Facebook Ads for Contractors: The Operator Playbook

In roofing, windows, and HVAC, the same pattern shows up in ad accounts that struggle. The campaigns are often structurally fine. Targeting looks reasonable. Budgets are real money. The problem is everything that happens after someone clicks. There is often no speed-to-lead, no follow-up sequence, and no appointment confirmation logic. The ads generate form fills that die in a CRM nobody touches until Monday morning. Facebook ads for contractors are not a creative problem. They are a system problem, and most contractors are solving the wrong end of it.

Why Facebook Ads Work Differently for Contractors

Paid search captures demand that already exists. A homeowner types "roof replacement near me" and Google shows your ad. Facebook is different. You are interrupting someone who was not looking for a roofer thirty seconds ago. That gap changes everything about how the campaign has to be built. Meta targeting runs on demographics, life events, and behavioral signals, not search intent. A homeowner in a coastal area who has engaged with home-improvement content is a reasonable proxy for someone who needs impact windows. That is the logic you are building into your audience. The creative has to manufacture the reason to act. A search ad can be two lines of text because the query created context. A Facebook ad has to create that context from scratch: a problem the homeowner recognizes, a reason your company is the answer, and a next step that feels low-friction. That is a harder brief. Most contractors hand it to a freelancer with no knowledge of the trade and wonder why the leads are soft. For a full comparison of how paid social stacks up against paid search for home services, read our guide to Google Ads for Contractors.

The Audience Structure That Actually Produces Appointments

Most contractors run one cold audience and call it a campaign. That is not a campaign. That is a single bet with no fallback. A structure that works layers three audience types. The first is warm retargeting: website visitors, video viewers, and anyone who engaged with the page in the last thirty days. These people already know the brand. The cost to convert them is lower and the lead quality is higher. This audience is small, but it should never be off. The second layer is lookalikes built from your best customers, closed jobs, not form fills. Upload a list of homeowners who spent above your target job size and Meta finds people who look like them across the platform. The quality gap between a lookalike built from closed jobs and one built from raw leads is real. The third layer is cold interest-and-demographic targeting. This is where most of the spend lives, and it is the hardest to make profitable. Location and a few home-improvement interests are the main levers, and if Meta puts your ad in the Housing special ad category, which can happen with home repair and improvement ads, age, gender, zip-code and lookalike targeting are off the table. Under Meta's default Advantage+ audience, interests and lookalikes act as suggestions Meta can go beyond, so stacking interest filters adds little and mostly narrows the signal you start with. Broader audiences with tighter creative usually outperform narrow audiences with generic creative.

Creative That Books Jobs, Not Just Clicks

The creative brief for a contractor Facebook ad has three parts: a problem the homeowner already recognizes, visual proof that you solve it, and one call to action. Video beats static for home services because the product is physical and the transformation is visible. A roof before and after, a window installation in sixty seconds, a homeowner explaining why they finally called, those convert better than a logo on a blue background. Creator-style video, shot on a phone and edited to feel native to the feed, routinely beats polished production. We offer creator video by Marissa Snyder, who has 5.79M total views and 4.4M on her biggest single video, because the format that looks casual is often the one that actually stops the scroll. The FTC Endorsement Guides require that any paid or incentivized testimonial be disclosed. If you run UGC that involved compensation, the disclosure has to live inside the content itself. Burying it in a caption does not count. This is not optional and the penalties are real. Test one variable at a time. Change the hook, measure it against the control, then move to the next element. Running five completely different ads at once can find a winner, since Meta's delivery favors creative variety, but it will not tell you which element made the difference.

• A CRM with instant lead routing Every form fill needs to land in a pipeline immediately, trigger a notification to the person making contact, and log the time stamp. If you cannot see how fast your team responds to leads, you cannot improve it.
• An automated follow-up sequence The sequence sends a text, an email, and a callback within the first few minutes after a form fill, once the form has captured the homeowner's consent to automated texts. The sequence runs whether the lead comes in at 9 AM or 9 PM. Manual follow-up at home-services lead volume is not a system.
• A dedicated landing page, not your homepage The homepage has too many exits. A landing page with one offer, one form, and no navigation menu removes the friction between the ad click and the conversion event.
• Conversion tracking on the thank-you page Meta needs a conversion signal to optimize. A pageview on the thank-you URL after a form submission is the minimum. Without it, the algorithm is flying without instruments.
• A clear offer the ad can lead with A free inspection, a free estimate, a same-day assessment, something that makes the next step obvious and low-risk for the homeowner. 'Learn more' is not an offer.

