Hi there,
Thanks for reaching out! It's good to hear you've got a CBO setup that's performing well, especially with how volatile things can be on the platform. Hitting your target CPA consistently is a great place to be. Your question about scaling is a very common one, and it's smart to be cautious so you don't disrupt what's already working. I'm happy to give you some of my initial thoughts and guidance on how I'd approach this based on my experience scaling these types of campaigns.
TLDR;
- Duplicating CBO campaigns with the same audiences is a bad idea. It creates auction overlap, which will almost certainly increase your CPMs and cause delivery problems.
- Your cost cap is the main reason you're underspending. It's a tool for stability, not aggressive scaling. You're telling the algorithm to stop spending if it can't find cheap results, which limits your reach.
- The best way to scale vertically is by gradually increasing the budget on your existing, winning CBO. A slow and steady approach of around 20% every 2-3 days is usually the safest bet.
- True scaling involves more than just increasing spend. You need to scale horizontally by constantly testing new creatives and expanding into new, untapped audiences (new interests, lookalikes etc).
- This letter includes an interactive calculator to help you plan your budget increases and a flowchart to decide on the right bidding strategy for your goals.
Don't Duplicate Your CBOs - You're Bidding Against Yourself
Let's get straight to your main question: "Will multiple CBOs with the same creative/audience combinations but different cost caps compete against each other in the auction?"
The short answer is yes, absolutely. And it’s a surefire way to ruin a perfectly good campaign.
Think of the Meta ads auction like a real-world auction house. Your audience – the specific group of people you're targeting – is the item up for sale. When you run one campaign, you're a single bidder in the room, competing against other advertisers. When you duplicate that exact campaign, you've essentially sent a clone of yourself into the same auction to bid on the exact same item. What happens when two people are desperate for the same thing? They drive the price up for each other.
In the world of Meta ads, this is called auction overlap. Your two (or more) campaigns start bidding against each other for the same limited ad inventory in front of the same users. The algorithm doesn't see these as one cohesive strategy; it sees them as separate competitors. The immediate result is that your CPMs (Cost Per Thousand Impressions) will almost certainly rise. You end up paying more just to reach the same people you were already reaching efficiently. This directly impacts your CPA, pushing it up and undoing all the good work you've done. You also risk really erratic delivery as the algorithm struggles to decide which of your identical campaigns should win the bid for any given user. It's a common mistake, but a costly one.
I've seen accounts where people have dozens of campaigns targeting slight variations of the same audience, and they wonder why their costs are through the roof and performance is unpredictable. It's because they've created their own internal competition. You've found a winning combination; the goal now is to feed that winner, not to create rivals that will cannibalise its performance.
The Real Reason You're Underspending: Your Cost Cap
You mentioned that you're consistently hitting your target CPA but only spending 30-40% of your budget. This isn't a mystery; it's the direct result of using a cost cap. A cost cap is an instruction to the algorithm that says, "Get me conversions, but *only* if you can get them at or below this specific price (£30 CAD in your case). If you can't, don't bother spending the money."
This is a great tool for maintaining stability and profitability. It puts a ceiling on your CPA, which is fantastic when you've found a sweet spot. However, it's fundamentally at odds with the goal of aggressive scaling. When you scale, you need to reach more people. The "low-hanging fruit"—the users most likely to convert at the lowest cost—get exhausted first. To reach more users and spend a larger budget, the algorithm inevitably has to go after people who are slightly less likely to convert, which means the average CPA will naturally start to creep up.
Your cost cap is preventing this from happening. It’s acting like a governor on an engine. The algorithm finds all the cheap conversions it can, hits your cap, and then simply stops spending because you've forbidden it from pursuing the slightly more expensive (but still potentially profitable) conversions needed to exhaust the full budget. You've essentially told it that you prioritise a strict CPA over spending your full daily budget. And the algorithm, being the obedient machine it is, is doing exactly what you asked.
So, the idea of setting up different CBOs with slightly different cost caps (£28, £32) is flawed for two reasons. First, as we've discussed, it creates auction overlap. Second, it doesn't solve the fundamental problem. The £28 cap campaign will spend even less, and the £32 campaign might spend a bit more, but you're still putting artificial limits on the algorithm's ability to scale effectively. To truly unlock a higher spend, you need to reconsider your bidding strategy altogether.
Best for profitable but limited delivery. The algorithm will not spend if it cannot meet your exact CPA target.
Best for aggressive growth. The algorithm will try to spend the full budget to get the most conversions possible, even if CPA fluctuates.
How to Scale Vertically the Right Way: Gradual Budget Increases
So, if duplicating the CBO is out, what's the correct approach? The simplest and most effective method for scaling a winning campaign vertically (i.e., increasing spend on what's already working) is to gradually increase the budget of your single, successful CBO campaign.
The key word here is *gradually*. The Meta algorithm is powerful but sensitive. A sudden, massive increase in budget can shock the system, potentially forcing it back into the learning phase and causing performance to become unstable. The algorithm has learned how to find you conversions efficiently at £500/day. You need to give it time to learn how to do the same at £600/day, then £720/day, and so on.
A good rule of thumb that we follow is to increase the budget by no more than 20% every 2-3 days. This gives the algorithm enough time to adjust, find new pockets of users, and stabilise performance at the new spending level before you increase it again. If you raise the budget and see the CPA spike and stay high for more than 48 hours, you might need to pull it back down to the previous level and let it settle before trying again. It's a bit of a dance, and it requires patience and close monitoring. Don't make changes daily. Let the algorithm do its work.
This method is far more reliable than duplication. You're working *with* the algorithm's learning, not against it. You're keeping all your performance data and social proof (likes, comments, shares on your ads) consolidated within one campaign structure, which helps maintain efficiency.
