Hi there,
Thanks for reaching out! Happy to give you some initial thoughts on your question. It's a common one for clothing brands with a decent sized catalogue, and tbh, the way most people approach it is a surefire way to burn through cash with very little to show for it.
Your instinct to just run a full catalog campaign and let the algorithm sort it out is tempting, I get it. It seems easy. But in my experience, it's usually the wrong move, especially when you're starting out. You end up spreading your budget so thin that nothing gets real traction, and the algorithm often just picks a random shirt that gets a few cheap likes but never actually sells. You're left scratching your head wondering what went wrong.
The better attack plan is more strategic. It's about being deliberate, testing smart, and scaling what's proven to work, not just what you hope will work. Let's get into it.
TLDR;
- Don't start with a full catalog campaign. It spreads your budget too thin and often optimises for low-quality engagement, not sales. You're essentially paying Meta to find non-customers.
- Before spending a penny, define your customer by their identity and 'pain points', not just demographics. Your designs solve a need for self-expression or belonging. Figure that out first.
- Group your 60+ products into 3-5 thematic collections. Pick one 'hero product' from your top 2-3 collections to test with dedicated creatives. This focuses your budget and gives you clear data.
- The most important piece of advice is to structure your initial campaigns to test these hero products against specific, niche audiences. Only after you find a winning product-audience combination should you scale and introduce broader catalog ads for that specific winning collection.
- This letter includes an interactive POD Profitability Calculator to help you understand your numbers and a flowchart to help you map out your customer's identity.
You probably should forget about demographics... for now
Right, first things first. Forget "men aged 18-35 who like streetwear". That tells you absolutely nothing useful. It's the kind of generic profile that leads to generic ads that get ignored. You're in the identity business. People don't buy a t-shirt, they buy a piece of who they are, or who they want to be. Your first job is to become an expert in that identity.
Your Ideal Customer Profile (ICP) isn't a demographic; it's a problem state, a belief system, a subculture. The 'pain' you're solving isn't a lack of clothes. It's the feeling of not being seen, of not belonging, of wanting to signal to the world "this is me". Your customer isn't just a job title; they're someone who is frustrated that no other brand 'gets' their specific sense of humour, their passion for a niche hobby, or their particular aesthetic.
Once you nail that identity, everything else gets easier. Targeting isn't guesswork anymore. You know the niche subreddits they're in, the specific influencers they follow on TikTok (not the massive ones, the niche ones), the bands they listen to, the other small indie brands they buy from. This intelligence is the foundation of your entire ad strategy. You have to do this work first, or you've no business spending a single pound on ads.
Core Identity
What tribe/subculture does your brand represent? (e.g., retro gamers, minimalist architects, sarcastic cat lovers)
Their 'Pain'
What frustration do they have? (e.g., "Mainstream merch is cringe," "I can't find designs that match my aesthetic")
Watering Holes
Where do they hang out online? (e.g., Specific subreddits, niche Instagram accounts, Discord servers)
Their Language
What slang/in-jokes do they use? (Your ad copy needs to speak this language)
I'd say you need to engineer a winner, not guess one
Okay, so you have 60+ products. That's too many to test properly. Your next step is to curate. Instead of seeing 20 shirts, 20 sweatshirts, etc., I want you to see 3-5 distinct 'collections' based on the ICP work you just did. A collection is a group of products that share a theme, an aesthetic, or a message.
For example, you might have:
- -> The 'Dark Humour' Collection: Shirts and hats with sarcastic, witty slogans.
- -> The '8-Bit Nostalgia' Collection: Designs referencing classic video games.
- -> The 'Architectural Lines' Collection: Minimalist, geometric designs on sweatshirts.
This immediately makes your catalogue manageable. Now, from each of your top 2 or 3 collections, you're going to pick ONE 'hero product'. This is the product you believe best represents the entire collection's vibe and has the broadest appeal within your target niche. This is the product you will build your initial creatives for. Not 20, but 2 or 3. This focuses your creative energy and your ad spend on the highest-potential items.
You're not guessing. You're making an educated hypothesis based on your deep understanding of your customer. Which design perfectly captures that 'in-joke'? Which one would make someone in your tribe stop scrolling and say "that is so me"? That's your hero.
| Thematic Collection | Chosen Hero Product | Primary Audience (ICP) | Core Ad Message/Angle |
|---|---|---|---|
| Example: Dark Humour | "I Need Coffee" Skeleton Sweatshirt | Millennials in creative/tech jobs, burnt out but self-aware, follow accounts like @TheOnion. | "Your official work-from-home uniform has arrived." |
| Example: 8-Bit Nostalgia | Pixelated Space Invader Hat | Gen X / Older Millennials who grew up with arcades, members of retro gaming forums. | "Wear your high score. If you know, you know." |
| Example: Architectural Lines | Minimalist Grid T-Shirt | Designers, architects, people who follow @dezeen, value clean aesthetics. | "A design so clean it'll organise your entire wardrobe." |
See how this forces clarity? Instead of a vague "let's sell some shirts", you now have a precise testing plan. You're not just testing products; you're testing messages to specific groups of people.
We'll need to look at a proper testing structure
With your hero products selected, now you can build your first campaign. But we're not just going to throw them all into one ad set. We need to structure it for learning. The goal of this first ~£500-£1000 in spend isn't necessarily profit; it's data. We need to find a winning combination of Creative + Audience.
I usually start with a simple CBO (Campaign Budget Optimization) campaign aimed at conversions (Purchases). Inside that campaign, I'd set up 2-3 ad sets. Each ad set will target a different audience hypothesis based on your ICP research.
- -> Ad Set 1: Interest Stack A. Group together 5-7 highly relevant interests. E.g., for the 'Dark Humour' collection: The Onion, McSweeney's, 'Black Mirror', 'Adult Swim', etc.
