Scaling rules that don't blow up your CAC

A campaign hitting target CAC at $200 a day feels like an obvious candidate to scale to $1,000 a day tomorrow. It's also the most common way to watch that same campaign's CAC double within a week, because the algorithm has to find four times the audience overnight.

Ad platforms' delivery systems optimize within the audience and budget you give them, and a sudden budget jump forces them out of the pocket of users who were converting well into a much larger, less-qualified pool, and the platform doesn't pause to ask if that's a good idea.

The rule I hold to: no more than a 20–30% budget increase every 48–72 hours on any single ad set, letting the algorithm re-stabilize between jumps rather than lurching. Horizontal scaling (duplicating a winning ad set into new but adjacent audiences) usually outperforms pure vertical scaling past a certain budget size anyway.

Patience here isn't caution for its own sake. It's the difference between a campaign that compounds and one that spikes, overspends, and gets blamed on "ad fatigue" when the real cause was pacing.