning, pest control, landscaping, pressure washing, gutters — visual results, repeatable, easy to bundle into an offer.
- Anything seasonal — pre-summer AC tune-ups, fall gutter cleaning, spring landscaping. Use Google Trends to confirm when your demand curve actually rises, then turn Meta on a few weeks ahead to create the itch before the season peaks.
- Retargeting — people who already visited your site or watched your video. This is the highest-ROI Meta play for almost every trade.
Where Meta usually burns cash:
- Pure-emergency, “need it now” work — burst pipes, no heat, lockouts, electrical faults. By the time the problem exists, the homeowner is on Google, not scrolling. Interrupting someone with a “burst pipe?” ad does almost nothing, because the person seeing it doesn’t have a burst pipe right now.
A blunt rule we use: if your customer’s problem has a deadline measured in hours, Google wins. If it’s measured in weeks or seasons, Meta has a shot.
The mechanics, in plain English
Everything below is configured in Meta Business Suite and Ads Manager — the same place you’ll watch CPL, ROAS, and cost per acquisition. First rule of thumb: never run a boosted post. The “Boost” button is a stripped-down ad with almost no targeting, optimization, or tracking control. Build a real campaign in Ads Manager instead — it’s the single most common mistake owners make.
Lead Ads (Instant Forms) vs. landing pages
You have two ways to collect a lead on Meta.
- Lead Ads with Instant Forms open inside Facebook or Instagram and pre-fill the person’s name, email, and phone. Two taps and they’re a lead. This drives the most volume and the lowest cost per lead — but also the lowest intent, because submitting took almost no effort. Meta’s higher-intent form option adds a review/confirmation step that filters out accidental and low-quality submissions; for home services, turn it on.
- Landing pages send the click to your own site. Fewer leads, but warmer ones, because they had to click through and read. Instant Forms typically convert at 8–12% versus 2–4% on a landing page, but a landing-page lead has shown more commitment.
For high-ticket work, the friction of a landing page often filters out tire-kickers. For volume services, Instant Forms usually win — as long as your follow-up is fast.
Creative is the targeting
On Meta, the ad itself does most of the targeting work — especially now that Advantage+ automation increasingly decides who sees it. What pulls in home services:
- Before-and-after photos and short video. Real jobs, real crews, real results. This is non-negotiable for remodeling, roofing, and cleaning.
- A specific offer. “$89 AC tune-up,” “free roof inspection,” “$50 off first cleaning.” Vague brand ads don’t interrupt anyone.
- Faces and local proof. Your team, your trucks, your neighborhood. People hire people.
Because your service radius is small, the same people see your ads quickly. Watch ad frequency — when it climbs past ~2–3 per week, creative fatigue sets in, clicks drop, and CPL rises. Plan to refresh creative every few weeks; local audiences burn out far faster than national ones.
Audience: Advantage+, lookalikes, and retargeting
Modern Meta leans on automation, so the old “stack five interest layers” approach is mostly dead:
- Retargeting (Custom Audiences) — past site visitors, video viewers, lead-form openers, and your customer list. Converts at several times the rate of cold audiences. Always run this first. One catch: website Custom Audiences shrank after Apple’s privacy changes, so engagement-based Custom Audiences (video viewers, Instagram/Facebook engagers, Instant Form openers) are now the more reliable, ATT-resistant way to rebuild a warm pool — and you need steady pixel/CAPI traffic before retargeting has anyone to chase.
- Lookalike audiences — Meta finds new people who resemble your existing customers. Your warmest cold audience.
- Advantage+ Audience + cold geo/homeowner targeting — set your service radius and let Advantage+ Audience find likely buyers, using your homeowner/age/interest inputs as a suggestion rather than a hard fence. Tight radius beats city-wide — but go too tight and your audience falls below Meta’s local reach floor, delivery stalls, and the campaign can’t gather enough data to exit the learning phase. If your service area is only a few zip codes, that constraint, not your creative, is often why Meta underperforms.
