Solved: Campaign CPM/CPC Same, 0 Leads After Audience Change

Hi there,

Thanks for reaching out!

Happy to give you some initial thoughts on the campaign fluctuations you're seeing. It’s a really common situation, especially with new campaigns, so don’t worry, you haven't broken anything. The short answer is that what you're experiencing is pretty normal behaviour for ad platforms like Meta, but I'll walk you through exactly why it happens and what you should be looking at instead of the daily ups and downs.

Let's get into it.

We'll need to look at the 'Learning Phase' first...

Right, the first and most important thing to understand is something called the "learning phase." When you launch a new campaign, the ad platform's algorithm doesn't magically know who the perfect customer for you is. It has to learn. It does this by showing your ad to a wide variety of people within your targeting parameters to see who clicks, who engages, and ultimately, who converts.

Think of it like hiring a new salesperson. On their first day, you wouldn't expect them to close ten deals. They need time to learn the product, understand the customer, and figure out what sales pitch works. The first couple of days might be a mix of beginner's luck and failed attempts. The algorithm is exactly the same. It needs to gather data—typically around 50 conversions per ad set within a 7-day period—to get a clear picture of your ideal customer. Until it reaches that point, performance will be all over the place. It's designed to be that way.

Those first two days where you got great results? That was likely a bit of luck. The algorithm found a small, cheap pocket of people who were ready to convert. But that pocket was small and quickly exhausted. Now, it's having to work harder, explore different segments of your audience, and that's why your costs might go up and your results might dip before they stabilise. Three days is simply not enough data to judge a campaign. I tell all our clients to not even look at the results for the first 7-10 days. Making changes based on one bad day is the quickest way to kill a potentially successful campaign because you're constantly resetting the learning process.

The key here is patience. You have to let the system do its job and gather enough data to make intelligent decisions. Constantly tinkering with targeting, placements, or creatives in the first week is like pulling a plant up every day to see if its roots are growing. You just end up killing it.

I'd say you need to understand the impact of your change...

You mentioned you removed the Audience Network from your placements. On the surface, this is often a good move. The Audience Network consists of placements in third-party apps and websites, and while it can deliver a high volume of cheap clicks, the quality is often dreadful. It's notorious for accidental clicks and low-intent traffic. So, removing it was probably the right instinct for improving lead quality in the long run.

However, you need to understand the immediate effect of that change. By removing a major placement category, you fundamentally changed the instructions you gave the algorithm. You told it, "Stop looking for people over there, and focus all your efforts on these more expensive, more competitive placements like the Facebook and Instagram feeds."

This does two things:

1. It Resets the Learning Phase: Because you've made a significant edit, the algorithm has to start learning again from scratch with the new placement constraints. This is likely the primary cause of the performance drop you saw on day three. The system discarded its previous learnings and started a new test.
2. It Can Increase Costs: Placements like the main feeds are more in-demand. There's more competition from other advertisers. This naturally drives up the cost (CPM and CPC). So while the quality of clicks might eventually be better, the initial cost to acquire them will almost certainly be higher than what you were getting on the Audience Network.

The mistake wasn't removing the Audience Network; the mistake was panicking when the system reacted predictably to the change. My advice is to leave it off now and let the campaign run for another full week without touching anything else. Only then will you have a baseline of data to see how it's *actually* performing on the higher-quality placements.

You probably should diagnose the real problem...

Fluctuations are a distraction. The real work in paid advertising isn't reacting to daily changes; it's systematically diagnosing where your funnel is leaking money. You need to stop thinking about "getting links" and start thinking about the entire customer journey, from the moment they see your ad to the moment they become a lead.

Here's how you break it down:

• Step 1: The Ad (Attention)
  This is about the creative and the copy. The main metric to watch here is your Click-Through Rate (CTR). Is it low (e.g., below 1%)? If so, your ad isn't compelling enough. It's not stopping people from scrolling. Your image might be boring, your headline might be weak, or your copy doesn't speak to a real pain point. High CPCs and CPMs are often a symptom of a low CTR, because the platform penalises ads that users ignore.


• Step 2: The Click to the Landing Page (Interest)
  Are you getting a lot of "Link Clicks" in Ads Manager but seeing a much lower number of "Landing Page Views"? This points to a technical problem, usually page load speed. If your website takes more than 3 seconds to load, a significant percentage of people who clicked your ad will simply give up and leave before the page even appears. You're paying for clicks that never even see your offer.


• Step 3: The Landing Page (Desire & Action)
  This is where most campaigns fail. You're getting people to the page, but they aren't converting. This is almost never an "ad" problem; it's an "offer" problem.
  • -> Is your value proposition crystal clear within the first five seconds?
  • -> Does the copy on the page match the promise made in the ad?
  • -> Is the page cluttered and confusing, or does it have a single, clear call to action?
  • -> Is your offer compelling enough? We'll talk more about this later, but "Contact Us" is not a compelling offer.
  • -> Does your page look trustworthy? Social proof, testimonials, and a professional design are not optional.

To make this easier, I've mapped it out in a flowchart. Follow the questions logically to find your bottleneck.

You'll need a better campaign structure...

Another common mistake I see is cramming everything into one campaign and one ad set. You can't learn anything that way. If you have three different audiences and five different ads all jumbled together, how do you know what's working and what's not? You need to isolate variables so you can test methodically.

