Solved: Avoid Auction Overlap when Duplicating Ad Sets

Hi there,

Thanks for reaching out. That's a great question about scaling campaigns by duplicating ad sets. It's a common strategy people discuss, and I'd be happy to share some insights from my experience on what works best.

It's a really common tactic people talk about, this idea of duplicating a winning ad set to scale up without resetting the learning phase. On the surface, it sounds like a clever little hack to trick the system. But honestly, it's one of the biggest myths in Facebook advertising, and it's probably costing you more money than it's making. The real path to scaling is a bit more methodical, but it’s far more reliable and profitable in the long run. Let's get into why that is and what you should be doing instead.

The Big Problem with Duplicating Ad Sets

I get why people do it. You've found an ad set that's working, the conversions are coming in at a good price, and you're scared to touch it. The fear is that if you increase the budget, you'll shock the system and send it back into the dreaded "learning phase," killing your performance. So, duplicating it seems like a safe way to tell Facebook, "Hey, just do more of that, please."

Unfortunately, that's not how the auction works. When you duplicate an ad set—even with broad, open targeting—you are creating an almost identical competitor to your original ad set. You now have two of your own ad sets trying to show ads to the same pool of people at the same time. This is called Auction Overlap.

Think of it like this: you've sent two salespeople from your own company to bid for the same contract. They end up just bidding against each other, driving the price up for you, and confusing the client. That's exactly what you're doing inside your own ad account. The algorithm sees two ad sets from you trying to reach user Jane Doe, and it forces them to compete. The result? Your CPMs (Cost Per 1,000 Impressions) go up, and your overall CPA (Cost Per Acquisition) increases. You're literally paying more to reach the same people because you're fighting yourself.

The other major issue is data fragmentation. The Facebook algorithm learns and optimises based on the data it collects. It needs a steady, concentrated stream of conversion events to get really good at finding you customers. When you duplicate an ad set, you split that data stream in two. Instead of one ad set getting 50 conversions in a week and exiting the learning phase with a strong signal, you now have two ad sets each getting 25 conversions. Neither of them learns as fast or as well. You're diluting the very data that the algorithm needs to succeed.

I'd say you should focus on 'Vertical Scaling' first

So, what's the correct approach? It’s called vertical scaling. It’s simple, effective, and works *with* the algorithm, not against it.

When you have a winning ad set, you scale it by directly increasing its daily or lifetime budget. The key is to do it gradually. A common rule of thumb that works well for us is to increase the budget by no more than 20-30% every 2 to 3 days. So if you're spending £50 a day, you can bump it to £60-£65. Let it run for a couple of days, check the performance, and if it's still stable and profitable, you increase it again.

This gradual increase gives the algorithm time to adjust. It can expand its reach and find more customers for you without the sudden shock that a massive budget change would cause. Yes, sometimes a small budget increase might temporarily push you back into the learning phase, but because you're feeding it data within the *same* ad set, it will exit that phase much faster and with better performance than two fragmented, competing ad sets ever would.

You scale an ad set vertically until you hit a point of diminishing returns—where increasing the budget further causes your CPA to rise beyond a profitable level. This is normal and expected. It means you've saturated the most responsive part of that audience at that budget level. And that's when you move to the next stage.

You'll need to know your numbers to scale confidently

Before we even talk about the next stage, there’s a piece of homework that is absolutely essential. You can't scale anything properly if you don't know what a customer is actually worth to you. The question shouldn't be "How low can I get my CPA?", but rather "How high a CPA can I afford to acquire a valuable customer?"

This is where understanding Customer Lifetime Value (LTV) comes in. It's the total profit you expect to make from a single customer over the entire duration of your relationship. I remember one SaaS client we worked with who were hesitant to spend more than £50 per trial. After we worked through their LTV, we found each paying customer was worth over £8,000 to them. Suddenly, paying £150 for a trial that had a 1 in 10 chance of converting looked like an incredible bargain. This mindset shift is what unlocks real, aggressive scaling.

The formula is pretty straightforward:

LTV = (Average Revenue Per Customer Per Month * Gross Margin %) / Monthly Churn Rate %

To make this tangible for you, I've built a small calculator. Play around with the sliders to see how small changes in your business metrics can dramatically change what you can afford to spend on ads.

We'll need to look at 'Horizontal Scaling' for real growth

Once you've maxed out vertical scaling on your winning ad set, it's time for horizontal scaling. This is where you find *new* pockets of customers. Critically, this does NOT mean duplicating your winning ad set. It means launching new, different ad sets to test new variables.

There are two main levers you can pull here:

1. New Audiences: This is where you hunt for more people who look like your best customers. Based on what usually works best, I'd test these in order:

• High-Value Lookalikes: Create lookalike audiences from your best customer lists. Not just all customers, but lookalikes of people who have purchased multiple times, or your highest LTV customers. Start with a 1% lookalike and expand from there.
• Lower-Funnel Lookalikes: If you don't have enough purchase data, create lookalikes from people who have Initiated Checkout or Added to Cart. These are still high-intent signals.
• New Interest Stacks: Brainstorm new categories of interests your customers might have. If you sell outdoor gear, you might have already tested hiking and camping interests. Now try targeting brands of outdoor clothing, magazines about nature, or even related activities like rock climbing.

2. New Creatives: This is often the most powerful lever for scaling. Your audience gets tired of seeing the same ad over and over (ad fatigue). Introducing fresh creatives can revitalise a campaign and unlock new performance. We worked on a campaign for a medical job matching SaaS where we reduced the CPA from £100 to just £7, and a massive part of that was a relentless creative testing process. We tried everything: static images, short videos, carousel ads, and even user-generated style content which performed amazingly well for them. Don't assume the creative that worked for one audience will work for another. Test different angles, messaging, and formats constantly.

You probably should be using Campaign Budget Optimisation (CBO)

This leads to the final peice of the puzzle: Campaign Budget Optimisation, or CBO. This is a setting at the campaign level where you set one central budget, and you let Facebook's algorithm automatically distribute that budget across your ad sets in real-time. It will give more money to the ad sets that are getting you the best results.

This is the perfect tool for horizontal scaling. You can put your proven, winning ad set in a CBO campaign alongside two or three new test ad sets (with new audiences or new creatives). The algorithm will continue to spend the majority of the budget on your proven winner, ensuring your results stay stable, while siphoning off a small portion of the budget to see if your new tests can find traction. If a new test ad set starts performing well, the algorithm will automatically give it more budget. If it performs poorly, it'll get very little spend. This removes the guesswork and manual budget management from the testing process. It’s the smart way to scale and the opposite of the manual, inefficient duplication method.

This is the main advice I have for you:

Forget the shortcuts and hacks. The path to sustainably scaling your Facebook ads is built on a solid, repeatable process. I've summarised the strategy I've outlined into a simple table for you below. This is the framework we use for our clients, from small eCommerce stores to large B2B software companies, and it works.

Scaling ads isn't about finding a secret button to press; it's about having a solid testing framework and understanding the machine you're operating. Trying to outsmart the algorithm with tactics like duplication usually backfires. The real secret is to give it what it needs: concentrated data, clear objectives, and enough room to do its job.

This stuff can get complex, and managing a proper scaling structure takes time and expertise. It involves knowing which metrics to watch, when to kill a failing test, and how to interpret the data to inform your next move. If you ever feel like you're spending too much time guessing and would rather have an expert team handle this kind of strategic work for you, we offer a completely free, no-obligation initial consultation where we can look through your account and give you some tailored advice.

Hope this helps!

Regards,

Team @ Lukas Holschuh