Big catalogs don't win AI search. Better data does.
Independent fashion brands have a structural advantage in AI recommendations that most don't know they have — because it has nothing to do with catalog size.
The assumption most fashion brands make is that AI search works like Google Shopping — that the biggest catalogs win. If you’re a mid-market brand going up against ASOS or Zara, that assumption probably feels deflating.
It’s also wrong.
Why catalog size doesn’t matter in AI recommendations
When a shopper asks ChatGPT “what to wear to a garden party if you’re broad-shouldered and petite,” the model isn’t looking for the brand with 50,000 SKUs. It’s looking for products it can confidently match to that specific person and occasion. Scale only helps if the product data underneath is rich enough to enable that match.
ASOS has over 850,000 products. Most of them have descriptions written for a shopper who’s already browsing — short, punchy, optimised for a grid. “Floral midi dress, regular fit.” That’s not enough for a model to decide whether it works for a petite woman with broad shoulders attending a garden party.
A brand with 400 products and honest, specific descriptions of how each piece actually fits? That catalog is more useful to an AI assistant than one with 800,000 items written for a different era of search.
What AI needs to recommend a product
AI assistants match natural language questions to product attributes. The shopper’s question contains real information: body shape, occasion, climate, aesthetic. Your product data either contains enough to answer that — or it doesn’t.
For that garden party query, a useful product description might read: “structured linen midi with a nipped waist, square neckline that helps balance broader shoulders, skims without clinging through the hips, hits mid-calf on a 5’4” frame." That’s specific enough to match. “Midi dress with floral print” is not.
Independent brands have an inherent advantage here. Their catalogs tend to be more considered. The buyers and founders often know their pieces deeply — which styles work on which body types, how fabrics behave across seasons, what occasions a piece actually suits. That knowledge exists. It just usually hasn’t made it into the product data.
The gap most small brands don’t realise they have
The reason most independent brands aren’t showing up in AI recommendations isn’t that their products aren’t good enough. It’s that their catalog data doesn’t reflect what they already know about those products.
A boutique that hand-picks 200 pieces for a specific woman — the one who travels for work, wears a 14, needs clothes that go from client breakfast to dinner without a wardrobe change — has a goldmine of implicit product knowledge. It’s in the buyer’s head, occasionally in the product copy, but rarely structured in a way AI search can actually use.
Translating that knowledge into enriched product data — detailed attributes covering fit, occasion, body type compatibility, colour season, and more — is what makes a catalog AI-legible. It’s a one-time infrastructure investment, not a permanent overhead.
The brands already doing this
The independents showing up in AI recommendations for category queries right now — “best bags for work travel,” “swimsuits for short torsos,” “jewellery for cool undertones,” “linen dresses for a summer wedding” — didn’t get there by luck. They structured their product data to answer the questions their shoppers are actually asking.
They didn’t need to out-scale ASOS. They needed to be more useful than ASOS at answering a specific question for a specific person. That’s a competition they can win — and increasingly are.
The timing matters too. Most big retailers are moving slowly on catalog enrichment because it’s a large, complex, cross-functional project at that scale. Independent brands can move faster. That window won’t stay open indefinitely.
Where to start
The fastest way to see where your catalog stands is an AI visibility audit — it takes about five minutes and shows you how your products perform against the queries your target shopper is actually running. Most brands are surprised by the gaps. Some are surprised to find they’re already showing up.
Either way, you learn something. And for brands with deep product knowledge and a curated catalog, the gap between where they are and where they could be is usually much smaller than they think.
The brands winning AI search right now aren’t the biggest. They’re the most legible. That’s a race worth entering.