Hi there,
Thanks for reaching out. I saw your post about structuring campaigns for your eCommerce store and thought I'd give you some of my thoughts. It's a really common problem, especially for clothing brands with multiple collections, so don't worry, you're not alone in struggling with it. What you're experiencing with CBO favouring one ad set is exactly what it's designed to do – find the lowest cost results and pour the budget there. That's great if you only have one goal, but a nightmare when you're trying to shift inventory across different product lines.
I've managed quite a few campaigns for eCommerce brands, including apparel, and have spent a lot of time figuring out the best way to structure accounts to avoid this exact issue while still getting a good return. I'll walk you through how I'd approach it, based on what's worked for us.
We'll need to look at your campaign structure...
First off, your question about creating a seperate CBO campaign for each collection. Can you do it? Yes. Should you? Probably not. It's a bit of a plaster on a bigger problem.
The main issue, as you guessed, is audience overlap. Let's say you create one campaign for your Jeans collection and another for your T-shirts collection. You set a budget for each. Great. But who are you targeting? Likely, you're targeting very similar, if not identical, audiences for both. Maybe it's women aged 20-35 who are interested in fashion, online shopping, and brands like ASOS or Zara.
When you have two seperate campaigns targeting the same pool of people, you are essentially bidding against yourself in the ad auction. Your 'Jeans' campaign and your 'T-shirts' campaign are both trying to show an ad to the same person. This drives up the cost for both campaigns (your CPMs will rise) and leads to lower overall efficiency. Meta's system gets confused about which campaign's performance data to prioritise, and your audience gets spammed with ads from the same brand, which can lead to ad fatigue much faster.
While it gives you the budget control you want, it often comes at the expense of profitability. It's a short-term fix that creates long-term problems and makes it very difficult to scale your ad spend effectively. There is a much more robust way to set up your account that lets you promote all your collections, control budgets where it matters, and avoids you competing with yourself. It's all about thinking in terms of a customer journey, or a funnel, rather than thinking in terms of product collections.
I'd say you need to switch to a funnel-based approach...
This is probably the single biggest change you can make. Instead of campaigns for 'Jeans', 'Hoodies', etc., you should structure your account based on the temperature of your audience. We call it a ToFu/MoFu/BoFu funnel.
-> ToFu (Top of Funnel): Prospecting
This is your 'cold' audience. These are people who have never heard of your brand before. The goal of your ToFu campaign is to find new potential customers and introduce them to your store. This campaign will have the largest audiences and will likely get the biggest chunk of your budget. This is where you test different interests, behaviours, and lookalike audiences to see who responds best to your brand.
-> MoFu (Middle of Funnel): Re-engagement
This is your 'warm' audience. These are people who have shown some interest. They've visited your website, watched one of your video ads, engaged with your Instagram page, but they haven't taken that next step like adding a product to their cart. The goal of the MoFu campaign is to bring these people back and get them to look at your products more closely.
-> BoFu (Bottom of Funnel): Retargeting
This is your 'hot' audience. These people are on the verge of buying. They've added items to their cart, they've started the checkout process, maybe they even added payment info but got distracted. These are your most valuable audiences, and your BoFu campaign is designed to give them that final nudge to complete the purchase. The return on ad spend (ROAS) here is usually the highest.
By structuring your account this way, with three main long-term campaigns (one for each funnel stage), you are aligning your strategy with how people actually shop online. You can set a CBO budget for each of these campaigns, giving you control over how much you want to spend on finding new customers versus retargeting existing ones. The audiences for each campaign are naturally exclusive (e.g., your BoFu campaign will only target people who added to cart, and you'll exclude purchasers), so the self-competition problem disappears. Tbh, this is the standard for any serious eCommerce store.
I remember one client we worked with who sold women's apparel. They had a messy structure with lots of campaigns for different products. We moved them over to this exact funnel structure and saw a 691% return on ad spend. It works because it lets Meta's algorithm do its job properly within a logical framework.
You probably should focus on who you're targeting...
Once you have the funnel structure, the next question is who to put in it. Getting your audiences right is more important than almost anything else. A great ad shown to the wrong person will do nothing. Here's how I'd prioritise audiences for your clothing store, working from the top of the funnel down.
For a new account, you'd start with detailed targeting to gather data. As soon as you have enough data (you need at least 100 people in an audience to use it, but really you want more like 1,000+ for it to be effective), you can build retargeting and lookalike audiences. You should test these in order of priority. An audience further down the funnel is almost always going to perform better.
Here's a breakdown of the audiences I'd prioritise. You can basically copy this structure.
| Funnel Stage | Audience Type | Specific Examples for Your Store |
|---|---|---|
| ToFu (Prospecting) | 1. Detailed Targeting (Interests, Behaviours) |
Start here. Don't just target "Jeans" or "Fashion". Get specific. Think about: -> Competitor Brands: People who like Zara, H&M, Levi's, etc. -> Fashion Publications: People interested in Vogue, Elle, Who What Wear. -> Related Interests: Online shopping, specific fashion styles (e.g., streetwear, minimalist fashion). -> Layering: Combine an interest like 'Online Shopping' with a demographic like 'Women aged 25-34'. |
| 2. Lookalike Audiences | Once you have sales data, these are gold. Create them in this order of priority: 1. Lookalike of your highest value customers. 2. Lookalike of all purchasers. 3. Lookalike of people who initiated checkout. 4. Lookalike of people who added to cart. 5. Lookalike of all website visitors. |
|
| MoFu (Re-engagement) | Warm Custom Audiences | Target people who have shown interest but not converted. Remember to exclude purchasers and people who added to cart. -> All website visitors (past 30-60 days). -> People who viewed 50% of your video ads. -> People who engaged with your Instagram/Facebook page. |
| BoFu (Retargeting) | Hot Custom Audiences | Your lowest-hanging fruit. Target these with urgency and offers. -> Added to Cart (past 7-14 days) but didn't buy. -> Initiated Checkout (past 7-14 days) but didn't buy. |
| Past Customers | Don't forget them! It's cheaper to get a repeat purchase than a new customer. -> All purchasers (past 180 days). Target them with new collections or special offers. -> Cross-sell: Target people who bought jeans with ads for t-shirts. |
When you're building interest audiences, the key is to be specific. A common mistake is picking an interest that's too broad. For instance, if you were selling eCommerce software, targeting the interest "Amazon" would be a waste of money because you'd reach millions of shoppers, not store owners. For you, targeting the interest "clothing" is similarly useless. You need to target interests that a person in your target audience is much more likely to have than the general population. That's the secret.
