Across the accounts we've audited in the $1M to $5M range, the same pattern shows up almost every time. One ad is responsible for roughly a quarter of all conversions, running at a frequency well past healthy, and the founder has no idea, because everything else on the dashboard looks completely fine.

One apparel brand we looked at had exactly this. A single ad was driving 25% of conversions at a frequency of 4.4, meaning the same people had already seen it more than four times. ROAS looked healthy. Conversions kept rolling in. Underneath, the account was standing on one leg, and that leg was about to give out. A fragrance brand we reviewed had it even worse, its entire account resting on two influencer videos with no creative pipeline behind either one.

Why the same pattern shows up everywhere

Three reasons keep catching smart, capable founders off guard.

Meta rewards concentration in the short term. When a winning ad emerges, the algorithm pours budget into it because that's the highest probability play right now. ROAS climbs, CPA drops, and it feels like you've cracked the code. Meta doesn't factor in that the ad is hitting a frequency of 4.0, it will keep showing it to the same audience until performance degrades, then shift budget to whatever's next, which often isn't very good either.

Creative production is genuinely hard to keep up with. The best performing brands test five to eight new concepts a week. Most founders at this stage have one person on Canva, or a freelancer delivering three videos a month. There's typically a two to three week gap between noticing a winner is fatiguing and having a real replacement ready to test, and that gap is exactly where performance craters.

And almost nobody watches frequency. Founders track ROAS, CPA, and CTR religiously. Frequency, the actual leading indicator of creative fatigue, barely gets a glance, so by the time CPA starts climbing, the ad has usually been overexposed for weeks already.

What actually protects an account

You don't need a full content studio to fix this. You need a bench deep enough that no single ad can take the whole account down when it dies, which it eventually will.

The cadence that works is three to five genuinely new concepts tested per week, not five crops of the same idea with a different thumbnail, real variation in hook, angle, and format. Run new creative through a small dedicated testing campaign first, something like fifty to a hundred dollars a day, let winners graduate into scaling campaigns, and cut losers within three to five days without sentimentality.

Keep at least five ads actively spending and converting at any given time. Below that, creative production becomes the single most important thing on your list, above everything else on it.

The check every founder should run this week

Open Ads Manager, sort by conversions at the ad level over the last 30 days, and look at what percentage your top ad represents, and its frequency. Most founders assume they're diversified because they have ten ad sets running. Then they check, and eight of those ten are all pushing the exact same two ads.

Having many ad sets was never the same as having many creatives. Creatives are the only real protection against a single point of failure taking the whole account down, and the accounts that survive are the ones who found this out before it happened, not after.