CBO vs ABO: How to Structure Meta Ad Budgets in 2026

Few Meta ads debates run as long as CBO versus ABO. One camp says campaign budgets are the only way to let the algorithm work. The other says ad set budgets are the only way to test anything fairly. Both are right, about different jobs.

This guide explains what each setting does, where each one fits, and gives you account structures you can copy for testing, scaling and small budgets.

The definitions

ABO (ad set budget optimization) means you set a budget on each ad set. If you have three ad sets at $50 a day, each spends roughly $50 a day regardless of which performs best.

CBO (campaign budget optimization), which Meta now calls Advantage campaign budget, means you set one budget at the campaign level. Meta distributes it across ad sets in real time, pushing more spend to whichever it predicts will get the most results for the money.

In both cases, Meta still decides how to split spend among ads within each ad set. The difference is only who controls the split between ad sets: you (ABO) or Meta (CBO).

Advantage+ sales campaigns are effectively an extreme form of CBO, with one campaign budget and very few ad set-level controls.

How they compare

When to use ABO

Testing new creative or concepts

The main argument for ABO is fair testing. If you put a new concept in a CBO with an established winner, Meta will often send most of the budget to the winner because it has a track record. The new concept never gets enough spend to prove itself either way.

With ABO, each test ad set gets a guaranteed budget. You decide the threshold for success, and the data is cleaner.

Forcing spend on a priority

Sometimes you need a product, region or offer to get spend even if Meta would not choose it, such as a new product launch or a seasonal collection. ABO lets you guarantee that.

Very small accounts testing one thing at a time

If you run one ad set, CBO and ABO behave the same. If you run two at $15 a day each, ABO keeps them comparable.

When to use CBO

Scaling proven creative

Once you know which concepts work, CBO lets Meta shift budget minute by minute toward wherever results are cheapest. That is something no human can do manually, and it usually lowers blended cost per purchase at higher budgets.

Simplifying the account

One budget is easier to manage than ten. Fewer decisions also means fewer accidental resets of the learning phase.

Broad and Advantage+ setups

If you run broad targeting with a single ad set, the budget level hardly matters. If you run a few ad sets (for example, broad plus a customer-list suggestion), CBO lets Meta sort out which one deserves spend. See our broad targeting guide for why one broad ad set often beats many narrow ones.

The hybrid structure most accounts end up with

Many experienced advertisers run both, each for its own job:

1. Testing campaign (ABO). Each new concept or batch of creatives gets its own ad set with a fixed daily budget. Run for a set period, usually 5 to 7 days or until each ad set has spent a set amount.
2. Scaling campaign (CBO or Advantage+ sales). Winners graduate here. Meta allocates budget among them.
3. Retargeting (either). Usually one small ad set, so the budget type barely matters. See what still works in Meta retargeting.

A common split is 15% to 25% of spend on testing and the rest on scaling, but the right ratio depends on how quickly your creative fatigues and how much new creative you produce.

How to graduate a winner

When a concept wins in ABO testing, move it to the scaling campaign by adding the winning ad (using the same post ID if you want to keep its social proof) to the scaling campaign's ad set. Avoid simply raising the test ad set budget dramatically, which pushes it back into learning and blurs your testing data.

Setting test budgets in ABO

A test needs enough spend to tell a winner from a loser. A rough rule many buyers use: let each test ad set spend at least one to two times your target CPA before judging it, and ideally enough to get several purchases.

Hypothetical example:

• Target CPA: $40
• Each test ad set gets $40 a day for 5 days = $200, or 5x target CPA
• Testing 4 concepts per week = $800 a week, about $114 a day on testing

If your total Meta budget is $300 a day, that is about 38% on testing, which is high. You might test 2 concepts a week instead ($57 a day, about 19%). Plug your own numbers into the CAC calculator and the ad budget calculator to find a testing volume you can afford. Our guide on how to test creative on Meta without burning budget has more detail on thresholds.

