For years, the advice has been simple: Make more ads. More hooks. More variations. More volume. And on the surface, it makes sense. If you test more, you should find more winners. But when you actually look at the data, something else is happening. A lot of that creative never even gets a chance.
The Hidden Problem With Creative Testing
Inside a recent agency account analysis across 572 ads and over $1M in spend, one stat stood out immediately: 66.8% of new creatives received less than $100 in spend.
At a ~$32 CPA, that’s nowhere near enough data to evaluate performance. In other words, most ads didn’t fail… They were never tested.
Only 10.7% of new creatives reached the $1,000 threshold required for meaningful evaluation. And even more interesting… Legacy creatives were almost 2x more likely to break through and receive real budget.
So while teams are being told to “just make more ads,” the platform is quietly doing something else: It’s concentrating spend into proven winners and limiting distribution to everything new.
The Algorithm Isn’t Built for Equal Testing
If you zoom out, this is more than just a creative problem, it’s also a distribution problem.
According to the analysis, a tiny percentage of ads are capturing the majority of spend:
- Just 4.2% of ads drove 77% of total spend
- The majority of ads received almost no delivery at all
That means your “testing volume” isn’t actually testing volume. It’s just creative production. And that’s a very different thing.
Introducing: Creative Breakthrough Rate
This is where a better metric comes in. Instead of asking, “How many ads did we launch?” You should be asking: How many ads actually broke through?
Call it Creative Breakthrough Rate: The percentage of creatives that receive enough spend to be meaningfully evaluated.
Because that’s the real bottleneck. Not how many ads you make… But how many the algorithm chooses to test.
In this case:
- 346 new creatives launched
- Only 37 reached meaningful spend
- Breakthrough rate: 10.7%
That’s the number that actually matters.
Why Volume Alone Doesn’t Work
This also explains something most teams feel but can’t articulate: You increase output… but results don’t improve because the system isn’t designed to distribute spend evenly.
It’s designed to:
- Find early signals
- Double down aggressively
- Starve everything else
And once a few ads gain momentum, they dominate. In fact, older creatives in this account continued to capture significant budget months later, even outperforming newer ones in some cases.
So the real challenge isn’t “making more content.” It’s breaking into the distribution loop.
The Real Lever: Differentiation
One insight that came out of the conversation around this data was that the algorithm may not even be evaluating each ad individually. It’s grouping similar creatives together. Which means if your “new” ads look like slight variations of existing ones, they’re competing for the same pool of spend and most will lose.
That’s why differentiation matters more than volume. Different angles. Different formats. Different structures. Not just iterations.
What Actually Moves Performance
If you look at where performance is coming from, it becomes clear:
- A small number of ads drive the majority of revenue
- Improving top performers has a bigger impact than killing underperformers
- Most ads are operational noise, not growth levers
In fact, improving the top 5 ads would drive more impact than removing hundreds of low-performing ones.
So instead of asking: “How do we make more ads?” A better question is: “How do we create ads that earn distribution?”
A Better Way to Think About Creative Strategy
If you take this seriously, it changes how you approach creative entirely and you stop optimizing for output.
You start optimizing for:
- Breakthrough potential
- Structural differentiation
- Early signal strength
Because once an ad gets traction, the system will scale it. But getting to that point is the real game.
The Bottom Line
Creative volume isn’t useless. But on its own, it’s not the lever people think it is because the constraint isn’t in the production. It’s in the distribution.
And if most of your ads never get a chance to perform, then your strategy shouldn’t be “make more.” It should be: Make ads that can’t be ignored by the algorithm.
And start measuring the thing that actually matters: Not how many you launch, but how many break through.
