Fewer Meta ad sets almost always perform better
Splitting budget across many small audiences feels precise. In practice it starves the algorithm of data and keeps everything stuck in learning.
We regularly inherit Meta accounts with 15, 20, 30 active ad sets — one per interest, age band or country. Each one gets a small slice of budget. Almost none of them ever leave the learning phase.
Why it hurts
Meta's delivery system needs about 50 optimisation events per ad set within a week to stabilise. An ad set spending €20 a day on a €40 conversion gets maybe 3–4 conversions a week. It will never get there. Its costs stay volatile and higher than they should be.
Many small ad sets also overlap. Two ad sets targeting similar people end up bidding against each other in the same auctions.
What we do instead
- Consolidate into a few ad sets with enough budget to reach ~50 conversions a week each — or optimise for a higher-volume event until they can.
- Let broad or Advantage+ audiences do the targeting, and use creative to speak to different segments.
- Keep prospecting and retargeting clearly separated, with existing customers excluded from prospecting.
- Make fewer, bigger edits — every significant change can reset learning.
On Meta, creative is the new targeting. Put the effort into more and better ads, not more ad sets.
Free written review. We reply within two working days.