Hi there,
Thanks for reaching out!
Happy to give you some initial thoughts and guidance on the high lead costs you're seeing. To be honest, I reckon you might be focusing on the wrong problem. The "learning phase" is often a bit of a red herring that gets people chasing their tails, when the real issue is usually buried a lot deeper in the strategy itself. Your costs didn't just jump because of an algorithm reset; they jumped because the underlying engine of your campaign might not be as strong as it needs to be.
Let's unpack that a bit.
We'll need to look at the 'Learning Phase' for what it really is... a symptom, not the disease.
Right, first things first. Every time you make a significant change to an ad set – changing creative, text, targeting, or even the budget too drastically – you'll throw it back into learning. That's just how the system works. The algorithm needs to re-gather data to figure out who to show your ads to. So seeing a temporary spike in costs isn't completely unexpected. It's frustrating, I know, but it's part of the game.
However, being stuck there for 10 days and seeing costs jump from €0.90 to €6 suggests something more fundamental is at play. The learning phase just exposes weaknesses; it doesn't create them. A robust campaign with a killer offer and on-point targeting will usually exit learning relatively quickly and stabilise at a profitable CPL. A weak one will struggle, and the learning phase just magnifies the poor performance.
Too many advertisers fall into the trap of thinking they need to "beat" the learning phase. The truth is, you need to give the algorithm something genuinely good to learn from. Here’s a bit of a contrarian take for you: when you run a campaign, you're giving Meta a set of instructions. If you tell it to optimise for leads, it will go find people it thinks are most likely to convert based on the creative and targeting you provided. If those inputs are average, it's going to have a hard time. It's like trying to cook a gourmet meal with cheap ingredients. The chef can only do so much.
This is even more true for objectives like "Brand Awareness" or "Reach". If you select those, you're basically telling Meta: "Find me the largest number of people for the lowest possible price." The algorithm does exactly that, and serves your ad to people who are cheap to reach precisely because they never click, engage, or buy anything. You're actively paying to find non-customers. You're running a conversion campaign which is the right call, but the principle is the same: the quality of your inputs dictates the quality of the output. The learning phase is just the period where the machine figures out just how good (or bad) those inputs are.
So, instead of worrying about the learning phase itself, we need to fix the things that are causing it to fail. And that almost always starts with the offer.
I'd say you need to diagnose your offer and your customer's real problem.
This is probably the single biggest reason campaigns fail. Not the creative, not the targeting, but the offer. An offer that isn't compelling, doesn't solve an urgent problem, or asks for too much commitment too soon is doomed from the start. Your ads could be brilliant, but if what you're offering on the other side of the click is weak, your CPL will skyrocket.
You mentioned your CPL went from €0.90 to €6. While that's a big jump, a €6 CPL isn't inherently "high". It all depends on what that lead is ultimately worth to you. If a lead is worth €100 in profit down the line, €6 is an absolute bargain. If it's worth €5, you're losing money. This is why the first thing we need to do is stop guessing and start calculating.
You need to figure out your Customer Lifetime Value (LTV). This is the number that tells you how much you can actually afford to spend to acquire a customer, which in turn tells you what a "good" CPL really is. Without this, you're just flying blind. Let's run through a quick example. Imagine you run a service business or a SaaS product.
How to Calculate Your Lifetime Value (LTV)
| Metric | Example Value | Description |
|---|---|---|
| Average Revenue Per Account (ARPA) | €200 / month | What you make per customer, per month on average. |
| Gross Margin % | 75% | Your profit margin on that revenue after direct costs. |
| Monthly Churn Rate | 5% | The percentage of customers you lose each month. |
|
The Calculation: LTV = (ARPA * Gross Margin %) / Monthly Churn Rate LTV = (€200 * 0.75) / 0.05 LTV = €150 / 0.05 LTV = €3,000 |
||
In this scenario, each customer is worth €3,000 in gross margin. A healthy ratio of LTV to Customer Acquisition Cost (CAC) is at least 3:1. This means you could afford to spend up to €1,000 to acquire a single customer. If your sales process converts 1 in 20 leads, you can afford to pay up to €50 per lead (€1,000 / 20). Suddenly, that €6 CPL looks incredibly cheap, doesn't it? This math is what separates businesses that scale from those that stagnate.
