Hi there,
Thanks for reaching out! It's a really common problem you're describing, and honestly, one of the most misunderstood aspects of running ads on platforms like Meta. It feels counter-intuitive, right? You give the all-powerful algorithm more data, and it gives you worse results. It makes no sense on the surface, but when you understand *how* the algorithm actually works, it starts to become clear.
You're not doing anything wrong, you've just hit the predictable ceiling of the algorithm's first, easiest job. The good news is there's a clear, systematic way to break through this cycle. I'll walk you through my thoughts on what's happening and what to do about it.
TLDR;
- Your ads get worse because Meta's algorithm finds and exhausts the 'lowest-hanging fruit'—a small pocket of easy-to-convert users—within your broad audience very quickly. This isn't a failure, it's how the system is designed to work.
- "Broad targeting" isn't really broad. The algorithm immediatly narrows its focus to the people it thinks are most likely to convert based on trillions of data points. That initial success is it burning through that tiny, hyper-responsive segment.
- The solution isn't to keep resetting the same campaign. It's to build a structured system with different campaigns, creatives, and audiences that constantly finds new 'pockets' of buyers.
- Stop focusing solely on the immediate Cost Per Acquisition (CPA). You need to calculate your Customer Lifetime Value (LTV) to understand how much you can truly afford to pay for a customer, which gives you the confidence to scale intelligently.
- This letter includes an interactive LTV calculator to help you find your real numbers, and flowcharts illustrating how the algorithm works and how you should structure your account.
We'll need to look at why your best audiences die (and why it's normal)
The core of your frustration comes from a simple misunderstanding of what the "learning phase" is and what "broad targeting" actually means in practice. You think you're telling Facebook to show your ad to millions of people, and over time, it'll get smarter about who it picks. That's not quite what happens.
What you're actually telling it is: "Here are millions of people. Find me the cheapest, fastest sales you possibly can, and ignore everyone else."
The algorithm is a ruthlessly efficient, but very lazy, optimiser. It doesn't explore your huge audience methodically. Instead, it uses its vast data to make an educated guess and find a small, hyper-responsive "pocket" of users who are ready to buy *right now*. These are your easy wins, the lowest-hanging fruit. This is what causes that incredible first week of performance. It's found the gold seam and is mining it as fast as it can.
But that seam is always finite. Once it has converted most of the people in that initial pocket, it's forced to look for the *next* most likely group. This new group is, by definition, slightly less interested, slightly more expensive to reach, and slightly harder to convert. Your CPA starts to creep up. Then it exhausts that pocket and has to find the next, and so on. Each subsequent pocket is lower quality than the last. This is the death spiral you're experiencing. It's not audience fatigue in the traditional sense; it's 'easy-win-pocket' exhaustion.
1. Campaign Launch
You target a "Broad" audience of millions.
2. The "Easy Wins"
The algorithm instantly finds a tiny pocket of hyper-responsive users ready to buy.
3. Peak Performance
For 3-7 days, you get amazing, profitable results as this pocket converts.
4. Pocket Exhaustion
The easy wins run out. The algorithm must now find less-receptive users.
5. Performance Decay
CPA rises, sales slow down, and profitability vanishes.
So, the "learning" isn't about the campaign getting smarter forever. It's a very short phase where it identifies that first pocket. After that, it's just exploitation, followed by a slow decay. Your job isn't to try and keep one campaign alive forever; it's to become an expert at finding new pockets.
I'd say you need to define your customer by their nightmare, not their demographic
This brings us to the next crucial point. If the algorithm is finding these pockets for you, how can you help it? The answer is to stop thinking about your audience as a broad, faceless mass and start thinking about their specific, urgent problems. Even within your "broad" audience, the only people who ever bought from you were the ones experiencing a particular kind of pain that your product solves.
Forget the sterile profiles. "Men aged 25-45" tells you nothing. You need to get deeply specific about their internal world. What is the career-threatening, sleep-depriving nightmare they are living through? Your customer isn't a demographic; they are a problem state.
For example, if you sell project management software, the nightmare isn't 'needing better organisation'. It’s 'the terror a marketing manager feels when she realises a major product launch is about to miss its deadline because her team is communicating in five different places, and she’s going to have to explain the failure to the CEO.' That is a specific, expensive, and urgent pain.