A solid starting structure looks something like this:

• Campaign 1: Prospecting (ToFu - Top of Funnel)
  • -> Ad Set 1: Interest Group A (e.g., targeting people interested in your direct competitors)
    • - Ad 1 (Video)
    • - Ad 2 (Image)
    • - Ad 3 (Carousel)
  • -> Ad Set 2: Interest Group B (e.g., targeting people interested in industry publications or software they use)
    • - Ad 1 (Video)
    • - Ad 2 (Image)
    • - Ad 3 (Carousel)
  • -> Ad Set 3: Lookalike Audience (e.g., a 1% Lookalike of your past customers, once you have enough data)
    • - Ad 1 (Video)
    • - Ad 2 (Image)
    • - Ad 3 (Carousel)


• Campaign 2: Retargeting (MoFu/BoFu - Middle/Bottom of Funnel)
  • -> Ad Set 1: Website Visitors (Last 30 Days) - Exclude converters
    • - Ad 4 (Testimonial Ad)
    • - Ad 5 (Case Study Ad)

With this structure, you can clearly see which *audience* is performing best at the ad set level, and which *creative* is performing best at the ad level. After a week or two, you can confidently turn off the losing ad sets and ads and reallocate budget to the winners. This is how you systematically improve performence, rather than guessing and making random changes based on one day's data.

For a new account, you'd start with detailed targeting (interests). Once you get at least 100 conversions (and ideally more like 500-1000), you can start building powerful Lookalike audiences and effective retargeting campaigns. One campaign we're running for a software client generated 4,622 registrations at just $2.38 each using this exact method of testing interest groups on Meta before scaling with lookalikes.

Let's calculate a realistic cost per conversion...

It's easy to get anchored to the cheap results from your first two days, but that's rarely sustainable. It's much more productive to understand the maths behind your cost per lead (CPL) so you can set realistic expectations.

Your CPL is determined by two main factors: your Cost Per Click (CPC) and your Landing Page Conversion Rate (CVR).

CPL = CPC / CVR

In developed countries like the UK or US, a typical CPC on Meta ads might range from £0.50 to £1.50. A decent landing page might convert visitors into leads at a rate of 10-30%. Let's see what that means for your costs. Use the calculator below to see how these two metrics affect your final CPL.

As you can see, a CPL between £2 and £15 is perfectly normal. If your initial results were below this, they were an anomaly. Your goal shouldn't be to get back to an unsustainably low CPL; it should be to build a system that can reliably generate leads within an acceptable, predictable range. We once worked with a medical job matching SaaS client whose initial CPA was over £100. By systematically optimising their funnel—from ads to landing page—we reduced that to just £7. This wasn't a one-day fix; it was a process of methodical testing and improvement.

The number one reason campaigns fail is the offer...

I've saved the most important point for last. You can have the best targeting, the most beautiful ads, and the fastest website in the world, but if your *offer* is weak, your campaign will fail. This is the part that most people overlook. They spend weeks tweaking ad settings when the real problem lies in what they're asking people to do.

The cardinal sin in B2B marketing, for example, is the "Request a Demo" or "Contact Us" button. This is an incredibly high-friction, low-value call to action. You're asking a busy stranger to commit their valuable time to a sales call with you, before you've provided them with any real value. It's arrogant, and it's why conversion rates on these pages are often terrible.

Your offer's only job is to provide an "aha!" moment of undeniable value. It must solve a small, real problem for your prospect for free, to earn you the right to solve their bigger problems for a fee.

What does a good offer look like?

• For a SaaS company: A free trial (no credit card required) or a freemium plan. Let them experience the product's value firsthand.
• For a service business: A free, valuable asset. A marketing agency could offer an automated SEO audit tool. A financial consultant could offer a cash flow projection template.
• For an agency like us: We offer a free 20-minute strategy session where we audit failing ad campaigns. We provide real value and demonstrate our expertise, which makes the decision to work with us much easier.

You need to stop selling your service and start solving a problem. Re-evaluate your landing page and ask yourself, "What tangible value is a visitor getting by filling out this form, right now?" If the only answer is "a sales call," your offer is the problem. Fix that, and you'll see a much bigger impact on your lead flow than any amount of tweaking ad placements will ever deliver.

This is the main advice I have for you:

Here’s a summary of the actionable steps I’d recommend you take, based on everything we've discussed. Instead of making panicked, short-term decisions, follow this methodical plan.

As you can probably tell, running paid ads effectively is a lot more complex than just launching a campaign and hoping for the best. It's a continuous cycle of strategic planning, methodical testing, data analysis, and optimisation. It takes time, expertise, and a deep understanding of both the platform and marketing psychology to get it right.

Trying to figure all this out on your own can be a slow and expensive process, filled with costly mistakes and missed opportunities. Working with an expert can help you bypass that painful learning curve, implement proven strategies from day one, and start seeing a reliable return on your ad spend much faster.

If you'd like to have a chat about how we could apply these principles to your specific business and help you build a predictable lead generation engine, feel free to schedule a complimentary, no-obligation strategy session with us. We'd be happy to take a look at your setup and give you some more tailored advice.

Hope this helps!

Regards,

Team @ Lukas Holschuh