You'll need to structure your ads and creatives properly...
So how does this all solve your original problem of promoting specific collections? This happens at the Ad Set and Ad level, within your new funnel campaigns.
Inside your ToFu Prospecting Campaign:
Here you can have multiple ad sets, each targeting a different cold audience (e.g., one for a Lookalike of Purchasers, one for an interest group). Inside each ad set, you can run ads for your different collections. This is where Dynamic Creative (DCO), which you're already using, is brilliant. You can give it:
- 5 different images/videos (e.g., a lifestyle shot of the jeans, a flat lay of the t-shirts, a video of the hoodies).
- 5 different headlines (e.g., "Discover Your Perfect Fit," "New Season Hoodies Have Arrived").
- 5 different primary texts.
Meta will then mix and match them to create hundreds of combinations and automatically figure out which ad works best for which person within that audience. It might show the jeans ad to one person and the t-shirt ad to another. This lets you promote everything at once, while letting the algorithm do the heavy lifting of optimising.
Inside your MoFu & BoFu Retargeting Campaigns:
This is where it gets even more powerful. For these audiences, you should be using Dynamic Product Ads (DPA). This requires setting up a product catalog and connecting it to your Meta account. Once it's set up, you can run a DPA campaign that automatically shows people the exact products they looked at on your website. Someone viewed a specific pair of blue skinny jeans and a black hoodie? The ad will show them exactly those two items, reminding them to come back and buy. This is highly personalised and incredibly effective for eCommerce.
Alongside DPAs, you can also run standard retargeting ads in your BoFu campaign. For your 'Added to Cart' audience, you could run an ad with copy like "Did you forget something? Your cart is waiting" and offer a small discount like free shipping to push them over the edge. These simple tactics can make a huge diffrence to your conversion rate.
This is the main advice I have for you:
I know this is a lot to take in, so I've put the main points into a table to summarise the actionable strategy. This is the blueprint I would use if I were starting on your account tomorrow.
| Area of Focus | My Recommendation | Why This Is Important |
|---|---|---|
| Account Structure | Immediately stop creating campaigns per collection. Rebuild your account into three core CBO campaigns: ToFu (Prospecting), MoFu (Re-engagement), and BoFu (Retargeting). | This aligns your ad spend with the customer journey, eliminates audience overlap between campaigns, and allows Meta's algorithm to optimise far more effectively. It's the foundation for scaling profitably. |
| Budget Allocation | Allocate budget using CBO at the campaign level. A good starting point is 70% ToFu, 10% MoFu, 20% BoFu. Adjust based on performance. | This ensures you are consistently feeding the funnel with new potential customers while dedicating enough budget to convert your warmest audiences at a high ROAS. |
| ToFu Strategy | Inside your ToFu campaign, create ad sets for your best Lookalike Audiences and highly specific Detailed Targeting groups. Use Dynamic Creative (DCO) to test all your collections within these ad sets. | This lets you find new customers at scale. DCO automates the testing of which collection resonates with which cold audience, solving your original problem without fragmenting your campaigns. |
| Retargeting Strategy | Set up a product catalog immediately. Your BoFu campaign should heavily feature Dynamic Product Ads (DPAs) targeting 'Add to Cart' and 'View Content' audiences. Exclude purchasers. | DPAs are the most powerful tool for eCommerce retargeting. They are automated, highly relevant, and will recover a significant number of what would have been lost sales. This is where your highest ROAS will come from. |
| Measurement | Shift your primary success metric from just 'cost' to Return on Ad Spend (ROAS). Track your ROAS separately for ToFu, MoFu, and BoFu. | Focusing on cost alone is misleading. A £50 cost per purchase might seem high, but if that customer spent £300, it's a 6x ROAS which is fantastic. Analysing ROAS by funnel stage tells you where your money is working hardest. |
Implementing a structure like this isn't just a one-time setup. It requires ongoing management – monitoring which audiences and creatives are working, turning off the losers, scaling the winners, and constantly testing new ideas to feed into the system. You have to be realistic results wise; for eCommerce in a developed country, a cost per purchase can be anywhere from £10 to £75 or more. The goal isn't to get the cheapest sales, but the most profitable ones.
It can feel a bit overwhelming, and this is often the point where businesses decide they need an expert pair of hands. A good agency or consultant doesn't just set up ads; they build and manage this entire strategic system for you. They bring the experience of what audiences to test, what creative styles work, and how to interpret the data to make decisions that grow your revenue. We've helped eCommerce clients who were struggling with messy accounts achieve 633% return, and even 8x return, on their ad spend by implementing and optimising this exact kind of funnel.
Hopefully this detailed breakdown gives you a much clearer path forward. If you go through this and decide you'd rather have someone handle the complexities for you, we offer a free initial consultation. We can have a look at your account together and discuss how a robust strategy like this could be implemented for your brand specifically.
Hope this helps!
Regards,
Team @ Lukas Holschuh