CBO or ABO in Meta ads? Campaign budget vs ad set budget

What are CBO and ABO?

• CBO (Campaign Budget Optimization): The budget is set at campaign level. Meta distributes it across the campaign's ad sets, favouring whichever is delivering the best results at the time. In Meta's interface today it's called Advantage+ campaign budget.
• ABO (Ad Set Budget Optimization): Each ad set has its own budget. You decide how much goes to each audience.

Neither is “right” or “wrong”; they're used for different jobs.

When does CBO work better?

• When scaling: Once winning audiences and creatives are clear, raising budget from one place and leaving distribution to the algorithm makes management easier.
• With similar-sized audiences: If ad set audiences are close in size, CBO distributes more evenly.
• With broad targeting: In accounts using open targeting and large audiences, the algorithm's flexibility is an advantage.

When does ABO work better?

• When testing: When trying a new audience or creative, you want each to get an equal, guaranteed budget. In CBO the algorithm may spend almost nothing on some ad sets, leaving the test inconclusive.
• When audience sizes differ a lot: If a broad interest audience and a small retargeting audience share a campaign, CBO leans towards the big one and the small one doesn't get enough budget.
• When you need strict budget control: If you must allocate an exact amount to a product or region, ABO guarantees it.

Why does the learning phase matter?

Each ad set goes through a learning phase before its performance stabilises. According to Meta, an ad set needs around 50 optimisation events (such as purchases) within roughly 7 days of its last significant edit to exit learning.

So:

• Splitting budget across many small ad sets can leave none of them able to exit learning.
• Frequent big changes to budget, targeting or creative can reset learning.

CBO can help learning finish faster by concentrating spend on fewer ad sets; with ABO, you have to make sure each ad set's budget is enough to learn.

A practical setup: test first, then scale

Many teams use both:

1. Test campaign (ABO): New creatives or audiences run for 5–7 days in ad sets with equal budgets.
2. Decide: Pick the winners that stay above break-even ROAS. We explain how to find your target ROAS in our ROAS guide.
3. Scale campaign (CBO): Move the winners into a CBO campaign and increase budget gradually.

This structure keeps tests fair and lets winners grow quickly.

Common mistakes

• Treating an ad set that doesn't spend in CBO as a “bug”. If the algorithm isn't spending on it, the others are usually delivering better.
• Putting very different audience sizes in one CBO campaign. Small audiences won't get budget; move them to a separate campaign or use ad set spend limits.
• Changing budgets every day. Learning keeps resetting and results never stabilise.
• Judging campaigns only by Meta's ROAS. Also watch the effect on total revenue — see why ad and store data don't match.

Which should you choose?

Short answer: ABO when you're testing, CBO when you're scaling. If your account runs a few campaigns with broad audiences, CBO usually makes management easier. Whichever you pick, base decisions on weekly results and real revenue; CiroFly makes that easier by showing Meta spend next to store and marketplace revenue on one screen.