CBO vs ABO: how to choose a Meta budget structure for testing and scaling VSL offers

CBO vs ABO comes down to who allocates the money. With ABO (ad set budgets) you set each ad set's budget, so every test gets the spend you choose. With CBO, which Meta now calls Advantage campaign budget, you set one campaign budget and Meta moves it between ad sets in real time. Most VSL buyers test with ABO and scale with CBO, but the right answer varies by offer, so test both.

What CBO and ABO mean today

The names come from Meta's older interface, and buyers still use them.

• ABO (ad set budget optimization): each ad set has its own daily or lifetime budget. Meta optimizes delivery inside each ad set, but it never moves money between them.
• CBO (campaign budget optimization): the campaign has one budget, and Meta distributes it across the ad sets. Meta now calls this Advantage campaign budget. Its developer documentation describes it as finding "the best available opportunities for results" and shifting budget in real time (Meta for Developers).

A few details from the same documentation matter in practice:

• You can set a minimum spend target per ad set under CBO, but it is a best-effort target, not a guarantee. Spend caps, on the other hand, are hard limits.
• The bid strategy (lowest cost, cost cap, bid cap or minimum ROAS) is set once for the whole campaign, and a minimum ROAS campaign can't switch to another strategy after it is created.
• A campaign with more than 70 ad sets can't change its bid strategy or turn Advantage campaign budget off.

There is also a third option: the Advantage+ sales campaign, where budget, audience and placements are all automated. In Meta's current setup a sales campaign runs that way when Advantage campaign budget, Advantage+ audience and Advantage+ placements are all on (Advantage+ campaigns documentation).

CBO vs ABO side by side

The two structures, point by point:

ABO vs CBO at a glance

ABO (ad set budgets)

• Who allocates spend: you, per ad set
• Best for: creative and audience tests
• Test clarity: high (every ad set gets its budget)
• Daily work: more (you move budget by hand)
• Typical use in VSL offers: testing hooks, formats, leads

Main risk: keeps spending on losers until you act.

CBO (Advantage campaign budget)

• Who allocates spend: Meta, across ad sets
• Best for: scaling proven ads
• Test clarity: low (new ads may get little spend)
• Daily work: less (Meta rebalances)
• Typical use in VSL offers: the campaign that carries most of the spend

Main risk: starves new ideas; hides why it works.

Neither structure makes a weak ad profitable. They change how fast you find out, and how much of the budget the winner gets once you have one.

When ABO wins

ABO is the testing tool. When you launch five new hooks for a proven VSL, you want each one to get enough spend to be judged. Under ABO you give each its own ad set and budget, and the comparison is fair because the budget is identical.

ABO also suits offers with little data. If your budget only buys one or two sales a day, splitting it under CBO across many ad sets means no ad set gets enough results to learn. One or two ad sets with deliberate budgets give a cleaner signal.

At high volume, some teams keep ABO and manage it closely. Rubens Araujo, whose team took offers from Brazil to Latin America and the US, said in a 2026 podcast interview that his campaigns ran on ABO at high volume, reviewed every half hour or so from large spreadsheets. That is a staffing decision as much as a strategy: ABO at scale only works if someone is watching.

When CBO wins

CBO is the scaling tool. Once you know which creative sells, you want the budget to follow the cheapest results across audiences and placements without moving it by hand every few hours. Meta sees auction prices you don't, so it usually allocates faster than you can.

The trade-off is clarity. Under CBO, Meta often gives most of the budget to the ad set with the most history, and a new ad in the same campaign gets a few dollars. You learn that Meta preferred the old ad, not whether the new one is good. Keep tests and scaling in separate campaigns, which is the approach in our Facebook ads for a VSL guide.

Give automated structures time. Pedro Andrade, a direct-response traffic manager, said in a podcast interview that Advantage+ campaigns can trail for several days and then overtake a manual setup, so he judges them over a longer window than a normal test. He also tests CBO and ABO on each offer rather than assuming one is better, because different offers respond differently.

Test structures that give a clean read

Three structures cover most VSL testing.

Three test structures

Three structures cover most VSL testing

ABO, one creative per ad set

Equal budgets, same audience settings, one ad in each ad set. Simple and fair.

The cost is overlap between ad sets that bid on the same people.

One creative per CBO campaign

Each campaign holds a single creative, copied into several ad sets. Meta chooses the best ad set for that creative, while each creative keeps its own budget.

Pedro Andrade described using this structure to isolate creatives: ten creatives means ten campaigns.

ABO tests feeding a CBO scaling campaign

Find winners in an ABO test campaign, then add each winner to a CBO campaign that carries most of the spend.

How much to spend on a test

Tie the test budget to what a sale is worth, not to a round number. Pedro Andrade's rule of thumb for affiliates is to give each test at least half of the expected commission, spread over at least three days because weekdays perform differently, and up to the full commission when cash allows. As a formula:

Sizing a test from what a sale is worth

Pedro Andrade's rule of thumb for affiliates, and the clicks it buys

Test budget per creative=at least half of the expected commission, up to the full commission when cash allows

Daily budget per creative=the test budget, spread over at least three days

Clicks per creative=test budget ÷ cost per click

Offer owners: replace the commission with your CPA ceiling.

A worked example with illustrative numbers: the offer pays a $150 commission. Each creative gets $75 to $150 over three days, or $25 to $50 a day. At a $1.50 cost per click, that buys 50 to 100 clicks per creative. If the click-through rate is weak after two days, the hook is the problem; change it before blaming the offer. If clicks are healthy but nobody buys, look at the VSL's retention and play rate before you kill the ad.

