One of the most dangerous things you can do when running Meta Ads is decide whether a creative is good based purely on whether you like it.

We were reminded of that recently with a client we've worked with for more than two years.

They wanted to review some of the Meta Ads we had running and share a few ideas of their own.

Always welcome.

Then one particular creative came up.

It was AI-generated.

Quite busy.

A little more wordy than some of the ads we'd normally create.

Their immediate reaction?

"I don't like that one."

Fair enough.

There was just one problem.

Their audience absolutely loved it.

The Ad Had Generated 100+ Event Registrations In 48 Hours

When we looked at the actual performance, that single image had generated more than 100 registrations for the client's in-person events in the previous 48 hours.

Suddenly, nobody was particularly concerned about whether it was the prettiest advert in the account.

And the performance didn't stop there.

Fast forward, and that same creative generated another 130 registrations in a single day across three different event locations.

By that point, we were spending thousands of pounds putting it in front of more people.

Not because we thought it was beautiful.

Not because the client had suddenly decided it was their favourite creative.

Because it was working.

And, more importantly, it was generating tens of thousands of pounds in return.

Your Opinion Isn't The One That Matters

This is one of the biggest lessons we've learned managing Meta Ads over the years.

We're not the target audience.

And neither is the client.

It sounds obvious, but it's incredibly easy to forget.

When you're close to a business, you naturally develop opinions about what its advertising should look like.

You might prefer:

Cleaner designs.

Shorter copy.

Professional photography.

Video over static images.

Minimal text.

But the people seeing the ads might think completely differently.

That's why we try very hard not to let personal preference influence our Meta Ads decisions.

The audience gets the final vote.

The Best-Performing Meta Ad Isn't Always The Prettiest

Some of the best-performing creatives we've ever run would never win a design award.

And that's fine.

An advert has a job to do.

It needs to stop the right person.

Communicate something they care about.

Create enough interest for them to take the next step.

And ultimately contribute towards a profitable campaign.

If a beautiful, perfectly polished creative doesn't generate leads, registrations or sales, being beautiful doesn't make it a successful advert.

Likewise, if a slightly busy image generates hundreds of registrations and a strong return on ad spend, we're not going to turn it off because somebody doesn't like the design.

Performance has to come before preference.

The Same Applies To AI-Generated Meta Ads

There's another interesting lesson from this particular campaign.

The creative was AI-generated.

There's a lot of discussion around whether audiences like AI creatives, whether businesses should use them, and whether traditional designs will perform better.

Our view is much simpler.

Test them.

Your audience doesn't know the rule you've created in your head about what supposedly works.

They don't care whether an image was created with AI, photographed professionally or designed from scratch in Canva.

They care whether the advert resonates with them.

Sometimes AI-generated creatives perform brilliantly.

Sometimes they don't.

The same is true of professionally designed creatives.

The only reliable way to know is to put them in front of the market and measure the response.

This Is Why We Never Fall In Love With Creatives

When we create something we're particularly proud of, it's tempting to want it to work.

We've all done it.

You've spent ages getting the design right.

The copy feels perfect.

Everyone internally loves it.

Then you launch it...

And nothing happens.

Meanwhile, the creative somebody nearly didn't approve is generating leads all day.

Meta Ads can be humbling like that.

It's why we've learned not to become too emotionally attached to individual creatives.

We would rather be wrong and generate a fantastic ROI for a client than be right about which advert looked best.

The data tells the truth.

Every Meta Ads Account Is Different

This is also why we're wary of blanket statements about Meta advertising.

"Short copy performs better."

"Video always beats images."

"AI creatives don't work."

"People don't read text-heavy ads."

There may be trends worth paying attention to, but there are very few rules that apply universally.

Every business is different.

Every audience is different.

Every offer is different.

We've seen simple creatives outperform highly produced videos.

We've seen longer ads beat short ones.

We've seen angles we thought would perform brilliantly fall flat.

And we've seen creatives nobody expected much from become some of the biggest winners in an account.

That's why testing is everything.

Once The Data Speaks, We Act

Finding a winning creative is only the first part.

Once the data showed us how strongly this particular image was performing, our job changed.

Now we needed to maximise the opportunity.

That meant putting more budget behind it.

Watching whether performance held as spend increased.

Using the learning to inform future creative tests.

And getting it in front of more of the right people while the opportunity was there.

Scaling is not about spending more blindly.

It's about recognising when the data gives you permission to spend more, then acting quickly enough to capitalise on it.

The Real Takeaway

The client didn't like the ad.

If we'd made the decision based on personal preference, we could easily have switched it off.

Instead, we looked at the data.

More than 100 registrations in 48 hours.

Another 130 registrations in a single day.

Tens of thousands of pounds being generated in return.

That matters far more than whether any of us would have chosen that particular creative in a meeting.

Because the best-performing Meta Ad isn't necessarily the one you like most.

It's the one your market responds to.

That's why we test.

We measure.

We optimise.

And then we let the data make the decisions.

Built something significant and ready to scale it properly? See how our Done For You Meta Ads service works.