Once you've identified that nightmare, you can work backwards. Where do people with that specific pain hang out online? What podcasts do they listen to (maybe 'Acquired')? What newsletters do they actually read (like 'Stratechery')? What influencers do they follow on Twitter (like Jason Lemkin)? What other tools do they already use (like HubSpot or Salesforce)?
This intelligence becomes the blueprint for your targeting and your messaging. You don't need the algorithm to guess who's in pain; you can start telling it directly. This work is the foundation. Without it, you're just throwing money at the wall and hoping the algorithm gets lucky, which as you've seen, is a short-lived strategy.
You probably should create a message they can't ignore
Now that you understand the customer's nightmare, you can stop running just one generic ad. The ad that worked so well in the first week resonated perfectly with that first "easy win" pocket. To find a *new* pocket, you likely need a *new* message that speaks to a slightly different angle of their pain.
This is where proven copywriting frameworks come in. They are shortcuts to creating compelling ads that connect on an emotional level.
For a product, you can use the Before-After-Bridge framework. I remember one campaign we worked on for a women's apparel brand. They had a great product but were struggling to get consistent sales online. We used this exact framework for their ads.
- Before: "You've designed the perfect collection, but your online store feels invisible. You're seeing more 'add to carts' than actual sales, and every ad you run feels like a gamble."
- After: "Imagine seeing your revenue grow consistently month after month, with a clear return on every pound you spend on ads. You're not just selling clothes; you're building a brand people love and come back to."
- Bridge: "Our new collection is the bridge. Designed for confidence and style, it's what your wardrobe has been missing." The campaign we built around this kind of messaging helped them achieve a 691% return on their ad spend.
For a service, you can use Problem-Agitate-Solve.
- Problem: State the nightmare. "Are your cash flow projections just a shot in the dark?"
- Agitate: Pour salt on the wound. "Are you one bad month away from a payroll crisis while your competitors are confidently raising their next round of funding?"
- Solve: Present your service as the solution. "Get expert financial strategy for a fraction of a full-time hire. We build dashboards that turn uncertainty into predictable growth."
By creating multiple ads, each focusing on a different "before" state or agitating a different part of the problem, you give the algorithm more chances to find new, untapped pockets of customers within that same broad audience.
You'll need the math that unlocks real growth
One of the biggest mental traps you can fall into is obsessing over the daily CPA and ROAS. When it's high, you feel great. When it drops, you panic. This emotional rollercoaster forces you into making short-term decisions. The real question isn't "How low can my CPA go?" but rather "How high a CPA can I profitably afford?"
The answer lies in calculating your Customer Lifetime Value (LTV). This is the total profit you can expect to make from an average customer over the entire duration of their relationship with you. When you know this number, everything changes.
Here’s the basic maths:
1. Average Revenue Per Account (ARPA): What's the average amount a customer pays you per month/year? (Or for eCommerce, what's their average order value multiplied by their purchase frequency?).
2. Gross Margin %: What's your profit margin on that revenue after accounting for the cost of goods sold?
3. Monthly Churn Rate %: What percentage of customers do you lose each month? (For eCommerce, this is trickier, you might use '1 / average customer lifetime in months').
The formula is: LTV = (ARPA * Gross Margin %) / Churn Rate %
Let's imagine you run a subscription box that costs £50/month. Your gross margin is 70%, and you lose 5% of your customers each month.
LTV = (£50 * 0.70) / 0.05
LTV = £35 / 0.05 = £700
This means every customer you acquire is worth £700 in gross profit to your business. A common, healthy ratio for a growing business is a 3:1 LTV to Customer Acquisition Cost (CAC). This means you can afford to spend up to £700 / 3 = £233 to acquire a single customer and still have a very profitable business model. Suddenly, a CPA of £50, £80, or even £150 doesn't seem so scary, does it? It looks like a great investment.
This calculation frees you from the tyranny of short-term profitability and allows you to make intelligent, aggressive decisions to fuel your growth. Use the calculator below to plug in your own numbers.
