Solved: Advantage+ Serving Only Old Customers

Hi there,

Thanks for reaching out!

Happy to give you some initial thoughts on this. It's a classic situation and something I've seen a fair few times. Sounds like you've hit the Adv+ ceiling, where the algorithm gets a bit lazy and just defaults to the easiest possible conversions. The good news is, it's fixable, but it'll take a bit of a restructure and a shift in how you think about your campaigns.

We'll need to look at why your Advantage+ campaign turned into a glorified retargeting machine...

Honestly, what you're describing isn't that surprising. When you give an algorithm like Meta's a broad audience that includes your entire customer list and tell it to find conversions, it's gonna take the path of least resistance. It's like telling a delivery driver to find the closest house on their route—they're not gonna drive to the next town over if there's a drop-off next door.

Your old customers are the lowest-hanging fruit. They know you, they trust you, and they're infinitely more likely to buy again than a complete stranger. So, the algorithm targets them, gets a conversion, and reports back "job done". Forgetting to exclude your customer list was the critical mistake. You essentially gave it a big pot of gold and told it to look for treasure, so it never bothered leaving the pot.

This is the uncomfortable truth about these automated campaign types. When you set an objective like 'Sales' without tight controls, you are giving the algorithm a very specific command: "Find me the largest number of sales for the lowest possible price." The algorithm, being efficient but not strategic, seeks out the users who are most likely to convert cheaply. In your case, that's the people who have already bought from you. You were actively paying the world's most powerful advertising machine to find you the people you'd already found. This is why you're seeing your ROAS drop now; you've likely saturated that small group of repeat buyers, and the algorithm is struggling to find more easy wins.

The initial 5x ROAS was probably a sugar rush from hitting all those loyal customers at once. Now that's worn off, you're left with a campaign that has no idea how to actually find *new* people. It's a common trap, and getting out of it means taking back some control from the machine.

I'd say you need to completely rethink your account structure...

Your attempt to fix this with a new campaign that had exclusions was the right idea, but it's likely it failed because it was competing against a campaign with months of data and a pixel seasoned on your best customers. The algorithm favoured the old campaign because it had a proven track record, even if that track record was built on sand. A single campaign just can't do everything. You need a proper funnel structure.

I always structure accounts into three core campaign types based on the user's journey. Top of Funnel (ToFu), Middle of Funnel (MoFu), and Bottom of Funnel (BoFu).

• ToFu (Prospecting): This is your new customer engine. Its only job is to find people who have never heard of you. Here, you MUST exclude all past purchasers, all website visitors from the last 180 days, and all social media engagers. This is cold traffic. Its performance will look worse on paper (lower ROAS), and that's okay. We measure this on Cost Per Acquisition (CPA), not ROAS.
• MoFu (Consideration/Warm Retargeting): This is for people who've shown some interest but haven't bought. Think website visitors, video viewers, people who've engaged with your Instagram page. You're reminding them you exist and nudging them towards a purchase. Here you exclude past purchasers.
• BoFu (Conversion/Hot Retargeting): This is for people who are right on the edge of buying. They've added to cart, initiated checkout, or are on your customer list for repeat purchases. This is where your high ROAS comes from. This is basically what your Adv+ campaign has become.

By splitting it up, you tell Meta exactly who to go after in each campaign and you can judge performance properly. You can't expect a cold prospecting campaign to get a 5x ROAS overnight. But you *can* expect it to bring in new customers at an acceptable cost, who you then nurture into profitable repeat buyers through your MoFu and BoFu campaigns.

You probably should focus 80% of your effort on prospecting...

This is where your growth comes from. Retargeting is just harvesting the demand you've already created; prospecting is creating new demand. Based on my experience with eCommerce campaigns—I've run many that have hit over 600% ROAS for things like apparel and cleaning products—getting the prospecting right is the whole game.

So, how do you do it properly? You test audiences methodically.

For a new prospecting campaign, you'd want to prioritise your audiences like this:

1. High-Intent Lookalike Audiences: This is your best bet. You need to give Meta a strong signal of who your ideal customer is. Don't just upload your whole customer list. You want a lookalike of your *best* customers. If you can, segment your customer list by:
   • Highest lifetime value customers
   • Repeat purchasers (bought 3+ times)
   • Recent purchasers (last 30-60 days)
   Start with a 1% lookalike of your highest value purchasers in your target countries. This is usually gold. As you get more data, you can test lookalikes of people who initiated checkout or added to cart. You need at least 100 people for a custom audience, but honestly, you want more like 1,000 for a really solid lookalike.


