Solved: Maintaining 1 Adset/Ad Post-Meta Update

Hi there,

Thanks for reaching out! I had a look at your question about campaign structure, and it's a really common thing for people to get tripped up on. You're right, the way Meta wants you to run ads now is a bit different from a few years ago, and that old 1:1:1 structure, while it looks neat, is probably costing you more than you think.

I'm happy to give you some initial thoughts and break down how we approach this. The short answer is that moving to multiple ads in one adset isn't just about tidiness; it's about fundamentally working *with* Meta's algorithm instead of fighting against it. Let me explain a bit more about what that means in practice.

We'll need to look at why your 'clean' setup is a false economy...

I get the appeal of the 1:1:1 structure. It looks organised, you know exactly where every pound is going, and it feels like you have total control. But here's the brutally honest truth: in today's advertising world, that feeling of control is an illusion, and it's a very expensive one. You're essentialy telling the most powerful advertising machine on the planet that you know better than it does. You're forcing it to work with one hand tied behind its back.

Every time you create an adset, you're creating a separate little learning environment for the Meta algorithm. For that adset to become effective, it needs to exit the 'learning phase'. To do that, it needs to get a certain number of conversions (usually around 50) in a short period (about a week). When you split your budget across dozens of adsets, each with only one ad, you're making it almost impossible for any of them to get enough data to learn properly. You're sending a tiny trickle of budget to each one. As a result, they either stay in the learning phase forever, which means unstable performance and high costs, or they exit with very little confidence, so the results are never that great anyway.

Think of it like this: you want to find the best-tasting cake. The 1:1:1 method is like hiring 10 different bakers, giving each of them just one ingredient and a fiver, and telling them to make a masterpiece. You'll get ten half-baked, mediocre attempts. The modern approach is to hire one brilliant baker, give them a fully stocked pantry (your 3-5 creatives) and a proper budget, and let them experiment to find the winning recipe. This is what you do when you consolidate your ads into a single adset. You're giving the algorithm the ingredients and budget it needs to do its job properly. One of our software clients was struggling with this, and once we restructured their campaigns to test creatives properly, we got them 5,082 software trials at just $7 a pop. That just wouldn't have been possible with a fragmented 1:1:1 setup.

When you run campaigns with the objective set to 'Reach' or 'Brand Awareness', you're telling the algorithm to "Find me the biggest audience for the cheapest price". And it does exactly that. It finds the people inside your targeting who are least likely to click, engage, or buy anything, because their attention is cheap. You are literally paying to reach non-customers. The best awareness comes from a sale. That's why your campaign objective should almost always be a conversion event, like a lead or a purchase. You let the results create the awareness, not the other way around.

I'd say you need a simple system, not a complicated structure...

That "messy" feeling you described comes from a lack of a clear testing system. When you just throw four random creatives into an adset, it does feel chaotic. The key is to be methodical. You don't need a hundred campaigns. You probably only need a few, each with a very clear purpose.

Here’s a basic framework we use that keeps things simple but is incredibly effective:

1. The Campaign Level: Set the Goal and Budget.
Start with one campaign for your 'prospecting' efforts (finding new customers). Turn on Campaign Budget Optimisation (CBO). This is vital. CBO lets Meta automatically distribute your total campaign budget to the best-performing adset(s) in real-time. If you're manually setting budgets at the adset level, you're back to guessing. Let the algorithm do the heavy lifting.

2. The Adset Level: Test ONE Big Idea at a Time.
This is where you stop the mess. Your adsets should be for testing your big hypotheses. The most common thing to test here is the audience. So, inside your one CBO campaign, you might create three adsets:
-> Adset 1: Broad Targeting. Just location, age, gender. Let the algorithm find the audience based on your pixel data. This works surprisingly well once your account is seasoned.
-> Adset 2: Interest Stack. Group together 5-10 highly relevant interests that define your ideal customer. Think about what podcasts they listen to, what software they use, what influencers they follow. Your ICP isn't a demographic; it's a problem state. Target their pain.
-> Adset 3: Lookalike Audience. Create a 1% Lookalike from your best customer list or from people who have purchased. This is usually your strongest performer.

3. The Ad Level: Test Your Messaging and Creatives.
Inside *each* of those three adsets, you place the *same* 3-5 ads. This is where you test your creative angles. Don't just test different colours on a button. Test fundamentally different ideas. For instance:
-> Ad 1 (Video): A User-Generated Content (UGC) style video showing a real customer's testimonial.
-> Ad 2 (Image): A high-quality image that uses the 'Before-After-Bridge' framework. "Your life sucks now (before). Imagine if it was amazing (after). Our product is the bridge."
-> Ad 3 (Carousel): A carousel ad that breaks down 3 key benefits of your product.
-> Ad 4 (Image): A direct, punchy ad that uses the 'Problem-Agitate-Solve' copy. "Got this painful problem? It's even worse than you think. Here's the solution."