Budget, Bidding, and How Long to Wait Before Judging a Campaign

Meta's ad algorithm has a learning phase. During that phase, performance swings and the data doesn't reflect what the campaign will do at scale. Cut the budget or make a major change and the learning phase resets. How long that takes depends on daily spend and how many conversions you're generating, because Meta measures the learning phase in optimization events, not weeks. The budget question every contractor asks is: how much is enough? The real answer is that your budget has to be large enough to produce enough conversion events for the algorithm to learn from. Meta's own documentation says an ad set needs roughly fifty optimization events per week to exit the learning phase. At higher ticket sizes, roofing, impact windows, full remodels, the cost per conversion event goes up. That means you need more budget to generate enough events in a reasonable window. Bid strategy matters. Highest volume bidding (the default, formerly called lowest cost) tells Meta to spend your budget and get as many conversions as it can. A cost per result goal (formerly cost cap) lets you set the average you want to pay per conversion, which is useful once you know your numbers. Set the goal too low and it throttles delivery. Start with highest volume while the campaign is learning. Move to a cost per result goal once you have baseline data. For lead-generation campaigns, instant forms, Meta's native lead forms, typically produce higher volume at a lower cost per lead than click-to-website campaigns. The trade-off is lead quality. Native forms pre-fill from the user's profile, which cuts friction and sometimes cuts intent along with it.

The Lead-Quality Problem and How to Diagnose It

Bad lead quality is the most common complaint contractors have about Facebook ads. The leads come in. Nobody answers the phone. They give fake addresses or say they were just browsing. That complaint is almost always a targeting or offer problem, not a platform problem. Start with the audience. Cold interest-and-demographic targeting on Meta is loose. An income or interest filter does not mean the person owns a home, or that the home needs your service at all. Layer geo-targeting with areas that have the right housing age, income distribution, and weather exposure for your product, using zip codes where Meta allows them, since Housing-category ads cannot target by zip. Tighter inputs produce tighter audiences. Then check the offer. A "free quote" ad attracts price shoppers. An ad that leads with a specific problem, for example one aimed at Florida homeowners with older tile roofs worried about their next wind inspection, qualifies the reader before they click. The qualifier is in the creative. Anyone without an older tile roof removes themselves before they fill out the form. Then check the form. Longer forms with qualifying questions, homeowner or renter, type of roof, own or rent, reduce volume and raise quality. The right trade-off depends on your close rate and ticket size. At high ticket sizes, a smaller number of better-qualified leads almost always produces more revenue per dollar of ad spend than a large volume of low-intent contacts. The full math on that is at /roofing-leads.

Facebook Ads vs. Google Ads: Which One to Run First

Every contractor with a budget and a small team asks this question. The honest answer depends on where your constraint is. If your constraint is demand, people not finding you at all, Google is the faster path. Search ads capture homeowners who are already looking. Google's Local Services Ads put roofing and home-services companies at the top of local searches, and the "Google Verified" badge they carry, which shows only after Google's verification clears, cuts the friction to call. If your constraint is volume at a specific ticket size and your follow-up system is already working, Facebook is the right next layer. It reaches people who would never have searched for you. That is a different kind of prospect, but a real one. The contractors who fail at Facebook ads run them before the back end is ready. No CRM, no automated follow-up, no dedicated landing page. The platform is not the problem. The back end is the problem. For a full breakdown of how we think about paid search versus paid social for home-services, read our Google Ads for Contractors guide.