This is the main advice I have for you: build a campaign structure that never gets tired
So how do we put all this together? You stop thinking in terms of one magic campaign and start building a system—an engine that is constantly prospecting for new customers, converting them, and maximizing their value. You need to move from being a campaign manager to a portfolio manager.
Instead of one single broad campaign, you should have a structure with multiple, ongoing campaigns, each with a specific job. Here’s a simple but powerful structure we use for our clients that you could adapt.
Top of Funnel (ToFu) - Prospecting
- Your current setup
- Goal: Find new pockets
- Expect it to fatigue
- LAA of Purchasers
- LAA of Add to Carts
- Goal: Find similar people
- Based on 'nightmare' research
- Magazines, tools, influencers
- Goal: Test specific hypotheses
Middle/Bottom of Funnel (MoFu/BoFu) - Retargeting
- Target all website visitors from last 30 days (excl. buyers)
- Show them testimonials, reviews, or handle common objections.
- Goal: Convert the unconverted
Here’s how it works:
1. ToFu Prospecting Campaigns: You'll have several. Keep your Broad campaign, but treat it as an exploration tool. When it works, great. When it fades, don't panic. Alongside it, launch a second prospecting campaign targeting Lookalike Audiences. Take the data from your succesful first week (your list of purchasers) and create a 1% Lookalike. This tells Facebook "go find me more people exactly like these". Then, launch a third campaign based on your 'customer nightmare' research, testing specific interests like competing software, industry magazines, or influencers. This diversifies your risk. If Broad fails one week, your Lookalike campaign might be firing, and vice versa.
2. BoFu Retargeting Campaign: This is non-negotiable. It's a separate campaign with one job: target everyone who has visited your website or added a product to their cart in the last 30 days (but hasn't purchased). These people are warm leads. Your ads to them shouldn't be the same as your prospecting ads. They should show social proof (testimonials, reviews), handle objections ("Worried about the price? Here's our guarantee"), or offer a gentle nudge to complete their purchase. This campaign will almost always be your most profitable.
This structure turns the frustrating cycle into a predictable machine. You're constantly feeding the top of the funnel with new people from different sources, and efficiently converting those who show interest at the bottom. It requires more management than a single 'set-and-forget' campaign, but this is how you build a scalable and resilient advertising system.
I've detailed my main recommendations for you in a table below:
| Phase | Actionable Recommendation | Rationale |
|---|---|---|
| 1. Mindset Shift | Stop seeking one 'set and forget' campaign. Embrace a portfolio approach of continuous testing and management. | The algorithm's design means all prospecting audiences will eventually experience performance decay. A portfolio approach builds resilience. |
| 2. Strategic Foundation | Calculate your Customer Lifetime Value (LTV) to define your maximum allowable CPA. | This moves your decision-making from being reactive and emotional to being data-driven and focused on long-term profitable growth. |
| 3. Creative Development | Develop 3-5 new ad creatives based on different customer pain points using frameworks like Before-After-Bridge. | A new message can unlock a completely new "pocket" of buyers within the same broad audience that your old creative couldn't reach. |
| 4. Audience Diversification | Launch a new CBO prospecting campaign testing a 1% Lookalike of your initial purchasers against your best new creatives. | Actively seeks out new high-intent users based on proven data, rather than waiting for the broad campaign to completly die. |
| 5. Funnel Completion | Create a simple BoFu (Bottom of Funnel) retargeting campaign for all website visitors from the last 30 days, excluding purchasers. | This is the easiest way to recover lost sales and significantly increase your overall Return on Ad Spend (ROAS). It's your highest-leverage action. |
I know this is a lot to take in, but I hope it sheds some light on what's really happening inside your ad account. Moving from that single, frustrating campaign to a robust system like this is often the leap that separates businesses that stall from those that scale succesfully.
Implementing this kind of structure, doing the research, and creating the right ads can definitly be a significant amount of work. This is where getting expert help can make a huge difference, as we can build and manage these systems far more quickly based on years of experience.
If you'd like to chat through your specific situation in more detail, we offer a free, no-obligation initial consultation where we can look at your account together and map out a more tailored plan. Feel free to book one in if you think it would be helpful.
Hope this helps!
Regards,
Team @ Lukas Holschuh
