What actually causes ad fatigue on Meta?
Audience depletion, not visual boredom. Frequency is a symptom rather than the cause. Fatigue arrives once the finite set of people who resonate with your positioning has already seen the ad and decided. That is why refreshing the creative works for about a week and changing who the ad is for works for a quarter.
Two different problems share one name
Creative wear-out is what most people picture: the same people see the same ad too many times and stop noticing it. This is real. It is also small, and a genuinely new execution fixes it.
Pocket depletion is different. Everyone in the audience who was going to respond to this positioning has now seen it and made their decision. No execution fixes that, because the problem is not how the message looks. It is that the people who would say yes have already said it.
On a performance chart these are almost indistinguishable — both show a decline. The responses are opposite, which is why so much fatigue advice half-works.
Why frequency is a poor diagnostic
Frequency rises when reach stops growing while spend stays constant. Read that again: the metric people use to diagnose fatigue is itself a description of depletion, expressed as a ratio.
Treating frequency as the cause leads somewhere unhelpful — cap it, rotate creative, widen the audience slightly — all of which address the arithmetic rather than the situation.
It also misses the harder case. An angle can be depleting at low frequency, if Meta is serving it to plenty of new people who simply do not care. Frequency looks healthy. CPM rises and CTR falls anyway.
What to look at instead
- Reach growth at constant spend. When reach flattens while spend holds, you are at the edge of the pocket.
- CPM trend against CTR trend. Rising CPM with falling CTR is depletion. Falling CTR with stable CPM is more likely an execution problem.
- The lifespan of successive winners. Winners that used to last two months and now last two weeks describe a pocket being consumed, not a team getting worse at their jobs.
Why "refresh the creative" only half-works
A genuinely new execution can re-engage people who scrolled past the first one. That is a real lift and it is worth having.
It does not add a single new person to the pool.
So the refresh produces a few good days, then settles back to the trend it interrupted. The team concludes that creative is getting harder, and the response is to produce more of it, faster — which is more work aimed at the wrong variable.
What resets fatigue
New positioning, not new production.
The change has to happen at the level of who the ad is for, which means changing at least one of three things: the desire it names, the belief system it assumes, or the person it pictures. Change none of those and you have repainted the same ad, however different it looks.
This is the practical test for a creative brief. If the brief describes a new look, it is a refresh. If it describes a new buyer, it is a new pocket.
Refresh before the decline, not after
Once you know roughly how long an angle lasts in your account — and after two or three cycles you will — the next angle should enter testing before the current one peaks.
That single change in cadence is the difference between managing fatigue and reacting to it. Reacting means every refresh happens while performance is already bad, budgets are already under scrutiny, and nobody has time to research properly. Which produces a rushed angle, which fails, which confirms that "creative testing does not work here."
A standing testing budget, running whether or not anything is currently on fire, costs less than that cycle does.