This structure isn't messy. It's a scientific grid. You have one campaign. Inside, you're testing three audiences. Within each, you're testing four creative concepts. After a few days, CBO will have pushed most of the budget to the winning adset, and within that adset, the algorithm will be favouring the winning ad. You'll know which audience *and* which creative angle resonates most. It's clean, logical, and it gives you actionable data.

Here’s what that looks like in a table:

You probably should redefine what 'winning' looks like...

You mentioned you "turn off each one that wasn't spending". This is one of the most common mistakes I see people make. The algorithm is designed to quickly identify a potential early winner and push spend towards it. That doesn't mean the other ads are failures. Sometimes, an ad might have a slightly higher CPC but a much higher conversion rate, making it more profitable in the long run. By turning it off too early, you're robbing the algorithm of the chance to figure that out.

You need to give your ads a fair shot. A good rule of thumb is to let each *adset* spend at least 2-3 times your target Cost Per Acquisition (CPA) before you make any decisions. If your target CPA is £30, let the adset spend £60-£90. Then you can look inside and see what's happening. Is one ad getting all the spend and all the conversions? Great, that's a winner. Is another ad getting clicks but no conversions? The hook is good, but the landing page or offer might be the problem. You need to analyse, not just switch things off based on spend.

And this brings us to the most important metric of all, which most people ignore: Customer Lifetime Value (LTV). You can't know if your CPA is "good" or "bad" unless you know what a customer is actually worth to you. If a customer is worth £10,000 over their lifetime, then paying £300 to acquire them is an incredible bargain. But if they're only worth £100, then a £50 CPA is a disaster. Forget about cheap leads; focus on profitable customers.

Most businesses don't know their numbers. So, before you spend another penny on ads, you need to calculate your LTV. It's the only way to make intelligent decisions about your ad spend.

You'll need to focus on what really matters...

The truth is, campaign structure is only a small part of the equation. It's an important part, because a bad structure will cripple your efforts. But the best structure in the world won't sell a product nobody wants, or sell a great product with a terrible offer. The number one reason paid ad campaigns fail is the offer. It's either not valuable enough, or it's being shown to an audience that has no urgent need for it.

Before you obsess over the number of ads in an adset, you have to nail these things first:

1. Your Audience's Nightmare: Who are you *really* selling to? Not "small business owners". That's a demographic. You need to know their pain. A fractional CFO service isn't selling spreadsheets; they're selling a good night's sleep to a founder who is terrified of running out of cash. Your ads need to speak directly to that specific, urgent, expensive nightmare.

2. An Irresistible Offer: The "Request a Demo" button is dead. It's high friction and offers zero immediate value. Your offer must solve a small, real problem for free to earn you the right to solve the bigger one. For a SaaS company, that's a free trial with no credit card. For an agency, it could be a free, automated audit tool. For us, it's a free strategy session where we map out a plan for potential clients. You have to give value before you can ask for a sale.

3. A Full-Funnel Approach: Finding new customers (prospecting) is just the start. You also need campaigns to re-engage people who have visited your site but not purchased (retargeting), and campaigns to sell more to your existing customers (retention). A common mistake is putting 100% of the budget into prospecting and then wondering why the conversion rate is so low. People need to see your brand multiple times before they trust you enough to buy. You should be prioritising your audiences based on how close they are to converting, from top-of-funnel (ToFu) interest-based audiences down to bottom-of-funnel (BoFu) audiences like 'added to cart'.

This is the main advice I have for you:

I know this is a lot to take in, and it's a big shift in thinking from managing ads to managing a system. It can feel a bit daunting at first, but this is the approach that separates the accounts that struggle from the ones that scale profitably. It's what we spend all day implementing and refining for our clients, from eCommerce stores to complex B2B software companies. For one medical job matching SaaS, we used this exact kind of methodical testing to reduce their cost per user from a painful £100 down to just £7.

Getting this right involves a lot of moving parts, and sometimes it's more efficient to have an experienced pair of eyes on it. If you'd like to go through your account together and build a tailored strategy based on your specific goals, we offer a completely free, no-obligation consultation call. We can take a look at what you've been running and give you a clear, actionable plan to move forward.

Hope this helps clear things up a bit!

Regards,

Team @ Lukas Holschuh