- -> Ad Set 2: Interest Stack B. A different group of interests. Maybe competitor brands, related authors, or niche publications. This tests a different angle on the same ICP.
- -> Ad Set 3: Broad Targeting (No Interests). If you have some purchase data on your pixel already (say, 50-100 sales), you can test a broad audience. If you're brand new, skip this for now.
Now, inside *each* of those ad sets, you put the ads for your 2-3 hero products. This means your 'Skeleton Sweatshirt' ad and your 'Pixelated Hat' ad will both run to Audience A and Audience B. This setup allows Meta to allocate budget to the best-performing audience, and within that audience, to the best-performing creative.
After 3-5 days, the data will start to tell you a story:
- High CTR, No Adds to Cart? -> Your creative is good, it's stopping the scroll. But something on your product page is wrong. Is the price too high? Are the product mockups poor quality? Is the description weak?
- Low CTR? -> The ad creative itself isn't resonating with that audience. The image is bad, or the copy doesn't speak their language. Time to test new creatives.
- Lots of Adds to Cart, Few Purchases? -> This is usually a shipping cost issue or a lack of trust at checkout. Do you have reviews? Trust badges? A clear returns policy?
You're not just looking at ROAS (Return On Ad Spend). You're being a detective, looking at the whole funnel to diagnose the problem. This is where most people give up. They see no sales and turn everything off. We see no sales and ask "why?", then go and fix the leak. I remember one campaign for a women's apparel brand where we achieved a 691% return on ad spend, and it was all down to this methodical approach of testing and fixing leaks in their funnel.
Understanding your numbers is absolutly vital here. A low ROAS might be fine if you're acquiring a customer who'll buy again, but you need to know your break-even point from day one. I've built a little calculator for you to play with.
You'll need to scale the winners (and kill the losers)
So, what happens when you find a winner? Let's say your 'Skeleton Sweatshirt' is getting a 3x ROAS with Audience A. Brilliant. Now it's time to scale. This is where you move from testing to growth mode.
Step 1: Isolate the Winner. You'll duplicate your successful testing campaign. In the new campaign, you'll turn off everything that didn't work. You'll have one campaign, with one ad set (Audience A), and one ad (The Skeleton Sweatshirt creative). Now you can start gradually increasing the budget on this campaign, maybe 20% every few days, as long as performance holds.
Step 2: Go Deeper on the Winning Theme. If the sweatshirt worked, there's a good chance other products in the 'Dark Humour' collection will too. Now is the time to introduce a Dynamic Product Ad (DPA) or catalog ad. But here's the key difference: you create a product set in your catalog that *only* contains the 'Dark Humour' collection. You're not showing your whole catalog; you're showing them more of what they've already shown interest in. This is so much more powerful and relevent.
Step 3: Build High-Quality Lookalikes. Once you have 100+ purchases of that sweatshirt, you can create a 1% Lookalike Audience of those buyers. This is pure gold. You're telling Meta, "go find me more people who look exactly like the people who already gave me money for this specific product." This will almost always outperform a broad interest-based audience.
Step 4: Don't Forget Retargeting. You need a separate, always-on campaign for your Middle and Bottom of Funnel (MoFu/BoFu). This targets people who have visited your site, added to cart, or initiated checkout but haven't purchased. Show them a different ad. Maybe with a review testimonial, a reminder of your returns policy, or even a small discount code to nudge them over the line.
Your question about a "full catalog campaign" wasn't wrong, it was just premature. You use it strategically to scale a *proven concept*, not to find one in the first place. Doing it this way round is the diffrence between a profitable clothing brand and one that shuts down after three months because "ads don't work".
My final advice to you
I've detailed my main recommendations for you below. This is the exact process we'd follow. It's about being strategic, methodical, and data-driven, rather than just throwing things at the wall and hoping something sticks.
| Phase | Action | Why It Works |
|---|---|---|
| 1. Strategy (Pre-Spend) | Define your ICP based on identity/subculture. Group your 60+ products into 3-5 thematic collections. Select one 'hero product' from each of your top 2-3 collections. | This focuses your budget and creative efforts on the highest potential products, preventing wasted spend and ensuring your messaging is sharp and relevant. |
| 2. Testing (First £500-£1k) | Run one CBO Conversion campaign. Create 2-3 ad sets targeting different interest stacks based on your ICP research. Place ads for all your hero products in each ad set. | This structure is designed for maximum learning. It quickly identifies the winning combination of audience and creative, telling you what to scale and what to kill. |
| 3. Analysis | Analyse the full funnel: CTR, Add to Cart rate, and Cost Per Purchase. Diagnose where users are dropping off. Don't just look at the final ROAS. | Finds the specific 'leaks' in your sales process (e.g., bad creative, high shipping, confusing product page) so you can fix the actual problem. |
| 4. Scaling | Isolate winning ad set/ad combos in a new campaign. Build Lookalike audiences from purchasers. Launch a separate retargeting campaign. Use catalog ads ONLY for the winning collection. | This methodically grows what's proven to work, using your own customer data to find more high-quality buyers, leading to profitable and sustainable growth. |
This all probably sounds like a lot of work, and tbh, it is. Running paid ads effectively for an eCommerce brand isn't a 'set it and forget it' task. It requires constant monitoring, testing, and optimising. The difference between a 1.5x ROAS (where you're barely breaking even, as the calculator showed) and a 4x+ ROAS (where you're building a real business) lies in this attention to detail and strategic approach.
Getting this process right from the start can save you thousands in wasted ad spend and months of frustration. If you'd like to chat through your specific product collections and how we'd approach building out this exact strategy for your brand, we offer a free initial consultation where we can map this out together.
Regards,
Team @ Lukas Holschuh
