Lessons from a Meta Fashion Account Audit
· Articles

Phil here, Director of Paid Social. I audited a fashion brand this week that's been spending the same budget for two years while sales have slowly increased. They want to spend more. They want to grow faster. Paid is one of their main growth drivers.
The question they came to us with: "What happens when we actually increase spend? And how do we make sure it's still profitable?"
It's one of the most common questions we get from fashion brands. And the answer almost always starts with something they're not looking at.
The Red Flag: Spend Up 260%, Reach Only Up 115%
In the UK, spend is up 260% year over year. Sounds aggressive. But reach is only up 115%.
That gap tells you everything.
Frequency is up 77% to 13 over 30 days. That frequency increase is driving CPMr up 67%, from £38 to £63. It now costs £63 to reach 1,000 unique people. This time last year it was £38.
When I talk about the cost to advertise increasing, this is what I mean. It's not creative fatigue. It's not ROAS declining. Those are small fixes. The big thing that causes brands issues over a long period of time is the economics of the platform no longer working. CPM increases, frequency increases, your cost to reach a thousand users goes up. That's the start of the funnel. Then everything else changes. Your CPC goes up, and your CPC can only go up so much before your conversion rate and AOV don't support the cost of that traffic.
You can make improvements to AOV. You can make improvements to conversion rate. But the cost to advertise can always overtake them.
In this case, the CPM hasn't actually increased year over year, which is unusual and actually fortunate. The problem is entirely frequency driven.
Why Fashion Frequency Is Naturally Higher (And Why That Makes This Harder)
This brand serves between 5 and 10% of the UK market. That's a lot of market share. And the customer journey is messy. Referrals from WhatsApp groups, word of mouth, direct search, Instagram, Facebook. If you picked a customer who ordered five minutes ago and traced back through their journey, it would be chaotic. It's a considered purchase, not an impulse buy.
This matters because Meta is touching every part of that funnel. It's not just running top of funnel and bottom of funnel like it would for a supplement brand. It's involved at every stage. And that's why frequency is naturally higher for this type of brand. There's a lot of organic and direct traffic hitting the pixel, and Meta likes that. It serves those people ads as a second touch point, which pushes frequency up.
The question the brand was asking: are we hitting a ceiling in the UK? Is our money better spent in the US where there's a blue ocean of growth?
It's a fair question. But before answering it, you need to understand whether the current UK spend is actually working as hard as it could be.
The View-Through Problem
A high percentage of UK sales are coming through the view-through attribution window. Around 50%.
That's a flag. It means Meta is likely serving ads to people who are about to convert anyway. The platform looks better than it actually is because it's claiming credit for sales that might have happened regardless.
In the US, it's different. Only around 30% of sales are on view-through. That's one of the reasons why reported performance looks worse in the US. Less view-through bloat means the numbers are more honest. The US isn't necessarily performing worse. It's just being measured more accurately.
This is why I'm confident the UK account is addicted to warm audiences. When 50% of your attributed sales are view-through, and your frequency is 13 over 30 days, Meta is recycling the same people repeatedly and taking credit when they convert.
Phase One: Reduce Frequency Through Exclusions
The immediate fix isn't TOF campaigns or creative overhauls. It's exclusions.
Implement a 180-day purchase exclusion. For a brand where most customers are single-purchase, 180 days is the right window.
Exclude the Klaviyo email list. These people are already in your ecosystem. Stop paying to reach them through Meta.
For the test that I think would be most revealing: take one campaign per product and apply full exclusions. Pixel purchases, Klaviyo email list, website visitors, and social engagers. Strip it all out.
What should happen: frequency goes down, reach goes up, and the percentage of sales on view-through goes down. The reported performance might look worse initially. But it should be more incremental. You're finding genuinely new customers rather than paying to serve ads to people who were going to buy anyway.
This is the test that tells you whether the UK has more headroom or whether you really are hitting a market ceiling.
Phase Two: Top of Funnel (But Not How You Think)
The brand's instinct was to put 20% of spend into a traffic campaign. "That would completely transform the net-new reach numbers, right?"
Yes, it would. And that's exactly why it's the wrong move.
You can make the net-new reach chart look incredible by dumping 20% of budget into a traffic campaign. But it doesn't mean that traffic is any good. It's the easiest way to do TOF wrong.
Start with 2%. Not 20%. You want the budget low enough that you can run it for a long period of time. You don't want to do 20%, have it not work after two weeks, pause it, and realise you'd have been better off not doing it in the first place.
Three campaign types that work well for TOF:
Page Likes. If the audience skews older and is more active on Facebook. Sounds old school. Still works.
