Spent the last few years rebuilding Meta accounts for DTC brands crossing $100k/month, and the pattern is almost always the same:

1. Campaign structure is bloated from panic-adding. Every time ROAS dips, someone launches a new campaign instead of fixing the one that's underperforming. Six months later there are 40 campaigns, most of them ghosts eating budget with no real signal.

2. Creative volume is treated as a "nice to have." Brands doing real revenue are still running 5-8 ad variants a month. Meta's algorithm needs volume and variance to find winners — if you're not testing 2-4 new angles a week, you're not actually testing, you're guessing.

3. Nobody separates "we lost money on this test" from "this test failed." A killed ad isn't wasted spend if it told you something. The brands that scale fastest treat every dollar as a data-buy, not just a sale-buy.

4. Reporting is screenshots, not decisions. A ROAS chart doesn't tell you what to do Monday morning. What matters is: keep, kill, or scale — and why.

If your MER has been flat for 2+ months despite "optimizing," it's usually not a targeting problem. It's a structure + volume + decision-speed problem.

Happy to break down what a rebuild actually looks like if anyone's stuck on this.