Why Putting a Spend Limit on Your CBO Is Killing Your Sales

Most media buyers make a brutal mistake when trying to scale campaigns on Meta Ads: they try to be smarter than the algorithm.

Today I want to talk about the worst of all these control mistakes: using Minimum Spend rules inside CBO campaigns (Advantage+ Campaign Budget).

If you're doing this, stop. You're literally throwing away the whole reason a CBO exists.

CBO is designed for efficiency, not fairness

Let's remember how a CBO works. You hand Meta a global budget at the campaign level (say, $100 a day), and let artificial intelligence decide in real time which Ad Set deserves that money based on conversion probabilities.

The algorithm is a mercenary hunting for cheap results. If Ad Set 'A' is getting purchases at $5, and Ad Set 'B' is costing $15... CBO will send 90% of the money to 'A' and almost nothing to 'B'. That's pure efficiency.

The problem with forcing minimum spends

What happens when you set up a CBO but force Meta to spend a minimum of $20 on Ad Set 'B'?

• You destroy the algorithmic advantage: You're forcing manual distribution onto a system designed for computational overhead. You're forcing Meta to spend money on an Ad Set that statistically (according to the AI) has worse odds of producing a good result, just to satisfy your manual rule.
• It's a disguised, more complicated ABO: If you want every ad set to spend equally or you want to test fixed budgets per audience, just create an ABO campaign (ad set-level budget). Doing CBO with minimums is like doing an ABO but adding a layer of algorithmic and technical complexity that only works against you.

"But I do it to give every ad a fair chance"

Many argue they use a minimum spend in CBO "to make sure every ad gets a fair chance to print" before the winner takes it all, without having to worry about managing the budget of 20 Ad Sets manually (as you would in an ABO).

Short answer: It's a waste of budget. On Meta Ads, if there's no organic spend from the algorithm = no scale is possible.

You don't need to give an ad "a fair chance" if, after the first thousand impressions, Meta Ads has already calculated that the user patterns of that ad won't match the ROAS of your winning ad.

If you want to control which ads get served to ensure diversity, use ABO for isolated testing, find the winning creatives and then consolidate them in one big unrestricted CBO campaign to scale.

Stop tying the machine's hands. If you want granular control in the testing phase, use ABO. If you want relentless efficiency and scale, use pure CBO.