Hi there,
Thanks for reaching out! Happy to give you some of my initial thoughts on the new setup. It sounds like you're in a common spot – you've found something that works but hitting a wall when you try to scale up. It's a good problem to have, means the basics are there. Let's get into how you can break through that ceiling.
TLDR;
- Your core problem isn't the switch to CBO, it's the lack of a structured advertising funnel. Simply grouping your 'best' creatives won't work for scaling.
- You need to split your campaigns into Prospecting (ToFu), Retargeting (MoFu/BoFu), and potentially Customer Retention. This lets you control the message and budget for different audience temperatures.
- Don't judge performance after 15 hours. CBO and Meta's algorithm need 3-5 days minimum to exit the learning phase and stabilise. Cutting ads too early is a huge mistake.
- Before cutting a product like Product B, diagnose the *entire* customer journey from ad click to purchase. Use the interactive diagnostic calculator in this letter to find the real drop-off point.
- This letter includes several interactive calculators and diagrams to help you visualise a proper scaling structure and calculate key metrics like LTV, which is the real key to profitable growth.
We'll need to look at your campaign structure... not just the budget setting
Right, first things first. Moving to CBO (Campaign Budget Optimization) is the right move for scaling, so good on you for trying it. But it's not a magic switch. You've basically given Meta the keys to the car and told it to "drive," but you haven't given it a map. That's why it's spending weirdly and you're not seeing the results you want just yet.
Your main question about judging performance after 15 hours is easy to answer: no, it's absolutley not enough time. The Meta algorithm has what's called a "learning phase". During this time, it's experimenting like a mad scientist, testing who to show your ads to, what time of day, on what placement, etc. This phase is messy and performance will be all over the place. It usualy needs about 50 conversions *per ad set* within a 7-day period to exit this phase. Until then, the data is basically noise. You need to give it at least 3-5 days before you even think about making a decision. Pulling the plug early just resets the learning phase and you're back to square one.
And about the purchase lag in Ads Manager – yes, it's pretty normal. Especially with recent privacy changes like iOS14, atribution can be delayed by up to 72 hours. As long as your test events are firing in the Events Manager, I wouldn't worry too much about it for now. Focus on the bigger picture.
Day 1-3: Learning Phase
Performance is volatile. CPA/ROAS will fluctuate wildly. Don't touch anything!
Day 4-7: Optimisation
The algorithm starts finding pockets of performance. Results begin to stabilise.
Day 7+: Scalable Performance
Consistent results. Now you can make informed decisions about budget and creatives.
I'd say you need to structure your account like a proper sales funnel
The real issue here is your strategy. Lumping your best creatives into one CBO campaign is like throwing all your best sales reps into a room and telling them to "sell" without telling them who they're talking to. Some are talking to ice-cold leads, some to people ready to buy, and it's just chaos.
To scale an eCommerce store, you need to think in terms of a funnel. You have different audiences with different levels of awareness of your brand, and you need to talk to them differently. We structure this for pretty much every eCommerce client we work with, and it's fundamental.
Here's how I'd prioritise it:
1. Top of Funnel (ToFu) - Prospecting Campaign: This campaign's only job is to find NEW customers. People who've never heard of you. Here, you'll test your broad and interest-based audiences. Once you have enough data (at least a few hundred purchases), you'll start testing Lookalike audiences. This is where you put the majority of your budget to grow.
2. Middle of Funnel (MoFu) - Warm Retargeting Campaign: This is for people who've shown some interest but haven't got close to buying. Think website visitors, video viewers, Instagram engagers. Your ads here should be about building more trust, showing social proof (reviews, UGC), and highlighting benefits. You exclude anyone who has added to cart or purchased.
3. Bottom of Funnel (BoFu) - Hot Retargeting Campaign: This is your highest ROAS campaign. It targets people who have added to cart or initiated checkout but not purchased in the last, say, 7-14 days. The ads here are direct. "Still thinking about it?", "Forgot something?", maybe a small discount to get them over the line. This is where you close the deals.
By splitting it this way, you give CBO clear jobs to do within each campaign. The prospecting campaign can focus on finding cheap new traffic, while the BoFu campaign can focus on efficiently converting high-intent users. I remember one women's apparel client where we implemented this structure; their overall ROAS jumped to nearly 700% because we stopped showing cold prospecting ads to people who had already abandoned their cart.
Top of Funnel (ToFu): Prospecting
Audience: Interests, Lookalikes (from purchasers), Broad
Goal: Find new potential customers who have never heard of you.
Budget: ~70-80% of total spend
Middle of Funnel (MoFu): Engagement
Audience: Website Visitors, Video Viewers, Social Engagers (Excluding Purchasers & Add-to-Carts)
Goal: Build trust and educate.
Budget: ~10-15% of total spend
Bottom of Funnel (BoFu): Conversion
Audience: Add to Cart / Initiated Checkout (last 7-14 days)
Goal: Close the sale.
