Solved: High CPMs on New Advantage+ Campaigns

Hi there,

Thanks for reaching out! I've had a look at your question and it's a really common one we see all the time. It's frustrating when you're promised a simple, automated solution like Advantage+ but the reality is just sky-high costs and a feeling that you're just burning money. You're definitely not alone in seeing this.

The short answer is that Advantage+ with a fresh pixel is like asking a brand-new taxi driver to find a hidden side street in a city they've never been to before – they're going to drive around aimlessly for a long time, and the meter will be running the whole time. You can't expect the algorithm to work its magic when it has absolutely no idea who your customers are.

I'm happy to give you some of my initial thoughts and a bit of guidance on how we'd approach this. The solution isn't to wait for costs to go down; it's to take control and teach the algorithm who to look for. Below I've outlined the exact process we use to warm up a new ad account and build a solid data foundation before ever letting an automated tool like Advantage+ take the wheel.

The Big Myth: Why You're Paying Meta to Find Non-Customers

Here’s the uncomfortable truth about campaign objectives and automation on platforms like Meta. When you set your campaign objective to "Leads" and give an automated tool like Advantage+ a fresh, data-empty pixel, you are giving the algorithm a very specific, and flawed, command: "Find me the cheapest possible people who might, just maybe, fill out a form."

The algorithm, in its infinite wisdom, does exactly what you asked. It seeks out the users who are least in-demand, who click on anything, and are absolutely, positively least likely to ever become a valuable customer. Why? Because those users' attention is dirt cheap. You are actively paying the world's most powerful advertising machine to find you the worst possible audience for your product. This is why your CPMs and lead costs feel so high—you're paying a premium for low-quality traffic that doesn't convert into anything meaningful.

Real success on Meta isn't about finding the cheapest clicks; it's about finding actual customers. Awareness and leads are a byproduct of having a great offer that solves a real problem for a specific group of people, not a prerequisite for making a sale. To find those people, you have to temporarily fire the algorithm as your main strategist and become the strategist yourself.

I'd say you need to teach the algorithm what a good customer looks like

So, how do you fix it? You need to stop asking the algorithm to find customers and start telling it what your customers look like. This means pausing your Advantage+ campaigns and going back to basics with manual targeting. The goal for the first few weeks isn't necessarily a low Cost Per Lead (CPL); it's to gather at least 100-500 high-quality conversion events on your pixel. These events are the 'food' that will train the algorithm.

This is where most people get it wrong. They create broad, generic audiences because they think a bigger audience means more opportunities. It doesn't. It just means more room for the algorithm to find those cheap, irrelevant users we talked about. You need to become an expert in your customer's specific, urgent problems and interests.

Forget sterile demographics like "Men aged 25-45". That tells you nothing. You need to define your customer by their pain and their passions. What niche podcasts do they listen to? What industry newsletters do they actually open? What software tools do they already pay for? What specific influencers or thought leaders do they follow?

Let's take a practical example. Say you're selling a project management tool for small agencies.

The Bad Approach: Targeting broad interests like "Project Management", "Small Business", or "Marketing". This pool includes students, employees with no purchasing power, enterprise managers, and a million other people who will never buy your product. The algorithm will default to showing your ad to the students because they are cheapest to reach.

The Good Approach: Targeting a layered combination of specific interests. For instance:

• -> People who have shown interest in agency-specific software like Asana, Trello, or ClickUp.
• -> And who are also "Facebook Business Page Admins".
• -> And who also follow agency thought leaders like David C. Baker or Seth Godin.

This creates a much tighter, more relevant audience. Every person in this audience is far more likely to be your ideal customer. Yes, the audience size will be smaller, and your CPM might even be slightly higher initially, but the quality of the clicks and subsequent leads will be exponentially better. You're no longer wasting money on uninterested eyeballs. You are paying to get in front of people with a pre-qualified need. This is the first and most important step to feeding your pixel with the right kind of data.

I remember one B2B software client we worked with who was struggling with a £100 Cost Per Acquisition. They were using broad targeting. We paused everything and rebuilt their audiences from the ground up, focusing on niche software interests and job titles. Within a few months, we got their CPA down to just £7. It wasn't magic; it was just a logical process of giving the algorithm better data to work with.

You probably should follow a phased campaign structure

Alright, so we've established you need to manually feed the pixel. But you can't just run one ad set and hope for the best. You need a structured, phased approach that builds on itself. This is how you systematically create a powerful data asset in your ad account that you can then scale from.

We typically break this down into three core phases: Foundation (ToFu), Optimisation (MoFu/BoFu), and Scaling.

Phase 1: Foundation (Top of Funnel - ToFu)

This is where you start. The only goal here is to gather data by driving relevant traffic. You'll run a 'Conversions' campaign optimised for your primary goal (e.g., 'Lead'). Inside this campaign, you'll create multiple ad sets, each testing a different 'interest cluster' based on the deep customer research you did.

• -> Ad Set 1: Competitor Targeting. Target interests related to direct and indirect competitors.
• -> Ad Set 2: Tools & Software Targeting. Target interests in software your ideal customer uses daily.
• -> Ad Set 3: Media & Influencer Targeting. Target interests in publications, podcasts, and people your customer follows.
• -> Ad Set 4: Job Title / Behaviour Targeting. Target specific job titles or behaviours (like 'Small Business Owners').