(Apply this increase and monitor for 2-3 days before repeating)
You'll Hit a Ceiling: The Need for Horizontal Scaling
Increasing the budget is only one part of the equation. Eventually, even with gradual increases, you will hit a point of diminishing returns. Your CPA will start to rise consistently because you've saturated the most responsive parts of your current audiences. This is totally normal. No audience is infinitely scalable. This is when you need to shift your focus from purely vertical scaling to *horizontal scaling*.
Horizontal scaling means expanding your reach by finding *new* audiences and *new* winning creatives to run to them. Your current setup with 5 ad sets is a great start, but there are likely dozens more audience pockets you could test. This is the real work of a media buyer and where true, sustainable growth comes from. It's about building a robust testing system to constantly find new winners to which you can allocate budget.
We usually structure this using a funnel-based approach, breaking audiences down into Top-of-Funnel (ToFu), Middle-of-Funnel (MoFu), and Bottom-of-Funnel (BoFu).
Top of Funnel (ToFu): This is your cold traffic. People who have never heard of you. Your goal here is broad prospecting.
- Detailed Targeting: Interests, behaviours, and demographics. This is where you started, but have you exhausted all possible angles? Think about shoulder niches, competitor brands, relevant software your customers use, influencers they follow. The goal is to build dozens of themed ad sets to test.
- Lookalike Audiences: This is huge. Once you have enough data, you should be building lookalikes from your highest-intent custom audiences. Don't just make a lookalike of "all website visitors." Make lookalikes of your *purchasers*, of people who initiated checkout, or even of your highest LTV customers if you can upload that data. These are far more potent. Start with a 1% lookalike in your target countries and expand to 2%, 3-5% etc., as you scale.
Middle of Funnel (MoFu): These are people who have shown some interest but haven't taken a key action yet. They need another nudge.
- Retargeting Engagers: People who have watched a certain percentage of your video ads (e.g., 50% or more), or engaged with your Facebook or Instagram page. They know you, but haven't clicked through.
- Retargeting Website Visitors: Anyone who has visited your site but didn't make it to the checkout or purchase stage. You can segment this further (e.g., viewed specific product pages).
Bottom of Funnel (BoFu): Your hottest audience. These people are on the verge of converting.
- Retargeting Cart Abandoners: People who added an item to their cart or initiated checkout but didn't complete the purchase. These often have the highest ROAS of any audience. You can hit them with a specific offer or just a reminder.
The strategy is to systematically test new ToFu audiences within your CBO. Let Meta's algorithm find the winners and allocate the budget accordingly. As you find new winning interests or lookalikes, you can keep feeding the campaign, allowing it to spend more money efficiently because you're giving it fresh, fertile ground to explore. At the same time, you need a relentless creative testing process. Your ~10 ads per ad set is a good number, but are you swapping out the losing ads for fresh ones every week or two? Ad fatigue is real, and new creative is the fuel for any scaled campaign.
I've managed several campaigns for software clients where a structured approach to scaling was key to breaking through performance plateaus. For one software client running ads on Meta, we helped them achieve 5,082 software trials at a $7 cost per trial. For another, focused on B2B registrations, we scaled their Meta Ads campaign to achieve 4,622 registrations at only $2.38 each. It's a proven process, but it requires discipline and a structured approach to testing.
(Interests, Lookalikes)
(Engagers, Site Visitors)
(Cart Abandoners)
My Recommendations for You
To put it all together, trying to scale by duplicating your CBO is a trap that will likely lead to higher costs and frustration. The path forward is more methodical. It involves careful budget management on your winning campaign combined with a structured plan for expansion.
This is the main advice I have for you:
| Step | Rationale | Action to Take |
|---|---|---|
| 1. Consolidate | Avoid auction overlap and consolidate all learnings and social proof into a single, powerful campaign. | Stick with your single winning CBO campaign. Do NOT duplicate it. |
| 2. Scale Vertically | Safely increase spend without shocking the algorithm, allowing it to maintain performance as budget grows. | Increase the budget of your existing CBO by 15-20% every 2-3 days. Monitor CPA closely after each increase. |
| 3. Re-evaluate Bid Strategy | Your cost cap is limiting spend. To truly scale, you may need to give the algorithm more freedom. | Once you want to scale more agressively, consider testing removing the cost cap and switching to the 'Highest Volume' bid strategy to prioritise spending the full budget. Expect some initial CPA fluctuation. |
| 4. Scale Horizontally | Your current audiences will eventually saturate. You need to constantly find new people to target to sustain growth. | Start systematically testing new ToFu audiences (new interests, new lookalikes from high-intent actions like purchases) as new ad sets within your CBO. |
| 5. Refresh Creative | Ad fatigue will kill performance over time. Fresh creative is essential for keeping scaled campaigns effective. | Establish a process to introduce 2-3 new ad creatives into your ad sets every 1-2 weeks, turning off the worst performers. |
Scaling a campaign from promising to genuinely significant is often where many advertisers get stuck. The strategies that get you to £500/day aren't always the ones that get you to £5,000/day. It shifts from finding one winner to building a machine that reliably produces new winning audiences and creatives. This requires a huge amount of time, dicipline, and experience to know which levers to pull and when.
If you're looking for a partner to help manage this process and take your campaigns to the next level, we offer a free, no-obligation consultation. We can take a proper look at your ad account together, analyse your current setup in detail, and build out a more robust scaling strategy tailored specifically to your business. It's a great way to get an expert second opinion and see if we'd be a good fit to work together.
Regards,
Team @ Lukas Holschuh
