How to Structure a CBO Campaign on Meta for DTC Brands

A CBO (Campaign Budget Optimization) campaign on Meta lets the algorithm distribute your budget across ad sets in real time, concentrating spend on whichever audiences and creatives are converting best rather than forcing fixed budgets per ad set. Last updated: February 2026

Table of Contents

What Is CBO and How Does It Work?

Campaign Budget Optimization (CBO) moves budget control from the ad set level to the campaign level. Instead of telling Meta "spend $100 on audience A and $100 on audience B," you tell it "here is $200, find the best opportunities across all my ad sets."

Meta's algorithm then monitors performance in real time and shifts budget dynamically. If Audience A is converting at $18 CPA and Audience B at $45 CPA, Meta will push 80-90% of budget to Audience A. You do not have to manually reallocate, it happens automatically and continuously.

This approach works because Meta has significantly more data about when and where to show ads than any human operator. The algorithm considers thousands of signals per impression: user behavior history, time of day, device type, recent purchase patterns, and more.

The tradeoff is less granular control. You cannot guarantee a specific ad set gets specific spend, which matters if you are trying to test audiences fairly. That is why the CBO vs ABO decision requires strategic thinking.

CBO vs ABO: Which Should DTC Brands Use?

ABO (Ad Set Budget Optimization): You set a fixed budget per ad set. Good for controlled testing where you want equal spend across variables. Gives you visibility into performance by audience but requires more manual management. CBO: Budget set at campaign level. Good for performance scaling once you know what works. Less control over individual ad set spend but better overall efficiency when the algorithm has enough data. The MHI Media recommendation: Use ABO for creative testing (where you need equal data per variation) and CBO for scaling (where you want Meta to find the best opportunities). Many brands use both simultaneously: ABO for their testing campaign, CBO for their scaling campaigns.

For DTC brands spending $5K-$50K per month, CBO with Advantage+ audience settings tends to outperform ABO for prospecting by 15-25% on CPA. Below $5K/month, the algorithm does not have enough data to optimize effectively, so ABO often performs more predictably.

The Recommended CBO Structure for DTC

The cleanest structure for most DTC brands using CBO is:

Campaign 1: Prospecting CBO Campaign 2: Retargeting ABO (keep this separate) The reason to keep retargeting separate is control. In a CBO with mixed cold and warm audiences, Meta will often dump budget into retargeting (which converts more easily) and starve prospecting. Separating them lets you control the prospecting-to-retargeting ratio intentionally.

How to Set Up a CBO Campaign

Step 1: Create the Campaign

    • In Meta Ads Manager, click "Create"
    • Select "Sales" as your objective
    • Name your campaign clearly (e.g., "Prospecting CBO - US - Feb 2026")
    • Toggle on "Campaign Budget Optimization"
    • Set your daily or lifetime budget
    • Select "Lowest cost" bid strategy for most DTC brands

Step 2: Set Campaign Budget

Start with a budget that allows at least 10-15 conversions per day across the campaign. The algorithm needs purchase events to optimize effectively. Below 50 purchases per week at the campaign level, CBO performance is inconsistent.

Rule of thumb: if your target CPA is $30, set a minimum campaign budget of $300-$450/day to give Meta enough room to learn.

Step 3: Configure Spend Limits (Optional)

CBO allows you to set minimum and maximum spend limits per ad set. Use these sparingly. Maximum limits help ensure test ad sets get some exposure. Minimum limits prevent the algorithm from completely ignoring an ad set you want data on.

Do not lock down spend limits so tightly that you eliminate the CBO benefit. If you constrain every ad set heavily, you might as well use ABO.

Ad Set Strategy Within CBO

Option A: Broad Targeting CBO

Create 3-4 ad sets with minimal audience restrictions:

This approach trusts Meta's algorithm to find buyers. For brands with strong pixel data (500+ purchase events), broad targeting in CBO often outperforms interest-based targeting.

Option B: Interest-Based CBO

Create ad sets targeting different interest clusters:

CBO will allocate to whichever interest group converts best. Over time, you learn which interests your buyers cluster in.

Option C: Advantage+ Audience CBO

Meta's recommended approach for 2026: use Advantage+ audience at the ad set level within CBO. This gives Meta maximum flexibility to find buyers anywhere on the platform. MHI Media has seen this outperform interest-targeting CBO by 20-30% CPA for brands with 300+ monthly purchases.

Creative Strategy for CBO Campaigns

Creative is the primary lever in CBO campaigns. Since Meta controls audience and budget, you control creative and offer. Load each ad set with 3-6 strong creatives and let CBO find which perform best with which audience segments.

Creative diversity within a CBO campaign: Avoid duplicating identical creatives across multiple ad sets within the same CBO campaign. Meta will either consolidate delivery (showing only one) or create competition between your own ads.

Refresh creatives every 2-3 weeks as performance drops. In a well-run CBO campaign, creative fatigue is the most common reason for deteriorating performance, not audience saturation.

Scaling CBO Campaigns

Once a CBO campaign is profitable and stable for 7+ days, you can scale budget. The standard approach:

Conservative scaling: Increase budget by 15-20% every 3-4 days. Small increments avoid disrupting the learning phase. Aggressive scaling: Duplicate the campaign at a higher budget (2x-3x) and run it alongside the original. If the duplicate performs, kill the original and continue scaling the new one. Horizontal scaling: Add new ad sets with fresh audiences or fresh creatives rather than increasing budget. This gives the algorithm new optimization opportunities.

Never reduce budget more than 20% in a single change. Cuts trigger a new learning phase which can destabilize performance for 3-7 days.

FAQ

How much budget do I need to start a CBO campaign? The minimum practical budget depends on your CPA. You need enough to generate at least 7-10 conversions per week at the campaign level for the algorithm to learn. If your CPA is $50, you need at least $350-$500/week, or $50-$70/day. Should I use daily budget or lifetime budget in CBO? Daily budget for ongoing prospecting campaigns. Lifetime budget for time-limited promotions (Black Friday, product launches) where you want Meta to optimize toward your highest-performing windows within the campaign flight. Can I mix prospecting and retargeting in one CBO campaign? Technically yes, but it is not recommended. Meta will typically over-invest in retargeting audiences because they convert more easily, starving your prospecting ad sets. Keep them separate campaigns. How do I know if my CBO is in learning phase? The "Learning" tag appears next to your campaign or ad set name in Ads Manager. The learning phase typically lasts 1-7 days and requires 50 optimization events to complete. During this phase, performance is unstable. Avoid making changes. What bid strategy should I use for CBO? Start with "Lowest cost" (no bid cap). This gives Meta maximum flexibility. Once you have stable performance data, you can experiment with "Cost cap" if you need to control CPA tightly, but cost caps often restrict delivery significantly. How many creatives should I put in a CBO ad set? 3-6 is the sweet spot. Too few and Meta has nothing to test. Too many and individual creatives get too little spend to generate statistically meaningful data. Refresh the bottom performers every 2-3 weeks. Does CBO work for retargeting? CBO can work for retargeting campaigns but ABO often provides more control over small, warm audiences. If your retargeting audiences are small (under 10,000 people), ABO lets you ensure each tier gets consistent exposure.