Run these with a small daily budget (£20-£50 per ad set is fine). Let them run for at least 4-7 days, or until each ad set has spent about 1-2x your target CPL. Some will fail spectacularly. That's fine. You are paying for data. Turn off the clear losers and reallocate budget to the winners. The goal is to get your first 100+ 'Lead' events on the pixel from these manually targeted audiences.

Phase 2: Optimisation (Middle & Bottom of Funnel - MoFu/BoFu)

Once you have a steady stream of traffic and some conversions (at least a few hundred unique website visitors), you can start the next phase. Now you'll build your 'retargeting' audiences. These are almost always your most profitable campaigns.

Create a new campaign, again for 'Conversions'. This time, your audiences will be 'Custom Audiences' made up of people who have already interacted with you:

• -> Ad Set 1 (BoFu): Target people who visited your landing page but didn't convert in the last 14-30 days. These are warm leads.
• -> Ad Set 2 (MoFu): Target people who engaged with your Facebook or Instagram page in the last 90 days.
• -> Ad Set 3 (MoFu): Target people who watched 50% or more of one of your video ads in the last 90 days.

These audiences are smaller but incredibly high-intent. Your CPL from these campaigns should be significantly lower than your prospecting (ToFu) campaigns.

Phase 3: Scaling with Lookalikes & (Finally) Advantage+

This is where the magic you were initially hoping for starts to happen. Once your pixel has at least 100 conversions (but ideally 500+) from a single country, you can start creating 'Lookalike Audiences'.

A Lookalike Audience is Meta taking your list of customers (e.g., everyone who became a lead) and finding millions of other users who share similar characteristics. This is infinitely more powerful than manual interest targeting.

Create a new prospecting campaign and test the following Lookalike ad sets:

• -> Ad Set 1: 1% Lookalike of Leads. This is your highest quality audience.
• -> Ad Set 2: 1% Lookalike of Website Visitors. Broader, but still very effective.
• -> Ad Set 3: 1-3% Lookalike of Leads. A slightly larger, but still potent audience.

Now, and only now, is when you can consider re-introducing Advantage+. With a pixel rich with thousands of visitor events, hundreds of lead events, and several high-performing custom and lookalike audiences, Advantage+ finally has the data it needs to work properly. It can use your existing audiences as a strong 'suggestion' and will be far more effective at finding new customers that actually resemble your existing ones.

You'll need realistic cost expectations

So, the big question: "how long does it take before costs go down?". The honest answer is that it depends. But costs don't just "go down" on their own; they go down because you implement a better strategy like the one I've just outlined. Following this phased approach, you should start seeing better quality leads and a more stable CPL within 2-4 weeks, as your retargeting and then lookalike campaigns come online.

But what is a "good" CPL anyway? The truth is, it varies massively by industry, country, and offer. It's much more productive to stop asking "how low can my CPL go?" and start asking "how high a CPL can I afford to acquire a great customer?". The answer to that lies in your Customer Lifetime Value (LTV).

Once you know what a customer is worth to you, you can work backwards to determine an acceptable CPL. For example, if your LTV is £5,000 and you have a 10% lead-to-customer conversion rate, a single customer is 'born' from 10 leads. This means you can afford to spend up to £500 on those 10 leads (£50 per lead) and still break even. A healthy business model would aim for a 3:1 LTV:CAC ratio, meaning you could afford to spend £1666 to acquire that customer, which translates to a £166 CPL. Suddenly, a £40 lead doesn't seem so expensive, does it? It looks like a bargain.

This is the maths that unlocks intelligent growth. It frees you from the tyranny of chasing cheap, low-quality leads. Use the calculator below to get a rough idea of your own numbers.

That said, it's still useful to have some general benchmarks. Based on the campaigns we've run for dozens of clients, here are some very rough ranges you might expect for lead generation in developed countries like the UK, US, or Australia.

As you can see, there's a huge range. Your job, using the phased approach, is to systematically move your costs from the right side of that table to the left. For one of our software clients, we got their cost per trial down to $7 using this exact methodology. It's absolutely achievable with a structured approach.

This is the main advice I have for you:

To pull it all together, here is a summary of the actionable steps I'd recommend you take, starting today. This isn't a quick fix, but it's a reliable process for building a profitable and scalable advertising machine instead of just a money pit.

I know this can seem like a lot of work, and it is. Getting this structure right, doing the audience research, writing compelling ad copy, and managing the budgets across these different phases is a full-time job. It requires expertise and a disciplined, data-driven approach. While the principles are straightforward, the execution can be tricky and time-consuming.

This is where working with a specialist can make a massive difference. An expert has been through this process hundreds of times and can navigate the initial data-gathering phase much more quickly and efficiently, saving you time and wasted ad spend. They can help you identify those winning audiences faster and scale them more aggressively once they're found.

If you'd like to chat through your specific situation in more detail, we offer a free, no-obligation initial consultation. We could have a proper look at your account together and give you a more tailored plan of action. Feel free to get in touch if that sounds helpful.

Regards,

Team @ Lukas Holschuh