The Meta Pixel, Conversions API, and the iOS tracking problem
This is where most accounts quietly leak money. Since Apple’s App Tracking Transparency rolled out, the Meta Pixel — the browser snippet that reports conversions — gets blocked for the large share of users who opt out (well over half of iOS users aren’t trackable). Apple Private Relay and US privacy rules like CCPA/CPRA thin browser signal further. Meta used to cap you at eight prioritized events per domain via Aggregated Event Measurement; as of mid-2025 that limit and the manual prioritization step are gone, and Meta aggregates eligible events automatically — but AEM still reports only a thinned, modeled slice of opted-out conversions, which is exactly why server-side signal matters.
The fix is the Conversions API (CAPI) — server-side tracking that sends conversions straight from your system to Meta, bypassing the browser and ad blockers. Run the Pixel and CAPI together with proper event deduplication, and Meta’s algorithm gets enough signal to actually optimize. Skip CAPI and you’re flying half-blind — which is exactly why so many owners conclude “Meta doesn’t work” when really their conversion tracking was broken. One more measurement note: Meta reports on a 7-day-click / 1-day-view attribution window by default, so some “Meta” conversions are view-through credit, not clicks — read the numbers with that in mind.
The real numbers (2026)
Set expectations correctly or you’ll quit too early — or too late.
- Average home-services CPL on Meta: roughly $30–$35 in 2026, up about 20% year over year.
- By trade: roofing commonly $60–$120 per lead; remodeling $100–$150+ in competitive markets.
- Minimum budget: plan on $1,000+/month to gather usable data. Meta’s algorithm needs around 50 conversions per ad set per week to exit the learning phase, so under-funding it guarantees noisy, expensive results. (The $1,000 floor is an agency rule of thumb, not a Meta rule — but it’s a realistic one.)
- Scale slowly: raise budgets on winners by 20–50% a week. Large edits can re-enter the learning phase; small, steady increases usually don’t.
But the number that decides whether Meta is worth it isn’t cost per lead. It’s cost per booked job (and, framed another way, your ROAS / cost per acquisition).
Meta leads are colder — and go cold fast
A Google lead raised their hand. A Meta lead got interrupted and tapped a pre-filled form. That’s a real quality gap, and it has two consequences.
First, judge Meta on cost per booked job, not cost per lead. A $25 Meta lead that books at 10% costs you $250 per job. A $120 Google lead that books at 40% costs you $300 per job — but those Google jobs might be bigger, faster emergencies. The headline CPL lies. Only the booked-job number tells the truth, and you can only see it if your conversion tracking ties booked jobs back to the channel. The mechanism that closes that loop is offline conversions / CAPI for CRM: when a lead becomes a booked job in your system, you send that event back to Meta so the algorithm learns to find more people who book — not just more people who fill a form. Without it, you’re optimizing toward the wrong outcome.
Second, speed-to-lead is everything. Meta leads decay fast — a lead contacted within five minutes converts dramatically better than one called an hour later, and a lead sitting in Meta’s Lead Center for two days is already dead. If a form fill doesn’t hit your CRM — a field-service platform like ServiceTitan, Jobber, or Housecall Pro — and trigger an instant text or call, you’ll waste the entire budget no matter how good the targeting is. Nurture matters too: most Meta leads need several follow-ups before they book.
This is the single biggest reason Meta “doesn’t work” for owners who try it. It’s not the platform. It’s that nobody called back in time.
The honest verdict
Meta is a complement to Google, not a replacement.
- If you run emergency, high-intent work, Google and LSA should own your budget. Add Meta only for retargeting site visitors and brand presence.
- If you run planned, visual, seasonal, or big-ticket work, Meta can be a genuine growth channel — if you have before-and-after creative, Pixel + CAPI tracking, offline-conversion feedback from your CRM, a real offer, and a five-minute follow-up process.
- For everyone, retargeting your existing site traffic on Meta is close to free money and worth turning on first — once you have enough pixel/CAPI traffic to build the audience.
Don’t ask “do Facebook ads work?” Ask “does my trade create demand or capture it, and is my follow-up fast enough to handle colder leads?” Answer those honestly and you’ll know exactly where to put the next dollar.
See if Meta fits your business
We run Meta ads alongside Google for US home-service businesses, and we’ll tell you straight whether your trade is a fit before you spend. If the math points back to Google Ads or local SEO instead, we’ll say that too. See if Meta fits my business →