Why Are My CBO Campaigns Sending 80% of Spend to One Ad Set: Budget Reallocation Bias Explained

You launched a CBO with five ad sets. By day three, one ad set is consuming 80% of the daily budget. The other four are starving — sometimes spending ₹200 against a ₹2,000 allocation.

This is budget reallocation bias, and it's how Meta's CBO is designed to work. The problem is that the algorithm's idea of a 'winner' often isn't your idea of a winner — and the concentration can lock you into the wrong cohort for weeks.

First: Confirm This Is Bias, Not Just Performance

Sometimes the winning ad set actually is the best. Validate before you intervene.

• Check ROAS, not just spend. If the dominant ad set has 4.2x ROAS and the others sit at 1.8x, CBO is doing its job.

• Compare cost per purchase, not CTR. CBO optimizes for the lowest-cost conversion, which can mean low-AOV buyers — see our [retargeting funnel guide](https://www.wittelsbach.ai/post/retargeting-funnels-for-d2c-beyond-abandoned-cart-sequences).

• Look at 7-day windows. Day-1 concentration is normal during learning.

The Root Cause: CBO Optimizes for Marginal Cost, Not Strategic Balance

CBO doesn't care about your portfolio strategy. It asks one question every few minutes: 'Where can I get the next cheapest conversion?' If ad set A has a 12% lower estimated cost per purchase than ad set B, every additional rupee goes to A until the marginal cost equalises.

For Indian D2C accounts the bias usually concentrates around:

• The cheapest cohort — often Tier-2 cities buying the lowest-AOV product.

• The retargeting ad set, which always looks 'better' on cost per purchase but cannibalises organic recoveries.

• The ad set with the strongest creative, even if it's targeting a tiny audience that will saturate in 5 days.

The 4-Step Diagnostic

Step 1: Audit AOV by Ad Set

Pull average order value per ad set. If the dominant ad set has ₹1,200 AOV and the starved ones have ₹3,800 AOV, CBO is sub-optimising for raw conversion count, not revenue.

Step 2: Check Audience Saturation

If the winning ad set is a 50K lookalike and reach has already hit 80%, you're 7-10 days from a CPM spike. Concentration plus saturation equals collapse.

Step 3: Look for Retargeting Cannibalisation

Is the dominant ad set a warm/retargeting audience? It will always win on cost-per-purchase because those buyers were going to convert anyway. CBO is taking credit for organic recovery.

Step 4: Check Creative Concentration

Sometimes one creative is doing all the work. Pause it for 24 hours — if budget redistributes evenly, the issue was a single ad, not the ad sets.

The Fix: Three Patterns That Actually Work

1. Set minimum ad set budgets at 15-20% of total. This caps CBO's ability to starve cohorts.

2. Split CBO by funnel stage. One CBO for cold prospecting, one for warm retargeting. Stops cannibalisation.

3. Use Bid Cap on the dominant ad set if it's an audience you don't want to over-saturate. Forces CBO to spread.

For most Indian D2C accounts, splitting CBO by funnel stage is the highest-impact change. Read [CBO vs ABO](https://www.wittelsbach.ai/post/cbo-vs-abo-in-meta-ads-which-budget-strategy-wins-for-d2c-in-2026) for the full framework.

How Wittelsbach AI Detects and Fixes CBO Skew

Bach AI watches CBO spend distribution every hour and flags concentration patterns that don't match revenue contribution. If your CBO is sending 80% of spend to a 1.8x ROAS ad set while a 4.1x ROAS ad set is starving, Bach AI shows the exact minimum budget rule to apply. Run a free Meta Ads audit at [app.wittelsbach.ai](https://app.wittelsbach.ai).

Frequently Asked Questions

Is 80% spend concentration always bad in a CBO?

No — if the dominant ad set has materially higher ROAS and isn't saturating, concentration is correct. The rule of thumb: concentration is healthy when the winning ad set delivers at least 1.5x the ROAS of the average of the others AND audience reach is below 60%. If either condition breaks, intervene.

Will setting minimum ad set budgets reset the CBO learning phase?

Minor edits like setting minimum budgets don't trigger a full learning reset, but they do cause a 24-48 hour redistribution period. Meta will throttle delivery on the previously dominant ad set while the others scale up. Expect short-term ROAS dip, then a more balanced steady state. Don't change minimums every day — let each setting run 7 days.

Should I just use ABO instead?

ABO gives you total control over distribution but loses CBO's automatic budget shifting during the day. For accounts under 100 purchases/week, ABO is often better — you avoid the concentration problem entirely. Above that volume, CBO with minimum budget rails usually outperforms ABO because the algorithm has enough signal to make smart shifts.

How long should I wait before manually rebalancing a CBO?

Minimum 4-5 days. Days 1-3 are learning — CBO is still figuring out which ad set has the cheapest conversions. By day 5 you have enough data to judge whether concentration matches actual performance. Don't intervene earlier; you'll just trigger a learning reset and lose another 3 days.

Does CBO bias get worse with more ad sets?

Yes — significantly. With three ad sets in a CBO, concentration is usually manageable. With six or more, Meta tends to pick one or two winners and starve the rest within 48 hours. For Indian D2C, the sweet spot is 2-4 ad sets per CBO, each representing a clearly distinct audience or creative angle. More than that and you're better off splitting into multiple CBOs.