2. Detailed Targeting (Interests & Behaviours): This requires a bit more thought than just typing in broad interests. You need to think about the *pain* your product solves. Who feels that pain? What podcasts do they listen to? What brands do they follow? What software do they use? If you sell high-end kitchenware, don't just target 'Cooking'. Target followers of 'Bon Appétit' magazine, people interested in 'Le Creuset', or users of recipe apps like 'Yummly'. You want to find niche interests that have a high concentration of your ideal customer. If an interest is too broad, you just end up paying to reach people who aren't a good fit.

You gotta make sure to test these in seperate ad sets so you know what's working. Don't lump a lookalike and an interest audience together. Give each ad set a decent budget and let it run for a few days until it's out of the learning phase before you judge it. Turn off what doesn't work and scale what does. It's a process of constant iteration.

You'll need to understand the numbers that actually matter...

Here's the bit that most people get wrong. You said your new prospecting campaign had "even worse results". I'm guessing that means it had a low or negative ROAS. That's totally normal and expected for a prospecting campaign. Trying to judge a ToFu campaign by its immediate ROAS is like judging a farmer by how many apples they have the day after they plant a seed. It's the wrong metric.

The real question isn't "How low can my CPL go?" but "How high a CPL can I afford to acquire a truly great customer?" The answer is your Lifetime Value (LTV). Once you know what a customer is worth to you over their entire relationship with your business, you can make much smarter decisions about how much you're willing to pay to get them in the door.

Let's do the maths. You need three numbers:

1. Average Revenue Per Account (ARPA): What do you make per customer, per month (or year)?
2. Gross Margin %: Your profit margin on that revenue.
3. Monthly Churn Rate: The percentage of customers you lose each month.

The calculation is: LTV = (ARPA * Gross Margin %) / Monthly Churn Rate

For an eCommerce business, you might calculate it slightly differently: (Average Order Value * Purchase Frequency * Gross Margin) / Churn Rate. But the principle is the same. Once you have your LTV, you can determine your target Customer Acquisition Cost (CAC). A healthy ratio is usually 3:1 LTV to CAC. So if your LTV is £300, you can afford to spend up to £100 to acquire a new customer and still have a very healthy business.

Suddenly, a prospecting campaign that's getting customers for £75 each doesn't look like a failure; it looks like a bargain. This is the maths that unlocks aggressive, intelligent growth. Use the calculator below to get a feel for your own numbers.

We'll need to look at your offer and creative...

Even with the perfect structure and targeting, a campaign can fail if the message is wrong. Your ad creative and offer for a cold audience (ToFu) has to be different from your retargeting (BoFu) ads.

For prospecting, you can't assume any prior knowledge. You need to grab their attention and speak directly to a problem they have. I often use the Before-After-Bridge framework.

• Before: Describe their current world. What's their frustration? "Tired of fast fashion that falls apart after two washes?"
• After: Paint a picture of the better world your product offers. "Imagine a wardrobe of timeless, quality pieces you'll love for years."
• Bridge: Introduce your product as the way to get there. "That's what we do. Discover our collection of ethically-made essentials."

This is way more effective than just showing a product photo and a price. You're selling a transformation, not just an item. For your lead magnet, the same applies. Is it solving an urgent, specific problem for your ideal customer? Or is it a generic "10% off your first order"? The more value you can provide upfront, the better your prospecting campaigns will perform.

You also need to be testing creative relentlessly. What works for your warm audience won't necessarily work for a cold one. Test different images, videos, headlines, and copy. I've seen some SaaS clients get incredible results with simple User-Generated Content (UGC) style videos because they feel more authentic and less like an ad. The same principle applies to eCommerce. A video of a real customer loving your product can often outperform a slick, polished studio ad.

This is the main advice I have for you:

To pull this all together, here is a table outlining the exact structure I'd recommend you implement. It's about moving from one "do it all" campaign to a strategic, multi-layered approach that builds a predictable pipeline of new customers.

This is a lot to take in, I know. Moving from a single automated campaign to a proper structured account is a big step, but it's the only way to build a sustainable and scalable advertising system. You've proven there's demand with your initial success; now it's time to build the machine that can capture that demand from new audiences consistently.

This kind of strategic overhaul is precisely what we do for our clients. We take accounts that have hit a wall and rebuild them for growth. It takes expertise to get the audiences right, calculate the LTV correctly, and write copy that converts cold traffic.

If you'd like to go through your account and map this out in more detail, I'd be happy to offer you a free 20-minute strategy session. We can look at your specific numbers and figure out the best audiences to test first.

Hope this helps!

Regards,

Team @ Lukas Holschuh