Solved: Running Multiple Ads in Facebook CBO Campaign

Hi there,

Thanks for reaching out! I had a look at your question and it's a really common one. It's great that you've found a winning interest and are getting consistent sales, that's half the battle won right there. A lot of people get stuck at that point, wondering how to scale up without breaking what's already working.

I'm happy to give you some initial thoughts and guidance on this. The short answer to your question is, yes, you *can* technically add more ads for different products into your existing CBO campaign. But the real answer, from my experience, is that you absolutely shouldn't. It's one of the quickest ways to mess up a good campaign and waste your budget. The solution isn't just about adding more ads; it's about building a proper testing system that runs alongside your successful campaigns. I'll walk you through how I'd approach this.

Why Just Adding More Ads is a Mistake...

I see this all the time. An advertiser finds a winning ad set and a winning creative, and things are going well. They want to make more money, which is the whole point, so the logical next step seems to be to add more products into the mix to "multiply sales". But this is where it goes wrong.

Your current CBO (Campaign Budget Optimization) campaign has one job: to take your $100 daily budget and spend it as efficiently as possible to get you the most sales. Over time, the Facebook algorithm has learned which ad creative and product combination works best within your "action figure interest" audience. It now knows, with a high degree of certainty, that if it shows Ad A to Person X, it's likely to get a sale. So, it funnels the majority of your budget towards that specific ad.

Now, what happens when you introduce a few new ads for different products into that same ad set? The algorithm looks at them and sees they have no performance history. It has no data on whether they will convert. So, faced with a choice between spending money on a proven winner versus an unproven newcomer, it will almost always stick with what it knows. This means your new ads might get a tiny fraction of the budget—maybe a few dollars a day. They will never get enough spend or impressions to gather meaningful data, and you'll conclude they don't work, when in reality they never even had a fighting chance. You're essentialy asking the algorithm to stop doing what it's good at and take a gamble, which it is programmed not to do.

You mentioned "overlap" in your question, and while that's a real concept in terms of audiences, it's not the main problem here. The problem is budget allocation and the disruption of a stable learning phase. You're effectively "polluting" your winning ad set with untested variables. The best-case scenario is your new ads get ignored; the worst-case scenario is you confuse the algorithm enough that it resets its learning, and the performance of your entire campaign drops. If it ain't broke, don't try to fix it by chucking more stuff into it.

I'd say you need a dedicated testing framework...

The professional way to handle this, and how we manage it for all our eCom clients, is to separate your "scaling" activity from your "testing" activity. You should think of your account as having two distinct types of campaigns running at all times:

1. Scaling Campaigns (Your CBO): This is where your proven winners live. Your current CBO campaign is a perfect example. These are the ads, products, and audiences that you know are profitable. You don't mess with these much. You let them run, and you only make small, gradual changes, like slowly increasing the budget by 15-20% every few days if performance is stable.

2. Testing Campaigns (Your New ABO): This is your laboratory. It's where all new ideas go to be proven or disproven. New products, new ad creatives (videos, images), new ad copy, and even new audiences are all tested here in a controlled environment. The key here is to use ABO (Ad Set Budget Optimization), not CBO. By setting the budget at the ad set level, you force Facebook to spend a specific amount on your test, ensuring each variable gets a fair shot. For example, you could set a £15 daily budget for an ad set testing your new products. You *know* that £15 will be spent on that test, unlike in a CBO where it might get £1.

Once a new product or creative proves itself in the Testing Campaign (i.e., it's profitable and gets consistent sales), you can then "graduate" it. You turn it off in the testing campaign and launch it within your main Scaling CBO campaign. This way, you're only ever adding pre-validated winners to your main campaign, which helps you scale reliably without disrupting performance.

This method lets you systematically test new products and creatives to find your next winner, while protecting the revenue being generated by your current one. It takes the guesswork out of the equation and turns scaling into a repeatable process. Here's a visual of what that process looks like:

You'll need to set up your testing campaign properly...

So, let's talk practical steps. Here’s how you'd set up your first test campaign to find a new winning product.

Campaign Setup:

• Objective: Sales. Always optimise for the final action you want someone to take.
• Budgeting: Choose 'Ad Set Budget Optimization' (ABO), which is the default setting. Do NOT turn on 'Campaign Budget Optimization'.
• Naming: Call it something clear, like "[TESTING] - New Products - [Date]". Organisation is going to be incredibly helpful as you scale.

Ad Set Setup:

• Ad Set Name: Name it after the audience you're targeting. In your case, "[TEST] - Action Figure Interest".
• Audience: Use the *exact same* "action figure interest" that is working for you in your CBO. Why? Because you want to isolate variables. You already know the audience is good. The only thing you're changing is the product/ad. If you test a new product on a new audience, and it fails, you won't know if it was the product or the audience that was the problem. Test one thing at a time.
• Budget: Set a daily budget you're comfortable losing. Testing is about buying data. A good starting point is often equal to your average cost per purchase (CPA) or a bit more. If a sale costs you $20, then a $20-30 daily budget for the test ad set is reasonable.

