Here's a conversation I've had more times than I can count. A brand comes in frustrated: "our media buyer isn't performing." I open the account expecting a mess: bad structure, broken pixel, budget scattered across forty ad sets.
Instead I find a perfectly reasonable account… running the same four ads it's been running for three months.
That's not a media buying problem. That's starvation.
The platform changed. Most accounts didn't.
Meta's delivery system got genuinely good. Advantage+ style campaigns, broad targeting, algorithmic budget allocation. The machine handles most of what buyers used to knob-turn all day. Which means the inputs you still control got more important, and the biggest input by far is creative.
Targeting used to be the lever. Now your ad is the targeting. A hook that speaks to skeptical 45-year-old moms finds skeptical 45-year-old moms. The algorithm reads the creative and matches it to people. Your job is to give it enough distinct creatives that it has something to work with.
Pool rule 003Be prolific. Volume finds winners.
Why four ads is starvation
Creative testing is a hit-rate game, and the honest hit rate is humbling. Across the hundreds of ads I've shipped, most tests lose. That's not failure. That's the price of finding the handful of winners that end up carrying most of the account's spend. If you only test four concepts a quarter, you're buying four lottery tickets and wondering why you never win.
And the winners you do find? They fatigue. Frequency climbs, CTR sags, CPMs drift up. A winning ad is a melting ice cube. If nothing new is entering the pipeline while it melts, the account's performance cliff is already scheduled. You just haven't seen the date yet.
What a healthy cadence looks like
The teams that win treat creative like a production line, not a special event:
- A fixed monthly volume: I deliver 50 creatives a month across my accounts, on a schedule, whether or not anything is "broken." Volume is insurance.
- Concepts vs variations: net-new angles to explore new territory, variations (new hooks, formats, crops) to squeeze proven territory. You need both, and you need to know which is which.
- Angles from data, not brainstorms: the account tells you what to make next: which hooks held attention, which angle's CPA is drifting, what competitors are scaling. Reading that signal is creative strategy.
The uncomfortable diagnosis
If your account has been running the same handful of ads for months, no amount of bid tweaking, budget shuffling, or campaign restructuring will fix it. That's rearranging furniture in a house with no groceries. Feed the account. The "media buying problem" usually disappears within a couple of testing cycles.
And if your team can't produce enough creative to feed it, that's the actual problem to solve. It's the exact reason I built an AI-accelerated production workflow. Not because AI creative is magic. Because volume with judgment beats perfection at a trickle, every single time.