Budget: ~5-10% of total spend
You probably should diagnose product performance before killing it
Now, let's talk about Product B. You've spent £90 with no sales. Instinct says to kill it. My experience says to diagnose it first. Where are people dropping off?
You need to look at the mini-funnel for that specific product's ads:
- Impressions -> Clicks (CTR): Is the CTR decent (say, over 1%)? If not, your ad creative or copy isn't grabbing attention. It's an ad problem.
- Clicks -> Landing Page Views: Are people actually landing on the page? A big drop-off here could mean your site is too slow.
- Landing Page Views -> Adds to Cart: Are people seeing the product but not adding it to their cart? This is the most likely culprit. It could be your product photos, the description, the price, or a lack of trust signals (reviews!). - Adds to Cart -> Purchase: Are they adding to cart but not buying? This often points to unexpected shipping costs or a complicated checkout process.
Spending £90 and getting zero sales is very different if you got 100 clicks and 0 adds-to-cart versus getting 20 clicks and 0 adds-to-cart. The first sugests a product/page problem, the second an ad problem. You can't make a good decision without knowing this. Don't just kill the ad set, be a detective.
To help with this, I've built a little diagnostic tool for you. Plug in your numbers for Product B and see where the biggest cost jump is. This will tell you exactly where the "leak" in your funnel is.
Funnel Cost Analysis
You'll need to understand what scaling really costs
This is probably the most important shift in mindset you'll need. You can't scale from 3 sales a day to 30 sales a day and expect to keep a 4-5x ROAS. It just doesn't happen. As you increase spend, you force Meta to find people who are less and less likely to buy. Your costs will go up. That's a fact of life. We had one eCommerce client who couldn't get past a certain daily spend without their CPA skyrocketing. The solution wasn't just in the ads; it was in improving their entire business model.
The real question isn't "How do I keep my ROAS high?" but "How high a Cost Per Purchase can I afford while still being profitable?". The answer is your Customer Lifetime Value (LTV).
For an eCommerce brand, a simple way to look at LTV is:
LTV = (Average Order Value x Purchase Frequency x Customer Lifespan) x Profit Margin
If you know a customer is worth £500 to you over their lifetime, then paying £50 to acquire them (a 10x LTV:CAC ratio) is a fantastic deal, even if the ROAS on that first purchase was only 2x. This is the maths that unlocks aggressive, intelligent growth. Without knowing your LTV, you're flying blind and making decisions based on short-term metrics that can stunt your growth.
I've put another calculator below for you. Play around with it. See how small improvements in how often people buy from you again, or how much they spend each time, can dramatically increase how much you can afford to spend on ads. This is the engine of a truly scalable business.
So, what's the plan?
There's a lot to take in here, I know. It's a shift from just running ads to building a proper marketing system. The good news is that you've already got profitable products, which is the hardest part. Now it's about building the machine around them.
I've detailed my main recommendations for you below in a table to make it a bit clearer. This is the roadmap I'd suggest you follow to get from where you are now to a place where you can actualy scale your sales volume predictably.
| Area of Focus | Problem | Recommended Action |
|---|---|---|
| Patience & Data | Making decisions too quickly (e.g., after 15 hours) based on volatile data. | Commit to letting any new campaign or ad set run for at least 5 days before judging its performance to allow it to exit the learning phase. |
| Campaign Structure | A single campaign trying to do everything (find new customers and convert warm leads). | Immediately restructure into three separate CBO campaigns: 1. Prospecting (ToFu), 2. Warm Retargeting (MoFu), and 3. Hot Retargeting (BoFu). Allocate budget accordingly (e.g. 80/10/10 split). |
| Product B Analysis | Killing a potentially good product based on top-level metrics (spend vs sales). | Use the diagnostic process outlined above. Analyse the funnel from CTR to ATC to Purchase to find the real drop-off point. Fix that leak before killing the ads. |
| Scaling Mindset | Focusing solely on ROAS, which will inevitably decrease as you scale. | Calculate your Customer Lifetime Value (LTV). Shift your primary success metric from ROAS to your LTV to Customer Acquisition Cost (CAC) ratio. Aim for at least a 3:1 ratio. |
Implementing this stuff takes time and it can feel overwhelming. You go from tweaking a couple of ads to managing a complex system with lots of moving parts. This is often the point where founders decide their time is better spent working on their products and business, and they bring in someone with expertise to build and manage the advertising system for them.
We've helped lots of eCommerce stores make this exact transition, achieving results like a 633% return for a cleaning products company and a 691% return for a women's apparel brand. If you'd like to have a chat and go through your account on a call, we offer a free initial consultation. We could dive into your specific numbers and build out a more concrete plan for you.
Either way, hope this has been genuinely helpful and given you a new way to think about growing your store.
Regards,
Team @ Lukas Holschuh