Once you know what a lead is worth, you can build an offer that truly delivers value. Forget generic offers like "Learn More" or "Contact Us". They are high friction and low value. Your offer needs to solve a small, tangible piece of your customer's biggest nightmare, for free, right now. It must be an "aha!" moment that makes them want more.
Think about it. Your ideal customer isn't defined by demographics. "Men aged 25-40 who like business" is useless. Your Ideal Customer Profile (ICP) is a person stuck in a specific, urgent, expensive nightmare. Your job is to find them and offer a painkiller. For example:
- -> If you're a marketing agency, don't offer a "consultation". Offer a "Free SEO audit that uncovers your top 3 missed keyword opportunities".
- -> If you're a SaaS company, don't just offer a demo. Offer a free trial with no card details, or a freemium plan. Let the product prove its own value. I remember a B2B SaaS client we worked with who saw a massive shift after moving from a demo request to a free trial. We got them 1535 trials on Meta Ads because the offer was so much stronger.
- -> If you're a service provider, don't just say "Get a Quote". Offer a "Free 15-minute diagnostic to identify the #1 bottleneck in your workflow".
This is about providing undeniable value upfront to earn the right to ask for a sale later. It pre-qualifies them and makes them sell themselves on your solution. This is a far more effective strategy than just hoping people will fill out a form.
You probably should re-evaluate your message and who you're talking to.
Once your offer is solid, your ad copy needs to communicate it in a way your audience can't ignore. It needs to speak directly to their pain. Most ads are boring, feature-focused, and talk at people. Yours needs to enter the conversation already happening in their head.
Here are a couple of frameworks that work wonders:
1. Problem-Agitate-Solve (PAS): This is perfect for service businesses.
- -> Problem: State the pain point they feel right now. "Struggling to keep up with customer support tickets?"
- -> Agitate: Pour salt on the wound. Make them feel the consequences of not solving it. "Are negative reviews piling up while your team is stretched thin and burning out?"
- -> Solve: Introduce your offer as the clear, simple solution. "Get our AI-powered chatbot setup in 24 hours and instantly resolve 70% of customer queries, freeing up your team to focus on what matters."
2. Before-After-Bridge (BAB): Great for software or transformational products.
- -> Before: Paint a picture of their current world of pain. "Another Monday morning, another 5 hours lost trying to reconcile spreadsheets from three different departments."
- -> After: Show them the dream world. "Imagine a single dashboard, updated in real-time, where all your key metrics are clear, accurate, and actionable."
- -> Bridge: Position your product as the bridge to get them there. "Our platform is the bridge. Connect your tools in minutes and see your business in a new light. Start your free trial today."
This kind of messaging, combined with a strong offer, gives the Meta algorithm the fuel it needs. It will start to find people who resonate with the problem, which means higher click-through rates, more relevant traffic, and ultimately, a lower cost per lead. I've seen this time and time again. One of our SaaS clients, a medical job matching platform, was struggling with a £100 CPA. By overhauling their offer and messaging to speak directly to the frustrations of both doctors and hospitals, we helped reduce their CPA to just £7. That's not an algorithm tweak; that's a fundamental strategy shift.
And this leads us to targeting. Your message is only effective if it reaches the right people. You need to target based on the problem, not just demographics. For example, if you're selling an e-commerce tool, don't just target the broad interest "eCommerce". That's full of customers, not store owners. Instead, target interests like "Shopify", "WooCommerce", or followers of e-commerce podcasts and influencers. You're looking for signals that they live inside the world where your solution is relevant.
You'll need a proper campaign structure to test this effectively.
Okay, so you've got a killer offer and sharp messaging. Now you need to structure your campaigns in a way that lets you test properly and find winning audiences without breaking the bank. A common mistake is lumping everything into one ad set. You need to separate things out to see what's actually working.
