Hi there,
Thanks for reaching out. The situation you described is one I've seen play out dozens of times. It's incredibly frustrating when a winning campaign suddenly falls off a cliff, especially when you try to scale it. It feels random, but it's usually a sign that you've hit a natural limit with your current approach.
I'm happy to give you some of my initial thoughts and guidance on this. The short answer is that simply duplicating adsets isn't a sustainable scaling strategy. It works for a bit, but it eventually breaks the system. What you need is a more robust structure that lets you scale deliberately without torching your profitability. Let's get into it.
TLDR;
- Your campaigns stopped working because you hit audience saturation. Duplicating campaigns with more budget just forces Facebook to find more expensive, lower-quality users.
- Rapidly duplicating campaigns and changing budgets also throws the algorithm back into the 'learning phase' over and over, creating instability and poor performance.
- The most important piece of advice is to stop focusing on daily ROAS and start calculating your Customer Lifetime Value (LTV). This tells you how much you can actually afford to spend to acquire a customer, which is the real key to scaling.
- You need to move from a single adset approach to a structured ToFu/MoFu/BoFu (Top/Middle/Bottom of Funnel) campaign model. This lets you target different audience temperatures with different messages and scale more predictably.
- This letter includes an interactive LTV calculator to help you find your real acquisition cost targets and a flowchart visualising the proper campaign structure.
We'll need to look at why 'just duplicating' campaigns is a trap...
Okay, so first things first. The method you used—duplicating a winning adset into a CBO to scale—is a common tactic. And it works, for a while. But it has a built-in expiry date. What you experienced isn't a glitch; it's the system working as designed, and you've just hit its limits. There's a couple of things going on here.
1. Audience Saturation and the "Low-Hanging Fruit" Problem
Think of your target audience on Facebook like an orchard. When you first launch your adset with a modest budget, the algorithm is smart. It goes and picks all the easiest, lowest-hanging fruit first. These are the people within your targeting who are most likely to convert, right now, for the cheapest cost. Your ABO adset was happily feasting on this segment for weeks.
When you duplicated it and added more budget, you told Facebook, "Go get me more fruit, and do it faster." The algorithm, having already picked the easy stuff, is forced to bring out a ladder. It starts going after the people higher up in the tree. They are harder to reach, more expensive to persuade, and less likely to buy. Your ROAS drops from 3 to 2. This is the cost of the ladder.
When you duplicated it a third time, you basically told the algorithm to get a cherry picker and go after the toughest, most expensive fruit at the very top. At this point, the cost to acquire a customer skyrockets, and your ROAS plummets. You start breaking even or losing money. You've exhausted the profitable segment of your audience at that specific moment in time.
It's a common misconception that if an audience has 10 million people, you can just keep pumping money in. The reality is that within that 10 million, there's probably only a small 'hot pocket' of a few hundred thousand who are actually good prospects for you. You burned through them.
2. The Dreaded 'Learning Phase' Reset
Every time you make a significant edit to a campaign or adset—like duplicating it or changing the budget dramatically—you risk pushing it back into the Learning Phase. During this phase, the algorithm is spending your money to figure out who to show your ads to. Performance is volatile and unpredictable. By constantly duplicating, you were likely creating multiple, competing adsets that were all in a state of flux, never allowing the pixel to gain real, stable learnings at a higher spend level. It's like trying to build three different foundations for the same house at once; none of them get properly set, and the whole structure is weak.
Tbh, you're essentially telling Facebook to find you the worst people. When you optimise for conversions, the algorithm tries to find buyers. But when you force it to spend more than the available pool of buyers can support, it has to lower its standards. It starts showing your ad to people who are *less* likely to buy, simply to fulfil your budget command. You're paying a premium for lower-quality impressions.
I'd say you need to know your real numbers before you scale...
Before you even think about relaunching a campaign, you need to step back from the daily ROAS metric. It's a useful health check, but it doesn't tell you the most important thing: How much can you actually afford to pay to acquire a customer?
