Facebook ads for roofers: why most of them fail before the money is spent

Start with the fact that decides everything else: nobody opens Facebook looking for a roofer.

That sounds obvious. It isn't, because most roofing ads are written as though the opposite were true. They lead with "licensed and insured," a stock photo of a shingle, and a call to action asking for a free estimate — which is exactly what you'd write if the person had gone looking for you.

They didn't. They were watching a video about someone's dog.

Search and social are opposite problems

On Google, demand already exists. Someone typed "roof leak repair" because water is coming through their ceiling. Your job is to be there and be trusted. That's demand capture.

On Facebook, no demand exists yet. Your job is to make someone who was not thinking about their roof start thinking about their roof. That's demand generation, and it's a completely different piece of work.

Run capture-style creative in a generation environment and you get the outcome nearly every roofer describes: money spent, some leads, all of them tyre-kickers.

What actually stops the scroll

The first second is the whole game. Not the offer, not the credentials — the first second.

Things that work in that first second, in rough order:

• Damage the homeowner recognises. A lifted shingle. Granules in a gutter. A stain spreading across a ceiling. People have seen these in their own house and not known what they meant.
• Something visually surprising. Drone footage moving over a roof reads as unusual on a feed built for faces and pets.
• A specific, local reference. "If your roof went on before 2005 in Pinellas County" outperforms "quality roofing services" because it makes a claim about you, the viewer.

Things that don't work: your logo, your truck, the words "licensed and insured," a stock photo of anything, and any headline containing the word "solutions."

The offer problem

A free estimate is not an offer. Every roofer offers it, it commits the homeowner to an appointment they have to be home for, and it signals that a salesperson is coming.

Better options for someone who wasn't looking for you:

• A free roof-age check from aerial imagery, with no visit required
• An honest answer on whether they need a replacement or four more years
• A short guide to what an insurance adjuster actually looks for

Notice the pattern. Each is lower commitment than an in-person estimate, and each is genuinely useful to someone who is merely curious. That's the state most of your Facebook audience is in.

What to shoot

You do not need a production company to make this work, but you do need real footage of your own jobs. Stock imagery is the single fastest way to look like every other roofer running ads in the county.

The short list:

• Drone footage over an active tear-off
• Close detail of failure — lifted, cracked, curling, or bare
• Decking underneath once the old roof is off, especially where there's rot
• Your crew working, from ground level, unstaged
• Thirty seconds of you explaining one thing homeowners get wrong

That last one matters more than it looks. A named human being from your county explaining something useful is the thing a national lead-generation company physically cannot produce.

The measurement mistake

Most roofing campaigns are optimised for form fills, and then the roofer is surprised to receive a pile of low-quality form fills. The campaign did exactly what it was told.

If your conversion event is a form fill, the platform will buy you the cheapest form fills available — which means the people least likely to buy. Set the conversion to something further down: a booked inspection, or a qualified call. Fewer leads, better ones, and a cost per lead that looks worse on a report while making more money.

The realistic timeline

Facebook ads for a roofer are not a switch. The people you reach in week one were not planning to replace a roof in week one.

What tends to happen: engagement in the first fortnight, a handful of early leads from the small slice of the audience who already knew they had a problem, and then a slow build as the retargeting pool fills with people who watched but weren't ready. Most of the value arrives in months two and three, from people who saw you weeks before they called.

Which is why running these ads only during storm season is backwards. The audience you build in spring is who you sell to in the summer.

The short version

• Nobody is searching for you on Facebook, so capture-style creative fails there.
• The first second decides everything. Lead with damage a homeowner recognises, not credentials.
• A free estimate isn't an offer. Give something useful that requires no appointment.
• Shoot your own jobs. Stock footage makes you look like everyone else.
• Optimise for booked inspections, not form fills, or the platform will buy you junk.
• Judge it at month two or three. The retargeting pool is where the money is.