Facebook Ads vs. Google Ads for Home-Services Contractors

The Follow-Up Stack That Turns Leads Into Booked Jobs

The ad gets the click. The follow-up system gets the job. These two things are built by different people, optimized by different metrics, and fail in different ways, which is why contractors who treat them as one problem stay confused about why their ads 'don't work.'

The stack has three layers. The first is routing: the lead hits the CRM the moment the form submits, a task is created for the setter, and a text notification goes to whoever is making the first call. The CRM should log the timestamp of every touch so you can measure speed-to-lead objectively.

The second layer is automated nurture: a text that arrives within minutes acknowledging the inquiry, followed by an email with the company's credibility markers (license number, years in business, photo of the truck). The automation runs on every lead regardless of when it came in.

The third layer is live outreach. A setter calls from a local number, not an 800 number, not a VOIP number with a suspicious area code. The goal on the first call is not to close. It is to confirm the appointment. Every call should be recorded, with the recording disclosure your state requires, so QA can identify where leads fall off.

This is the same infrastructure we build for contractors through our call center program: dialer, CRM, calendar, and payroll on one stack, with QA on every appointment before it counts.

One Hot Take: Stop Blaming the Algorithm

Here is the opinion most agencies will not say out loud: Facebook ads for contractors work fine. The algorithm is not broken. The targeting is not the core problem in most accounts. The problem is that contractors have been told the media buy is the product, when the media buy is actually the smallest part of the system.

Plenty of contractors have campaigns that are genuinely well-built, right audiences, good creative, reasonable budget, and still do not make money from them. The constraint is usually downstream: a setter who works a nine-to-five schedule on leads that come in at seven PM, a CRM with no automation, or a landing page that loads slowly on mobile.

The ad platforms are not going to tell you this. They want you to believe that one more creative test or one more audience segment will unlock the results. Sometimes that is true. More often, the unlock is in the process that runs after the click.

That is also why SEO builds up in a way that paid ads cannot. Our own site went from 600 to 1,300 organic clicks a month in 90 days after we fired our SEO agency, not by running more ads, but by fixing the asset that generates demand without ongoing spend. Paid and organic are not competitors. They are complements.

How to Launch Facebook Ads for Contractors in Five Steps

1. 1
   Build the back-end first

   CRM, automated follow-up sequence, dedicated landing page, and conversion tracking on the thank-you URL. Run a test form fill yourself and confirm every piece fires correctly before the campaign goes live.
2. 2
   Define the offer and the audience

   Pick one service, one geographic radius, and one specific offer. Choose the area by housing age and weather exposure, then target it by location and a few home-improvement interests. Start with your warm retargeting audience even if it is small, the data quality is better.
3. 3
   Build the creative around a specific problem

   Write the hook as the homeowner's problem, not your company's solution. Use video if you have it. If the video is creator-style and feels native to the feed, it will typically outperform a produced brand spot. Keep the call to action singular.
4. 4
   Launch and hold the budget steady

   Set a daily budget you can sustain without cutting the campaign, and do not make structural changes while the algorithm is in its learning phase. Patience here is not passive, it is the active choice that lets the data accumulate.
5. 5
   Measure what matters and iterate

   Track cost per booked appointment, not cost per lead. A campaign that produces expensive leads that book and close is better than a campaign that produces cheap leads that ghost. Once you have a baseline, test one creative variable at a time and document what you learn.

How Receipts Group Approaches Ads for Contractors

We are not a generic digital agency that bolted on a contractor division. The system we sell is the system we run. Our own windows and roofing operation in Southeast Florida, $23M a year, books about 650 in-home appointments a month from our own outbound floor. The call center and the site platform we sell are the ones we built for our own company. Eric Snyder is the founder of Receipts Group and Marketing Director at Safeguard Impact. He grew monthly revenue at Safeguard from $800K to $2M over fifteen months. Before that, he ran a Florida fix-and-flip and wholesale real estate operation as Operations Director, taking it from zero to $6M a year. The background is operator, not agency. For contractors in roofing, windows, HVAC, and remodeling, we run Facebook ads as part of a broader paid-social program. Management pricing is not published. We cover it on the call. A campaign built on a broken back-end wastes your budget, so build the back end first.