The same logic works for offer owners: replace the commission with your CPA ceiling. Watch hook rate for the first seconds and CTR for the click, and keep the testing discipline you would use on the VSL itself.

How to scale a winner: vertical, horizontal and the budget ceiling

Vertical scaling means raising the budget of the campaign or ad set that works. It is the fastest move, and it is also how campaigns break, because a higher budget pushes delivery into more expensive auctions. Many buyers step up in moderate jumps. Pedro Andrade's own example was bolder, a tenfold increase on a campaign that had proved itself; a jump like that only makes sense when results are strong and stable, and you should watch CPA closely afterward.

Horizontal scaling means duplicating the winner into new campaigns or ad sets at the same budget. He described going from five campaigns to twenty at the same budget per campaign, and duplicating winners to keep the number of active campaigns steady as older ones fade.

The budget ceiling is the spend level where a campaign's CPA stops holding. Every campaign has one. Find it deliberately before deciding a campaign or an offer has failed: raise the budget in steps, watch CPA for a few days at each step, and step back down when it breaks.

Cut gradually, not reflexively. When results fall below target, trimming the budget by around 30% often lets a campaign recover where a pause would end it. If two review rounds pass without recovery, then pause.

Every significant edit can send an ad set back into Meta's learning phase. Meta's API reports each ad set's learning status as LEARNING, SUCCESS (it exited learning) or FAIL (it isn't getting enough results to exit), along with the time of the last significant edit that reset it (learning stage reference). Check that status after each change before you judge the result.

Daily vs lifetime budgets

Both structures accept a daily or a lifetime budget, and the choice changes how Meta paces spend. According to Meta's documentation, a daily budget is an average: on some days Meta may spend up to 25% more than the daily budget. A lifetime budget is a total for the whole schedule, paced across it, so a $250 budget over five days might spend $25 one day and $75 another, depending on opportunities (Meta budgets documentation).

Gabriel Medeiros, who ran media buying at V Shred, described in a 2026 podcast interview a way to use that pacing. Sales were strongest at weekends, so before the weekend he shortened the campaign's end date, which made Meta spend the remaining lifetime budget faster. On Monday, when results dropped, he moved the end date back out. His site lists it as the "lifetime budget trick" (askgabe.co).

A worked example with illustrative numbers: $90,000 left on a lifetime budget with 90 days to run paces at roughly $1,000 a day. Move the end date to 30 days out and the same balance paces at roughly $3,000 a day. Move it back to 90 days on Monday and pacing falls back toward $1,000. Test it on one campaign first, and watch CPA and the learning status, because the pacing change is large. The same balance on the two schedules:

One lifetime balance, two end dates

Illustrative: $90,000 left on a lifetime budget, paced per day

90 days to runabout $1,000 a day

End date moved to 30 days outabout $3,000 a day

Which budget structure should you use?

Answer up to four questions about your offer and your account to find where to start:

Which budget structure to start with

Up to four questions about your offer and your account

1How many sales a day can your budget buy?

Only one or twoMore than that

2Do you have a proven creative for this offer?

No, the offer is newYes

3What is the next job?

Test new hooks or formatsGrow the budget

4What else is true of your account?

Steady purchase volume, many creativesA strong day-of-week patternNeither

Best fit

One ad set with a deliberate budget

Your budget buys only one or two sales a day. Split under CBO across many ad sets, no ad set gets enough results to learn, so one ad set with a deliberate budget gives a cleaner signal.

Best fit

ABO tests, or one creative per CBO campaign

A new offer with no proven creative yet. Give each new creative its own ABO ad set or its own CBO campaign, so every one gets enough spend to be judged.

Best fit

ABO tests kept apart from scaling

A proven VSL, and new hooks or formats to test. Run them in ABO tests, separate from the campaign that scales: a new ad inside a CBO campaign that already has a winner may get a few dollars and never prove anything.

Read more: Facebook Ads for a VSL: Tracking, Testing and Scaling

Best fit

A CBO scaling campaign

A proven creative and budget to grow. Put it in a CBO scaling campaign, so the budget follows the cheapest results without you moving it by hand.

Best fit

Try an Advantage+ sales campaign alongside CBO

A proven creative, budget to grow, steady purchase volume and many creatives. Keep the CBO scaling campaign, try an Advantage+ sales campaign alongside it, and give the automated campaign time before you judge it.

Best fit

Lifetime budget with a planned schedule

A proven creative, budget to grow and a strong day-of-week pattern. Start with a lifetime budget and a planned schedule, so spend can follow the strong days. Test it on one campaign first, and watch CPA and the learning status.

Read more: Gabriel Medeiros: How V Shred Scaled Cold Traffic

Then judge the structure the way you judge an ad: by sales recorded in your own checkout and the profit after refunds. Meta's dashboard tells you which structure Meta likes. Your orders tell you which structure makes money. The nutra marketing guide shows how to set the CPA ceiling those decisions depend on.

A checklist before you change budget structure

Tick these off before the change; the ticks stay in this browser:

Before you change budget structure

• Write down the CPA ceiling or commission the test is sized against.
• Separate tests from scaling: never judge a new ad inside a CBO campaign that already has a winner.
• Give every test at least three days unless it is clearly broken.
• Check the hook rate and CTR before blaming the offer, and the VSL's retention before blaming the ad.
• Scale in steps and note the CPA at each budget level, so you know the ceiling.
• Trim before you pause, and pause after two failed review rounds.
• Check the learning status after every significant change.
• Reconcile Meta's reported purchases with your real orders every week.

0 of 8 done