I run Meta ad campaigns for small e-commerce brands for a living, and I see the same mistake over and over: brands launch a single campaign with 5-6 ad sets targeting different "interests," let it run for 2 days, then panic and kill whatever isn't profitable yet.

Meta's algorithm needs ~50 conversions per ad set within a 7-day window to exit the learning phase and actually optimize delivery. If you're splitting a $30/day budget across 6 ad sets, none of them ever get enough data to learn. You're not testing — you're starving every ad set equally.

What actually works for brands under $1M/yr:

1. Start with 1-2 ad sets, broad targeting, CBO on. Let Meta's algorithm find your buyers instead of guessing at "interests" that are mostly noise now anyway.

2. Budget by conversion volume, not gut feeling. If you need 50 conversions/week per ad set to exit learning, work backwards from your CPA to figure out the minimum daily spend that's actually worth running.

3. Judge creative on hook rate and hold rate, not just ROAS, in week one. A new ad needs at least 3-4 days of real delivery before the ROAS number means anything. Killing it on day 2 kills your data, not the ad.

4. Layer in retargeting only after you have pixel volume. Too many brands build a 7-step funnel before they even have 100 site visitors a day. Simplify first.

If you're spending under $2k/mo on ads and not seeing consistent ROAS, it's rarely a "the algorithm hates me" problem — it's almost always a structure and patience problem.

Happy to answer questions if anyone's stuck on their account structure.