Ad Setup:

• Inside this one ad set, you'll create multiple ads. Let's say you have two new products to test (Product B and Product C). For each product, you should ideally have 2-3 different creatives (e.g., a static image, a short video, a carousel).
• So your ad set would look like this:
  -> Ad 1: Product B - Image 1
  -> Ad 2: Product B - Video 1
  -> Ad 3: Product C - Image 1
  -> Ad 4: Product C - Video 1
• Facebook's algorithm will then distribute your ad set budget across these 4 ads, quickly identifying which creative and product combination gets the most traction with this audience.

Analysing the Results: This is probably teh most important part. You need to have clear rules for what constitutes a "winner" and a "loser". Don't make emotional decisions based on one good or bad day. A simple rule is to let each ad spend at least your target CPA before you even look at it. If your CPA is $20, don't touch anything until an ad has spent $20. If it has no sales by then, it's on thin ice. If it has spent 2x your CPA (so $40) with no sales, it's almost certainly a loser. Turn it off.

If an ad gets a sale or two and is profitable, let it run. You're looking for consistency. An ad that gets you 3-4 sales profitably over a few days is a potential winner. That's the one you can graduate to your CBO campaign. This data-driven approach removes the guesswork and prevents you from wasting money on ads that are never going to work.

To make this easier, here's a little tool to help you set your "kill rules" based on your own numbers.

You probably should think beyond just one interest...

Finding that first winning interest is a massive step, but relying on just one is risky. Audiences can get saturated or "fatigued" over time, meaning performance will eventually decline as the same people see your ads too many times. Your next big lever for growth, alongside testing new products, is testing new audiences.

The testing framework I described above is perfect for this. You can create another ad set within your testing campaign, but instead of using your proven "action figure interest," you try a new one. Crucially, you should test the new audience with your *proven, winning ad creative*. Again, isolating one variable at a time. If the ad works, you know the new audience is good. If it doesn't, you know the audience is the problem.

So, what audiences should you test? There's a clear hierarchy that usually works best for eCommerce stores.

1. More Specific Detailed Targeting: Your "action figure" interest is a good start, but it's quite broad. Think deeper about your ideal customer.

• What specific brands do they follow? (e.g., Hot Toys, NECA, McFarlane Toys, Sideshow Collectibles)
• What franchises are they obsessed with? (e.g., Marvel Legends, Star Wars: The Black Series, DC Multiverse)
• What magazines or websites do they read? (e.g., ToyFare magazine)
• Who are your direct competitors? You can sometimes target people who like their Facebook pages.

2. Lookalike Audiences: This is where things get really powerful, and this is the main way we scale accounts. A lookalike audience is when you ask Facebook to find new people who are extremely similar to an existing group of people (your "source audience"). You can create lookalikes from various sources, but they are not all created equal. You should prioritise them based on how valuable the source audience is.

• Highest Quality: Lookalike of your past purchasers. These are people who share characteristics with those who have already bought from you. This is almost always the best performing cold audience. You'll need at least 100 purchases tracked by your pixel to create this.
• Good Quality: Lookalike of people who "Initiated Checkout" or "Added to Cart". These people showed strong intent but didn't finish.
• Lower Quality: Lookalike of all your website visitors, or people who have engaged with your Facebook/Instagram page. These are much broader.

3. Retargeting Audiences: These are people who have already interacted with your business but haven't bought yet. It's often the most profitable part of any ad account. You should have a separate, always-on campaign (can be CBO or ABO) that just targets these people.

• Bottom-of-Funnel (BoFu): People who added a product to their cart in the last 7-14 days but didn't buy. Show them an ad with that product, maybe with a small discount or a reminder about free shipping to get them over the line.
• Middle-of-Funnel (MoFu): People who viewed products or visited your website in the last 30 days. Show them your best-selling products or a brand ad.

This tiered approach gives you a nearly endless supply of new audiences to test and scale with. Here is a visual way to think about the priority:

This is the main advice I have for you:

To put it all together, here is a clear action plan. Following this structure is the difference between randomly gambling on new products and building a scalable, predictable sales machine for your store.

This structured approach is how professional media buyers and agencies scale eCommerce brands from $100/day to thousands per day. It takes a bit more setup than just dumping ads into one campaign, but it's infinitely more reliable and will save you a lot of money and headaches in the long run.

Managing this process—juggling testing and scaling campaigns, analysing the data correctly, creating lookalikes, and constantly developing new creative—can become a full-time job. It requires discipline and a deep understanding of how the ad platforms work. This is often where businesses decide to bring in an expert. An experienced hand can speed up the process, avoid common pitfalls, and bring insights from hundreds of other ad accounts to help you grow faster.

If you'd like to chat through your account in more detail and get a second pair of eyes on your strategy, we offer a free, no-obligation initial consultation. We could walk through your setup together and identify some specific opportunities for you. Either way, I hope this detailed breakdown gives you a solid plan to work with.

Regards,

Team @ Lukas Holschuh