Most underperforming paid social campaigns do not have a targeting problem. The audience is sensible, the budget is enough to learn from, and the landing page converts fine when people actually reach it. The ad is what is letting them down.
That matters more than it did a few years ago. Meta and TikTok now do most of the audience work themselves. You hand over a budget and a conversion goal, and the system goes and finds the people. What you still control is what those people see. Creative has quietly become the main lever you have left.
So it is worth being deliberate about how you build it. Here is the approach we use.
Lead with the hook, not the product
You get about two seconds. In that window someone decides whether to stop or keep scrolling, and they make that call before they have read a word of your copy.
Most business owners spend those two seconds on a logo, a product shot, or a headline that describes what the company does. All of that is information the viewer has not asked for yet. The hook has to earn the right to deliver it.
A hook can be a question your customer is already asking themselves. It can be a visual that looks out of place in a feed of polished ads. It can be someone talking straight to camera about a problem, in the words they would use down the pub. What it cannot be is a slow build. Front load the interesting part and explain later.
One idea per ad
The most common fault we see in client accounts is an ad trying to carry four messages. Quality, price, service and range, all in fifteen seconds.
Pick one. If you have four things to say, that is four ads, and running them separately tells you which message people respond to. An ad carrying one clear idea also gives the algorithm a cleaner signal to optimise against.
This is where a lot of brand guidelines get in the way. Guidelines built for a website or a brochure often produce ads that look tidy and say nothing. Ads need to be legible at a glance and a bit louder than your other collateral.
Make the proof specific
Vague claims are invisible. “Trusted by businesses across Melbourne” reads as filler, because every competitor says the same sentence.
Specifics land. A number, a timeframe, a named suburb, a real before and after. If you finished a job in Brunswick in six days that normally takes three weeks, that is the ad. If enquiries went from four a month to twenty, say the four out loud, because the starting point is what makes the jump believable.
The test we use on client work: could a competitor put their logo on this ad and run it unchanged? If yes, the proof is not specific enough yet. Most ad copy fails that test, which is why most ads get scrolled past.
Give people one thing to do
The call to action should ask for the next step the viewer is actually ready to take. Cold audiences are not ready to book a consultation. They might be ready to watch a two minute case study or read a guide.
Match the ask to the temperature of the audience and you will see cost per lead drop without touching the targeting. We go deeper on this in our guide to getting more from paid social.
Plan for fatigue before it happens
Every ad decays. Frequency climbs, the same people see the same thing repeatedly, and performance drifts down. This is not a sign the creative was bad. It is just what happens.
The practical answer is to have the next batch ready before you need it. When we shoot for clients we plan for volume: multiple hooks, several formats, and cutdowns at different lengths from the same session. It is far cheaper to capture fifteen usable assets in one shoot than to go back three times because each round only produced two.
A rough guide for a small account is three to five fresh creatives a month. Bigger spenders need more.
Test properly, or do not bother
Testing four ads at once with a small daily budget teaches you nothing. Each one gets a fraction of the spend, none reaches significance, and you end up making decisions on noise.
Two versions with one clear difference between them is a better test. Change the hook and keep everything else the same. Once you know which hook wins, change the next variable. It is slower and it produces answers you can trust.
Give each test enough budget and enough time. Most accounts we take over have been switching things off after two days. If you are weighing up where to spend at all, our comparison of Google Ads, Meta and LinkedIn is a good place to start.
Where Melbourne businesses tend to go wrong
Melbourne is a design literate city, and that cuts both ways. Local businesses often over invest in polish and under invest in message. We regularly see beautifully shot campaigns that never say what the business actually does or why anyone should care.
The other pattern is treating paid social creative like a scaled down TV commercial. Different medium, different rules. Sound off by default, vertical, thumb hovering over the skip. Work with those constraints rather than against them.
If you would rather hand this over, our creative team produces the concepts, video and design, and our paid social team runs the media. Same building, same strategy, no handballing between two agencies who never speak.
Frequently asked questions
Three to five is a sensible starting point for a small account, built around different hooks rather than different colours of the same idea. That gives the platform something to choose between without spreading your budget so thin that nothing learns.
Watch frequency rather than the calendar. Once the same people are seeing an ad repeatedly and results start sliding, it is time. For most small accounts that lands somewhere around monthly, though a high spend account will burn through creative faster.
On Meta and TikTok, yes, in the sense that the platforms handle most audience decisions automatically. Targeting still matters for exclusions, retargeting and budget structure. Creative is the variable with the most room left to move.
Static still works, particularly for retargeting and offer led ads. Video generally does more for cold audiences because it holds attention longer and gives you room to build an argument. Most accounts we run use both.
It varies with format and volume. A single day shoot producing a batch of assets is usually better value than commissioning ads one at a time. We scope it transparently, so you know what you are getting before anything is booked.
Want creative that earns its place in the feed?
Tell us about your business and what you are currently running. We will come back with an honest read on whether the creative is the problem, and what we would do about it.