For any account, I'd prioritise audiences in this order, moving from coldest to warmest. This is a structure we use for e-commerce, but the logic applies to lead generation just as well.
| Funnel Stage | Audience Type | Audiences to Test (in order of priority) |
|---|---|---|
| ToFu (Top of Funnel - Cold) | Detailed Targeting | Test separate ad sets for different themes of interests, behaviours, and demographics that align with your customer's 'nightmare'. This is your starting point. |
| Lookalike Audiences | Once you have enough data (at least 100-1000 conversions), create lookalikes of your best leads, website visitors, and video viewers. Prioritise lookalikes of higher-intent actions. | |
| MoFu (Middle of Funnel - Warm) | Engagement Retargeting | Retarget people who have visited your landing page, watched a significant portion of your videos, or engaged with your ads, but haven't converted yet. Exclude converters. |
| BoFu (Bottom of Funnel - Hot) | High-Intent Retargeting | This is your smallest but most valuable audience. Retarget people who started filling out your lead form but abandoned it, or visited your pricing page. This is where you can be more direct with your messaging. |
Start with a ToFu campaign. Create different ad sets inside it, each targeting a specific interest 'theme'. Let them run for a few days. Don't be too quick to judge. A good rule of thumb is to let an ad set spend at least 2-3x your target CPL before you make a call on it. If it's not delivering, turn it off. Funnel the winners' budget to test new audiences.
As soon as you have enough data, launch your MoFu and BoFu retargeting campaigns. These will almost always give you a lower CPL because you're talking to people already familiar with you. If your budget is small, you can combine MoFu and BoFu audiences into a single retargeting ad set to start with.
This structured approach lets you systematically find what works and scale it, rather than just randomly changing things and hoping for the best. It's definitely more work upfront, but it pays off massively in the long run by giving you stable, predictable lead flow.
I've detailed my main recommendations for you below:
This might all seem like a lot, but it's about shifting your focus from short-term tactics (like worrying about the learning phase) to long-term strategy. Getting the fundamentals of your offer, message, and targeting right is what will truly drive down your costs and allow you to scale.
| Step | Actionable Advice | Why It Matters |
|---|---|---|
| 1. Forget the Learning Phase | Accept that any significant ad change will reset it. Stop obsessing over the status and focus on the inputs you're giving the algorithm. | Frees you up to focus on what actually moves the needle: your core strategy. The learning phase is a symptom, not the cause of poor performance. |
| 2. Calculate Your Allowable CPL | Determine your customer LTV using the formula (ARPA * Gross Margin) / Churn Rate. From there, work backwards to find out what you can afford to pay for a lead. | This turns advertising from a cost centre into a profit driver. It gives you a clear KPI and lets you know if a €6 CPL is a disaster or a huge bargain. |
| 3. Strengthen Your Offer | Replace a generic "Contact Us" or "Learn More" with a high-value, low-friction offer that solves a real, tangible problem for your audience (e.g., a free audit, a tool, a specific guide). | A powerful offer is the #1 driver of lower lead costs. It makes your ad irresistible and pre-qualifies prospects before they even click. |
| 4. Rewrite Your Ad Copy | Use frameworks like Problem-Agitate-Solve or Before-After-Bridge to speak directly to your audience's pain points and desired outcomes. Make it about them, not you. | This grabs attention, increases relevance, and improves click-through rates, which signals to Meta that your ad is high quality, leading to lower costs. |
| 5. Implement a Funnel Structure | Separate your campaigns into ToFu (cold audiences), MoFu (warm engagement retargeting), and BoFu (hot intent retargeting). Test interest-based audiences methodically in your ToFu campaign. | This gives you a systematic way to test, find winning audiences, and scale your spend efficiently. It prevents you from wasting money on audiences that don't work. |
As you can probably tell, this isn't just about flicking a few switches in Ads Manager. It's about deep-diving into your business strategy, understanding your customer on a profound level, and translating that into a marketing system that works. It takes time, expertise, and a lot of testing to get right.
This is where professional help can make a huge difference. An experienced eye can spot the opportunities and weaknesses you might miss, and can implement these kinds of advanced strategies far more quickly and effectively. We do this day in, day out, and have seen what works (and what doesn't) across dozens of industries.
If you'd like to go through your campaigns and strategy in more detail, we offer a free, no-obligation initial consultation where we can give you some more specific pointers. It might be a good way to get a clear, actionable plan together.
Either way, I hope this has given you a new way to think about the problem and some concrete steps to take. Stop fighting the algorithm and start feeding it what it needs to succeed.
Regards,
Team @ Lukas Holschuh
