Why Creative, Not Targeting, Wins Meta Ads in 2026

For a decade, the pitch from every Meta Ads buyer sounded roughly the same: give us the budget, and we will find the audience through better targeting. Lookalikes, interest stacks, exclusion lists, bid caps. A whole discipline built around finding the right eyeballs and paying the right price for them.

That discipline is quietly becoming a smaller part of the job, and most advertisers have not updated their spending, or their expectations, to match.

The machine absorbed the targeting

Meta has merged what used to be two separate paths, manual campaign setup and Advantage+, into a single creation flow, with Advantage+ optimisation switched on by default for audience and placements.

Two things are worth being precise about here, because the loose version of this claim gets repeated a lot.

It is a default, not a mandate. Advertisers can still see an on/off state for Advantage+ and can turn it off. Automation is the path of least resistance now, not the only path.

Defaults decide what most accounts actually do. That is the real effect. When the automated route is the one the interface hands you, the median advertiser takes it. So two advertisers running a similar product to a similar rough audience on a similar budget increasingly get similar delivery from Meta’s side of the equation.

The lever that used to separate a sharp media buyer from a mediocre one has been absorbed into the platform for most accounts. What has not been absorbed is the creative.

The number people keep citing, and why we will not lean on it

A figure circulates widely in performance-marketing analysis putting creative quality’s share of the variance in Meta Ads results as high as 56%, ahead of targeting, bidding and placement.

Treat that number with caution. It comes from practitioner estimates rather than anything Meta has published, and estimates like it harden into facts through repetition alone. We have seen the same thing happen to half a dozen advertising statistics that everyone quotes and nobody can source.

You do not need the number to see the direction. Watch what the platform has automated over the past two years and the conclusion arrives on its own: creative is the largest lever still sitting in the advertiser’s own hands. That is either very good news or a real problem, depending entirely on whether the ad itself is any good.

What this actually changes

For brands buying “ad management.” If an agency’s main pitch is that they will optimise your campaigns, ask what is left to optimise once the platform automates targeting and bidding by default. Increasingly the honest answer is: less than a stronger piece of creative would do for the same spend. That is one of the questions worth asking any agency before you sign.

For brands running the same three video variations. The system needs something to choose between. Feeding it one hero ad and two afterthoughts gives it almost no room to work. Production capacity, the ability to actually make more and better creative, matters more than it used to, not less.

For anyone whose value stops at the dashboard. This shift is uncomfortable for an agency whose skill is in the ads manager. It is the opposite for one whose skill starts at the camera.

What to do about it in practice

The useful response is not “make better ads”, which is advice nobody can act on. It is closer to this:

1. Budget for volume, not just for one hero. Plan a shoot that produces a bank of cuts, formats and hooks rather than a single polished film. This is the same economic argument behind shooting once and cutting many: the crew is the expensive part, and it costs almost nothing to come away with more once they are already there.
2. Vary the thing that actually matters. Five versions of the same ad with a different colour grade is not a test. Different opening three seconds, different claim, different structure, different format: that is a test.
3. Judge the creative before the platform does. If nobody on your team would stop scrolling for it, the algorithm will not save it. It will just find the cheapest people who also will not stop.
4. Separate the two questions. “Is the account set up correctly” and “is the work any good” are different diagnoses with different fixes. Most underperforming accounts we look at are healthy on the first and failing on the second.

Where this leaves an agency like ours

We have always led with craft over spreadsheets, building work meant to hold attention rather than fill a media plan. That used to be a stylistic preference. With the platform doing most of the mechanical optimisation by default, it is closer to the whole game.

None of which means the media side stops mattering. Budget structure, measurement and the quality of the signals you feed the system are still real work, and doing them badly still wastes money. The point is narrower: the part of the job that used to create the biggest gap between two advertisers has largely been automated, and the part that has not is the part most brands underinvest in.

The brands that win the next few years of Meta Ads will not be the ones with the cleverest audience stack. They will be the ones who never stopped taking the ad itself seriously.

---

If your last “optimisation” conversation with an agency was mostly about targeting, that conversation is already out of date. See how we approach performance marketing.