A kitchen remodeling company came to us last spring having already burned through about $4,000 on Facebook ads for small business owners like themselves, and they had almost nothing to show for it. Boosted posts. A “get 3 free quotes” lead form. A pile of tire-kickers and exactly one signed job. They were ready to swear off Meta forever and put everything back into Google.
We asked for 90 days and the same monthly budget. By the end of it, that same $2,500 a month was producing 18 to 22 booked estimates a month, and roughly a third of those turned into signed jobs. Nothing magic happened. We just stopped doing the handful of things that quietly waste money on Meta, and started doing the boring stuff that works.
Here’s the whole story, numbers included, because “do Facebook ads work” is the wrong question. The right one is “what does it take to make them work for a business like mine.”
What they were doing wrong (and it’s what most small businesses do)
The remodeler wasn’t lazy or dumb. They were doing exactly what the “Boost Post” button and every Meta email nudge them toward. That’s the trap.
Boosting a post optimizes for engagement, not leads. Meta will happily find you people who like and comment because those are cheap actions. Likes don’t remodel kitchens. When we pulled their old account, their best-performing boosted post had 240 reactions and generated two form fills, both junk.
The second problem was the offer. “Get 3 free quotes” sounds fine until you realize it attracts people who are price-shopping five contractors and aren’t close to buying. A homeowner three weeks from tearing out a kitchen behaves very differently from someone idly wondering what a remodel might cost someday. Same ad, wildly different lead quality.
And the third problem, the quiet one, was no tracking worth the name. They had no idea which ad, which audience, or which creative produced the one job they closed. Flying blind on Meta is expensive, because the platform’s algorithm needs clean conversion data to find more of your good customers. Feed it garbage and it optimizes toward garbage.
First move: we changed what we asked Meta to optimize for
This is the single most important lever, and almost nobody outside the ad-buying world knows it exists.
Instead of boosting for engagement, we built a proper conversion campaign in Meta Ads Manager and told it to optimize for leads, then later for booked estimates once we had enough volume. We set up the Meta pixel correctly on their site (the old install was firing on every page load, which tells Meta nothing) and connected it to an actual “estimate requested” event.
Within about two weeks, cost per lead dropped from roughly $95 to $38. Same spend. The algorithm was finally pointed at the right target. That’s it. No new creative yet, no fancy audiences. Just telling the machine what “success” actually means to a remodeling company.
If you take one thing from this post, take that. Are Facebook ads worth it? They’re worth almost nothing when optimized for likes, and they can be very worth it when optimized for a real business outcome you’ve bothered to track.
The offer did more heavy lifting than the targeting
Everyone obsesses over audience targeting. In our experience running Meta for local service businesses, the offer usually matters more than the audience, at least at the small-business budget level.
We killed “get 3 free quotes.” In its place we ran a specific, lower-commitment offer. A “15-minute in-home design consult, no pressure,” with a real person named and a real timeframe promised. It asked for a smaller yes than “commit to comparing three contractors,” and it screened for people who actually wanted their kitchen looked at.
We also tested a second angle. A gallery ad showing three finished kitchens in their metro with the actual project cost ranges listed. Homeowners are starved for real pricing. Putting a number in the ad did two things. It pre-qualified people (sticker-shocked browsers self-select out) and it built trust, because most remodelers hide price like it’s a state secret.
The gallery ad became the workhorse. It carried most of the account for the rest of the engagement.
Budget: what small-business spend on Facebook actually buys
Let’s talk real money, because “facebook ads cost” answers online are uselessly vague.
At $2,500 a month, after the changes, this account settled around a $40 to $55 cost per booked estimate. That’s not a cost per lead. That’s per person who actually put a time on the calendar. At their average job value and close rate, every signed job was costing them a few hundred dollars in ad spend to acquire against a five-figure project. The math worked comfortably.
Would it have worked at $500 a month? Honestly, probably not well. Meta’s algorithm needs a certain volume of conversion events per week to exit its learning phase and get efficient. Under roughly $1,500 to $2,000 a month for a lead-gen campaign in a competitive local market, you often can’t feed it enough data, and performance stays jumpy. That’s the uncomfortable truth most agencies won’t tell a small business. Below a certain floor, you’re better off putting the money into [INTERNAL LINK: local SEO → /blog/local-seo-for-small-business] and Google’s [INTERNAL LINK: Local Services Ads → /blog/google-local-services-ads] first, then adding Meta once you can fund it properly.
We’re honest about this with clients because a Facebook campaign that can’t get out of the learning phase makes everyone miserable, including us.
Then the conversion rate doubled, and it wasn’t the ad’s fault
Here’s the part people skip. About six weeks in, leads were coming in fine but bookings lagged. The problem wasn’t the ad. It was what happened after the click.
Their old landing page took almost six seconds to load on mobile, and something like 80% of Facebook traffic is on a phone. Every second of load time on mobile bleeds a chunk of visitors before they ever see your form. We rebuilt a single, fast, mobile-first landing page that matched the ad’s promise word for word, put the consult offer above the fold, and cut the form to four fields.
Landing page conversion rate went from about 9% to just over 19%. We didn’t touch the ads that week. Same clicks, more than double the booked estimates. The ad gets people to the door. The page decides whether they come in.
This is why we track the whole chain and tie it back to revenue, not just clicks. If we’d only watched Meta’s dashboard, we’d have seen “leads are fine” and missed that the page was quietly eating half the pipeline.
The cheapest bookings came from people who already knew them
Once the account had steady traffic, we layered in retargeting, and it quietly became the best-performing part of the whole thing.
A kitchen remodel is a considered purchase. Almost nobody sees an ad on a Tuesday and books a $30,000 project that afternoon. They visit the site, look at the gallery, maybe watch a project video, then leave to think about it or ask their spouse. Most of those people never come back on their own. Retargeting is how you show back up.
We built two warm audiences in Meta. One was everyone who’d visited the landing page in the last 30 days. The other was everyone who’d watched at least half of the project walkthrough video. Then we ran a simple ad at them featuring a recent finished job and a soft nudge to book the consult they’d already been thinking about.
Cost per booked estimate on those warm audiences ran under $25, roughly half the cold-traffic number. That makes sense. These weren’t strangers. They were people already halfway convinced who just needed a reason to act this week instead of next month. If you only run cold prospecting and skip retargeting, you’re paying full price to earn interest and then letting it evaporate.
What this means for your account
So, do Facebook ads work for small business? Yes, with real conditions attached, and here’s the short version you can act on.
Stop boosting posts. Build a conversion campaign and optimize for a tracked business event, not engagement. Fix your pixel before you spend another dollar, because bad data poisons everything downstream. Lead with an offer that asks for a small, specific yes and screens out browsers, and don’t be afraid to put pricing in the ad. Fund it properly or wait until you can, because a starved campaign never gets efficient. And treat the landing page as part of the ad, not an afterthought, because on mobile it’s usually where the money leaks out.
The remodeler is still a client. Last time we checked the account, they’d raised the budget on their own, not because we pushed, but because for the first time they could see exactly what each dollar was bringing back.
If you’ve tried Facebook ads before and walked away frustrated, go pull your old campaign and check one thing first: what was it actually optimizing for? Nine times out of ten, that answer explains the disappointment. If you want a second set of eyes on an account before you give up on Meta, we’re glad to look. No pitch.