The answer to that is your Customer Lifetime Value (LTV). This is the total profit you can expect to make from a single customer over the entire time they do business with you. Once you know this number, scaling becomes a simple maths problem, not a guessing game. A healthy business can typically afford to spend about one-third of its LTV to acquire a customer. This is your target Customer Acquisition Cost (CAC).
Let's break down the calculation:
- Average Revenue Per Account (ARPA): What's the average amount a customer spends with you per month (or year)?
- Gross Margin %: What's your profit margin on that revenue after accounting for cost of goods sold?
- Monthly Churn Rate %: What percentage of your customers do you lose each month?
The formula is: LTV = (ARPA * Gross Margin %) / Monthly Churn Rate
This single number changes everything. If you find out your LTV is £500, then you know you can afford to spend up to £166 (a 3:1 LTV:CAC ratio) to get a new customer and still run a very profitable business. Suddenly, a campaign that's "breaking even" on the first purchase might actually be incredibly profitable in the long run. This is the mindset shift required for serious scaling.
Use the calculator below to get a feel for your own numbers. This isn't just a theoretical exercise; this is the bedrock of a scalable advertising strategy.
You probably should rethink your audience and campaign structure...
So, how do we use this new LTV/CAC target to build campaigns that can actually scale? We stop treating all audiences the same. We need to build a proper marketing funnel inside of your ad account. The classic model is Top of Funnel (ToFu), Middle of Funnel (MoFu), and Bottom of Funnel (BoFu).
This structure allows you to speak to people differently based on how familiar they are with your brand, and it creates a perpetual motion machine: you attract new people at the top, warm them up in the middle, and convert them at the bottom. It also stops your adsets competing with each other for the same users.
- ToFu (Top of Funnel - Cold Audiences): This is where you find new people who have never heard of you. Your goal here isn't necessarily immediate profit. It's to identify and attract potential customers at an acceptable cost. This is where you'll test your broad, interest-based, and Lookalike audiences. This campaign should get the bulk of your budget (maybe 70-80%).
- MoFu (Middle of Funnel - Warm Audiences): These are people who have shown some interest but haven't taken a key action yet. They've watched your videos, engaged with your ads, or visited your website but didn't add to cart. Your goal here is to build trust and re-engage them, reminding them of your value. This campaign gets a smaller slice of the budget (10-15%).
- BoFu (Bottom of Funnel - Hot Audiences): These are your hottest prospects. They've added a product to their cart, initiated checkout, or visited a key page multiple times. They are on the verge of buying. Your goal here is to get them over the finish line. These are your highest ROAS campaigns, but they have the smallest audience size. This gets the last 5-10% of your budget.
By splitting your campaigns like this, you can scale the ToFu budget to bring new people into your ecosystem, knowing that your MoFu and BoFu campaigns are there to efficiently convert them over time. You stop relying on one adset to do everything and instead build a system where each part has a specific job. This is how you can spend £1000 a day or more, because you're not trying to force a cold audience to convert immediately. You're building a relationship.
Here's a visual breakdown of how that structure might look:
1. Top of Funnel (ToFu)
Goal: Attract new, relevant eyeballs. Prospecting for potential customers.
- Lookalikes (1-5%): Based on Purchasers, High LTV customers
- Detailed Targeting: Interests, Behaviours relevant to your ICP
- Broad Targeting: (Once pixel is very mature)
2. Middle of Funnel (MoFu)
Goal: Re-engage and build trust with people who have shown interest.
- Website Visitors (Last 30 days, excl. purchasers)
- Social Engagers (Last 90 days, FB & IG)
- Video Viewers (Watched 75%+ of an ad)
3. Bottom of Funnel (BoFu)
Goal: Convert high-intent prospects into customers.
- Added to Cart (Last 7-14 days, excl. purchasers)
- Initiated Checkout (Last 7-14 days, excl. purchasers)
- Viewed Product (Last 3 days, excl. cart)
You'll need a different message for each audience...
A huge mistake people make is showing the same ad to everyone. A cold prospect needs to understand the problem you solve, while a hot prospect just needs a final nudge, like a reminder about free shipping or a testimonial.
