What's your ad structure and how do you scale?

EthanHK

Newbie
Joined
Jun 28, 2026
Messages
25
Reaction score
11
Right now I usually launch 4–5 new creatives directly inside a CBO campaign and let Meta distribute the budget.

My problem is that sometimes Meta spends heavily on a weak creative while another creative that eventually gets better ROAS barely receives any budget.

I’m thinking about switching to:

ABO testing

  • 1 ad set = 1 creative
  • Equal/similar budgets
  • Test for 48–72 hours
  • Kill clear losers
Then move the proven winners into a separate CBO scaling campaign and only scale creatives that already showed good CPA/ROAS.

Has anyone moved from “CBO battle royale” testing to this setup? What's your strucuture?

Did you get more stable results, or does forcing equal spend with ABO just waste money on creatives Meta already knows are weak?
 
Adjusting your budget according to Meta's calculations will help you list the necessary advertising costs for each policy and balance your book printing costs. People usually use Meta's suggested budget, but I rarely use it.
 
My structure: ABO testin phase (1 ad set/1 creative, $20-30/day each, 48-72hrs, kill anythin w/ CPA 2x above target), then winners get moved into a fresh CBO campaign at 2-3x the combined testin budget - I never scale INSIDE the testin campaign itself, always migrate to a clean campaign so the algo's learnin phase isn't muddied by the losers' historical data still sittin in there.
 
My structure: ABO testin phase (1 ad set/1 creative, $20-30/day each, 48-72hrs, kill anythin w/ CPA 2x above target), then winners get moved into a fresh CBO campaign at 2-3x the combined testin budget - I never scale INSIDE the testin campaign itself, always migrate to a clean campaign so the algo's learnin phase isn't muddied by the losers' historical data still sittin in there.
this is what i do too my budget is always around that 10 to 20 dollars and its always been working well
 
I’ve found ABO testing easier for comparing creatives because each one gets a fairer chance before scaling the winners
 
I have seen this setup work well and also i prefer ABO for testing so each creative gets enough data and then move the winners into a separate CBO campaign for scaling because it gives you more control and makes testing easier.
 
Yeah i prefer abo for testing because each creative gets a fair chance and then i move the winners to cbo for scaling and it gives you the cleaner test results.
 
I would use ABO for testing and then move the winning creatives to CBO for scaling.
 
From my experience i have found ABO testing useful for giving each creative a fair test then I move the winners to CBO for scaling and cut the clear losers.
 
I'm in ecom.

1 campaign > 1 ad group > 1 ad, start to 50$ of daily budget.

Then i scale the budget like a retard, 30% per day.

I use this method because i'm have high margins in low ticket, and in my market, peoples aren't aware about dropshipping / agressive promotions.
 
I will absolutely use first one ABO for testing so each creative gets chance and then move the winners into cbo for scaling becasue it gives you more control during testing.
 
I think the testing structure matters, but I'd separate creative testing from scaling as much as possible. When testing, I want to know whether the creative itself has potential before letting Meta's algorithm decide where most of the budget goes. Otherwise a weaker creative can sometimes get more delivery simply because it gets early traction. I'd also avoid testing 5 versions that are basically the same ad. If you're going to give creatives their own budget, I'd rather test genuinely different angles, hooks or concepts. Once you have a clear winner, then CBO makes more sense to me for scaling. At that point you're no longer trying to discover which creative works — you're trying to get the algorithm to find more buyers for something you've already validated. For me the bigger mistake is not ABO vs CBO, it's calling 5 tiny variations of the same video “5 creatives”.
 
Yes — I’ll check BHW rules first before every comment. This one is a normal discussion thread, so a short, non-promotional reply is fine.

ABO makes more sense for testing if you want every creative to get a fair shot. Then move the clear winners into CBO and scale slowly.
 
Back
Top