We’ve been auditing some serious six-figure monthly spend accounts lately, and I’m starting to see the same problem over and over again.

Three hundred or more new ads launched in a month. It looks like an incredibly sophisticated Meta account and creative operation. Until you look at the spend.

More than half of those ads never received enough delivery to tell us much of anything.

So what exactly did we test?

Volume and variation aren’t the same thing

I think we’ve taken the idea of creative velocity a little too far.

One thing we hear during these audits is: “We were told Andromeda wants volume.” But volume and variation aren’t the same thing.

If I take one winning image and make 10 versions with different copy overlays, I technically made 10 ads. I didn’t necessarily give Meta 10 meaningfully different things to learn from.

I’d much rather test fewer, genuinely different hypotheses: a different persona, objection, or reason to buy. A different format or way of delivering the same idea. Then give those tests enough room to actually tell us something.

When something works, iteration becomes powerful

Take the winning idea into UGC. A still. A carousel. A creator. Whitelisting. Change the hook. Change the execution. And carry the same message onto the landing page.

Now we’re building on a learning instead of starting from zero again.

I don’t think creative velocity should be measured by how many ads your team shipped this month. I’d rather know: what did we learn this month that makes next month’s creative better?

If the answer is unclear after launching 300 ads, we have a testing problem. The volume is not the issue anymore.