Why are the fastest-growing online stores almost always the ones leaning hardest into AI? Because they have quietly solved the oldest problem in retail: showing each shopper the right product at the right moment. ASOS, the UK fashion giant that adds thousands of new styles every week, is one of the clearest examples of what happens when you get that right.
The result: ASOS’s AI-powered product recommendations drove a 75% increase in email click-through rates and a 25% increase in average order value.
The problem AI recommendations solve#
ASOS lists hundreds of thousands of products and refreshes the catalog constantly. For a shopper, that scale is both the appeal and the obstacle. Nobody wants to scroll through ten thousand dresses. They want the handful that fit their taste, size, and budget — and they want them surfaced before they lose patience and leave.
Static merchandising can’t keep up with a catalog that changes daily. A human team can hand-pick a “trending now” row, but it can’t tailor that row to millions of individual shoppers in real time. That gap between what’s in stock and what this specific person wants is exactly where revenue leaks out of most stores.
How recommendation AI works#
Recommendation AI closes that gap by learning from behavior instead of guessing. ASOS uses collaborative filtering and machine-learning models that continuously process browsing and purchase data, so the system understands not just what a product is but who tends to buy it and what they buy alongside it.
In practice, the engine does three things at once:
It builds a profile of each shopper from the signals they generate — what they view, what they linger on, what they add to cart, what they return. It maps every product into the same behavioral space, so “items like this” means items that people like you actually buy, not just items in the same category. And it ranks recommendations in real time, reshuffling as the shopper’s session unfolds and as new inventory lands.
Because the model runs on live data, the recommendations stay fresh as the catalog turns over each week. The same engine powers on-site product rows, search results, and — crucially for ASOS — the email channel.
Why the email number matters most#
A 75% jump in email click-through is the stat worth dwelling on. Email is one of the few channels a retailer fully owns, and its economics are unbeatable when it works. The difference between a generic “new arrivals” blast and an email populated with products chosen for that individual recipient is the difference between a deleted message and a click.
ASOS folds its recommendation engine directly into email, so each send is personalized to the recipient’s recent behavior and predicted taste. The 75% CTR lift shows what happens when personalization moves from the website into the inbox: the same AI that helps shoppers on-site brings lapsed and casual browsers back, at almost no incremental cost.
The 25% lift in average order value compounds the effect. Better recommendations don’t just earn the click — they grow the basket, because complementary and higher-intent items surface naturally alongside the first purchase. When customers find what they need faster, they buy more, return less, and come back more often.
What this means for your store#
You don’t need ASOS’s data-science headcount to capture this advantage. The mechanics are the same whether you sell 200 SKUs or 200,000:
- Personalized recommendation rows on product and cart pages lift conversion and AOV by surfacing the right next item.
- Behavior-driven email turns your owned channel into a revenue engine instead of a broadcast.
- Real-time ranking keeps recommendations relevant as inventory and trends shift.
The brands winning today treat personalization as infrastructure, not a nice-to-have. The technology that once required a dedicated ML team is now available to any store.
Bring recommendation AI to your store with CartAmplify#
CartAmplify makes enterprise-grade recommendation AI accessible for every online store — Shopify, dropshipping, or marketplace. The same approach that helped ASOS lift email CTR 75% and AOV 25% works for your catalog: AI that matches each shopper with the right product, on-site and in the inbox.
Related reading#
- How ASOS’s AI Discovery Lifts Email CTR 75%, AOV 25%
- How ASOS Personalizes Feeds with Recommendation AI
- How Deliveroo’s AI Makes Plus Members Order 40% More
Figures cited from publicly reported ASOS AI case studies. Results vary by catalog, traffic, and implementation.