CBO settings worth knowing

• Ad set spend limits. Within CBO, you can set minimum or maximum daily spend per ad set. Useful sparingly, for example to guarantee a new ad set gets at least some spend. Overusing limits turns CBO back into ABO with extra steps.
• Bid strategy. Highest volume (the default) spends the full budget. Cost per result goal and ROAS goal tell Meta to spend only when it expects to hit your target, which can underspend. Bid caps give the most control and need the most skill.
• Budget scheduling. Raise budgets for known high-demand periods such as sale days.

Small budgets: what to do under $100 a day

With small budgets, structure matters more than the CBO/ABO label. The main rule is to consolidate.

• Under about $50 a day: one campaign, one ad set, three to six ads. Budget type does not matter. Swap creative in and out as a simple test.
• $50 to $150 a day: one ABO testing ad set at $20 to $40 a day and one main ad set with your winners. Or a single CBO with two ad sets and a minimum spend on the newer one.

Splitting $60 a day across six ad sets means each one gets $10, which is rarely enough to learn anything. Our lean Meta ads playbook for small businesses covers this in more depth.

Common mistakes

• Testing inside CBO without minimums. New ads get starved and you conclude they lost.
• Too many ad sets. Ten ad sets with overlapping broad audiences compete with each other and slow learning.
• Reacting daily. CBO spend shifts day to day. Judge on 3 to 7 days of data.
• Big budget jumps. Large single-day increases can reset learning. Many buyers raise budgets in steps of about 20% to 30% every few days, though Meta has relaxed this over time. See how to scale Meta ads.
• Judging ad sets by ROAS alone in CBO. Meta may give one ad set little spend because it is targeting a smaller, more expensive pool. Look at blended results.

Budget structure on other channels

The same logic applies off Meta. On ChatGPT ads, OpenAI's Ads Manager sets budgets at the campaign level, with a minimum of about $25 a day per campaign in the US as of mid-2026. That makes the "consolidate first" rule even more important for smaller advertisers: a few well-targeted campaigns beat many thin ones.

Where SecondWin fits

SecondWin does not manage Meta budgets. It is a done-for-you service for ChatGPT ads. It uses your niche's longest-running Meta ads to find proven messages, writes original policy-checked ChatGPT ads, and runs and refreshes the campaigns in your own OpenAI ad account. Plans start at $99 a month on the pricing page, and recommended starting spend is $50 to $200 a day, billed by OpenAI directly to you.

Before you commit budget anywhere, get a free analysis of your site's buyer questions and proven messages.

FAQ

Is CBO better than ABO?

Neither is better overall. CBO (Advantage campaign budget) is usually better for scaling proven creative, because Meta shifts spend toward the cheapest results in real time. ABO is usually better for testing, because each ad set gets a guaranteed budget. Most mature accounts use ABO for testing and CBO or Advantage+ sales for scaling winners.

Why does CBO spend all the budget on one ad set?

Because Meta predicts that ad set will deliver the most results for the money, often based on its history. New ad sets with little data tend to get less spend. If you need a new ad set tested fairly, use a separate ABO test campaign or set a modest minimum spend on it inside the CBO.

How much budget do I need for CBO?

There is no fixed minimum beyond Meta's platform limits, but CBO works best when the campaign budget can support enough conversions for learning, often cited as about 50 per week. With a $40 CPA that is roughly $285 a day. Smaller budgets can still use CBO, but should keep the number of ad sets low.

Can I switch an existing campaign from ABO to CBO?

Meta generally lets you turn Advantage campaign budget on for an existing campaign, but doing so can reset learning, and switching back may be restricted. Many advertisers prefer to build a new CBO campaign and move winning ads into it, keeping the original ABO campaign for testing.

What is the difference between CBO and Advantage+ sales campaigns?

CBO is a budget setting you can apply to most manual campaigns: Meta splits one budget across your ad sets. Advantage+ sales campaigns go further, automating audience, placements and much of the structure, with fewer manual controls. Both use campaign-level budgets, but Advantage+ hands more decisions to Meta.