Your creative and copy needs to match the audience's "temperature".
- ToFu Ads: These should be educational or entertaining. They need to stop the scroll and introduce a problem the user has, then present your product as the solution. Use broad appeal messages. The 'Problem-Agitate-Solve' formula works wonders here.
- MoFu Ads: Here, you can be more direct. Show customer testimonials, handle common objections, or explain key features and benefits. You're building trust and demonstrating why you're the best choice.
- BoFu Ads: These should be direct response ads. Remind them of the product they looked at (using Dynamic Product Ads), offer a small incentive if applicable, and use urgency to encourage them to complete their purchase.
Creative fatigue is also a real thing. Your ads, even the winning one, will stop working over time as the same audience sees it again and again. You must constantly be testing new images, videos, and headlines to keep performance strong. For instance, I remember several SaaS clients we worked with saw really good results from simply introducing some User-Generated Content (UGC) style videos into their ToFu campaigns. It feels more authentic and cuts through the noise.
Here's how that might look in practice:
| Funnel Stage | Ad Copy Angle | Example (for a hypothetical high-quality coffee subscription) |
|---|---|---|
| ToFu (Cold) | Problem-Agitate-Solve | "Tired of that bitter supermarket coffee? That morning dread of another bland cup? Imagine a perfect brew, every single morning. We deliver award-winning, freshly roasted beans right to your door." |
| MoFu (Warm) | Testimonial / Social Proof | "Over 5,000 coffee lovers gave us 5 stars. See why Sarah from Manchester calls it 'the best part of her day'. Our unique roasting process locks in flavour you won't find anywhere else." |
| BoFu (Hot) | Direct Offer / Urgency | "Did you forget something? The Ethiopian Yirgacheffe you were looking at is one of our most popular. Complete your order now and get free shipping. Don't miss out." |
I've detailed my main recommendations for you below:
Okay, that was a lot of information to take in. The key thing to realise is that you haven't failed; you've just outgrown your old strategy. To get back to profitability and scale beyond it, you need a more mature system. Here's a table summarising the actionable plan I'd recommend.
| Step | Action | Rationale |
|---|---|---|
| 1. Immediate Pause | Turn off all your current, underperforming campaigns. | Stop burning cash on a broken strategy. Give yourself a clean slate to rebuild from. |
| 2. Calculate Your Numbers | Use your business data (and the calculator above) to determine your LTV and a realistic target CAC. | This provides you with a North Star metric for success. You'll know exactly what performance you need to be profitable. |
| 3. Rebuild with a Funnel Structure | Create three new, separate CBO campaigns: ToFu, MoFu, and BoFu. | This organizes your account for scale, prevents audience overlap, and allows you to tailor your messaging effectively. |
| 4. Populate Audiences | Create and assign the relevant audiences to each campaign (Interests/Lookalikes for ToFu; Engagers/Visitors for MoFu; Cart Abandoners for BoFu). | Ensures you are targeting the right people at the right stage of their journey. |
| 5. Launch & Optimise | Allocate your budget (e.g., 70% ToFu, 20% MoFu, 10% BoFu) and launch. Let campaigns run for 5-7 days to exit the learning phase before making major changes. | This provides the algorithm with the stability it needs to find converters and gives you reliable data to make optimisation decisions. |
This is a fundamental shift from finding "hot pockets" to building a predictable customer acquisition engine. It requires more work upfront, but it's the only way to scale your ad spend without your ROAS falling through the floor. It takes the guesswork out of the equation and replaces it with a system.
Implementing a structure like this can be a bit daunting, and getting the audiences, budget allocation, and creative strategy right for each stage takes experience. It's a lot more involved than just duplicating an adset, and there are lots of small details that can make a huge difference to the final result.
If you'd like an expert hand in auditing your old campaigns and building out this new, scalable structure for you, we offer a free, no-obligation initial consultation. We can go through your account together on a call and map out a specific action plan tailored to your business. It's often the quickest way to get back on track and start growing again.
Hope that helps!
Regards,
Team @ Lukas Holschuh