Traffic to social profiles. Push people to your Instagram profile and Facebook page. This is especially valuable for brands where organic social is a big part of the customer journey. If that social profile visit matters in the path to purchase, driving more of them makes sense.
Google Referral custom conversion. This is the one most brands aren't running. Set up a custom conversion event that fires when someone sees a Meta ad and then comes back to the website via Google within seven days. If someone does that, it's a strong signal of high-quality traffic. They're real. They remembered you. They searched for you.
Because it's a custom conversion, you can track it across all your campaigns. You can see which campaigns drive Google referrals at what cost. And you can set it up as the campaign objective, so Meta goes and finds you better quality traffic rather than just the cheapest traffic available.
David Herrmann shared a great tactic for top of funnel campaign types...
Measure Before You Change Anything
Before any of this, run a conversion lift test. Account level, not campaign level. You can choose the location, so start with the UK.
This gives you your cost per incremental conversion. It's the benchmark you measure everything against.
And you don't need to increase budget to run it. Meta reps will tell you that you need to spend £120k or whatever number they pull out. You don't. Set it up at your current spend level. Run it for two weeks. Get the data.
Then make your changes. Exclusions, click-only attribution testing, TOF campaigns. Do the work. Then repeat the conversion lift test. Compare. That tells you if what you've done has actually improved incrementality or if you've just shuffled conversions between campaigns.
Fashion Creative: Let It Live
One thing that's different about fashion and apparel accounts versus supplement or health brands: you want ads to live as long as possible.
For a seasonal brand, think of your ad account like an organic social profile. You post a reel because it's right for this moment and you let it live on the profile. The same applies to paid. Launch an ad, let it build momentum, and that curve ends naturally with seasonality and the product being replaced by the next collection.
Compare that to a protein or creatine brand where you're in and out. Testing hooks, burning through creative, throwaway concepts. Fashion doesn't work like that. You want to be known for something. Put your flag in the ground. This is what we sell. This is what we're known for. Dig that hole deep because that supports the word-of-mouth piece too.
Our cadence is every Wednesday. Like clockwork. New ads launch into the account every Wednesday. Not every day. Brands that launch too much creative too frequently end up in a trap where nothing gets spend because the algorithm can't distribute budget effectively across too many options.
If you're seeing a lot of ads that don't get spend, or you're relying on winners from six months ago, that could be a sign you're launching too many similar creatives and not enough genuinely new concepts or formats.
"Creative Sameness" Might Be Nonsense
Meta has been pushing the creative diversity angle hard. You need more different ads. Your creative is too similar. The algorithm needs variety.
Some of that might be nonsense.
Here's why. If you're serving the same small audience repeatedly, it doesn't matter what your creative is. They've seen it all. The ads feel stale not because the creative is the same, but because the audience is the same.
As soon as you tap out of that pool and start reaching new people, those existing creatives are new to them. You don't necessarily need different ads. You need different audiences. TOF and exclusions solve that problem. Not more creative for the sake of it.
That said, creative diversity does matter. But the priority is net-new concepts and formats, not colour swaps and headline variations. When your internal team is doing the BAU content that's been working for three or four years, the job of an external partner is to find the new playbook: creator content, AI content, motion graphics, formats you haven't tried yet. A lot of that stuff doesn't work straight away. But that's what you're paying for. The discovery.
The UK vs US Question
The brand's hypothesis: maybe the UK is approaching a ceiling and the money is better spent in the US.
It's possible. But you can't answer that question until you've actually tested whether the UK has more headroom. Right now, the high frequency, lack of exclusions, and absence of TOF campaigns mean there's untapped potential being left on the table.
Fix the structural issues. Run the exclusion test. Launch TOF at 2%. Measure incrementality. If after all of that, the UK still shows diminishing returns, then yes, shift budget to the US where there's less competition and more room to grow.
But don't make that decision based on a hypothesis when there are obvious improvements that haven't been tried yet. The UK isn't necessarily tapped out. The current setup is just making it look that way.
The Takeaway
Every fashion brand audit tells a version of the same story. The cost to advertise is increasing. Frequency is climbing. CPMr is going up. Reach is flatlining despite higher spend.
The fix is always structural before it's creative.
Exclusions first. Then consolidation. Then measurement through conversion lift tests. Then TOF at a sensible budget with full exclusions and proper incrementality tracking. Then creative diversity to expand the addressable audience.
The brands that crack this will scale. The ones that keep increasing spend without fixing the system will keep paying more to reach fewer people.
And if you're sitting on a frequency of 13 over 30 days, the first thing to do isn't panic. It's understand why. Fix the inputs and the outputs will follow.