articles — digital growth

Facebook and Instagram ads for trades: creating demand you can count.

Google catches people who already know they have a problem. Facebook and Instagram interrupt people who don't — yet. That's a genuinely different machine, useful for genuinely different jobs, and the trades that lose money on it are almost always running it like search with pictures. Here's what it's actually for, and how to run it so the output is booked work, not applause.

skeelx — 24 aug 2026 · 6 min read

One housekeeping fact that removes half the confusion: Facebook and Instagram ads are one system. Same accounts, same Ads Manager, same targeting — Meta places your ads across both apps. "Should I advertise on Facebook or Instagram" is rarely the real question. The real question is whether you should be paying for capture or creation.

Capture versus creation

Search advertising is capture: the demand already exists, fully formed, typed into a box — your ad just competes to catch it. Paid social is creation: nobody scrolling a feed is looking for a concreter, so the ad's job is to plant the idea, prove you're the obvious choice, and be remembered when the need arrives. Creation is slower to pay and harder to measure, but it's the only paid channel that works before the search happens — and for discretionary work (renovations, landscaping, solar, decks) the customer who was warmed for weeks often never runs the comparison search at all. If you're weighing this against search spend generally, SEO or Google Ads first covers that arithmetic; this article is about the second engine.

existing demand — someone typed the problem attention mid-scroll — no demand yet, no query capture engine — search creation engine — social matches a query already formed intent high · clicks dear wins by being findable and provable plants the job in the feed intent none — yet · views cheap wins with proof people stop for one business, two different machines remembered — until the need arrives the enquiry — the only output that counts, whichever engine made it the boost button feeds neither engine properly — it buys applause, not enquiries
capture and creation — two engines, one output that matters

The boost-button trap

The blue Boost post button is where most trade ad money goes to die. Boosting is real advertising with the controls removed: it defaults to optimising for engagement — likes, comments, video views — because those are the cheapest things Meta can buy you. So the machine dutifully finds people who like liking things, your post "does well", and the phone stays quiet. Real campaigns are built in Ads Manager, where you choose what the machine optimises for. The button isn't evil; it's just answering a question you didn't mean to ask.

Objectives and events that mean enquiries

In Ads Manager, tell the machine the truth about what you want: a leads objective — instant forms or calls — or a sales/conversion objective pointed at a real event, like the quote-request submit on your site. That requires the plumbing to exist: a pixel and, done properly, server-side events, so Meta learns from actual enquiries rather than page views. Then judge the account on one number — what a booked-job enquiry costs — not reach, not clicks, not "engagement rate". Everything Meta shows you by default is denominated in attention. You pay the bills in jobs.

Creative from real work

On social, the creative is the targeting: the ad that stops the right person is footage of their problem being fixed. Before-and-afters, the mess mid-job, the fault you found that the last bloke missed, the homeowner's own words — shot on a phone, captioned honestly. This consistently beats polished brand ads for trades because the feed is where people smell stock photography instantly. You don't need a videographer; you need a habit of capturing jobs and creative built from that raw material. Three or four variants, retired when they tire — creative fatigue, not targeting, is usually why a working campaign stops working.

Geography honesty

Meta's location targeting is a radius, and radiuses don't care about your actual service area — they'll happily spend a third of your budget across a river you never cross. Draw the geography as tightly as the tools allow, exclude what you can't serve, and accept the honest limitation: for a single-suburb operator, the audience may simply be too small for the machine to learn on. That's not a settings problem. It's a sign this engine may not fit the business yet.

When search should get the money

Honesty about the boundary: if the work is urgent — burst pipes, dead switchboards, lockouts — the customer goes straight to a search box, and creation has no time to work. Fund search ads and the map presence first, and treat paid social as the second engine you add when capture is already fenced and paying. Where creation earns its budget is discretionary, visual, plannable work — and even then, only when the enquiry is being counted honestly enough to prove it.

next

Two engines. One number that matters.

Start a project