Why do generic “best seller” rows underperform so badly compared to personalized recommendations? Because a top-seller list answers the question “what’s popular?” when the shopper is really asking “what’s right for me?” Yves Rocher, the global cosmetics brand present in dozens of countries, quantified that gap — and it’s enormous.
The result: Yves Rocher’s AI-powered, real-time personalized recommendations achieved a purchase rate 11x higher than standard top-seller recommendations, alongside a roughly 17.5x increase in clicks on recommended items.
Top-seller lists are leaving money everywhere#
Most stores fall back on popularity: a “best sellers” row shown identically to everyone. It’s easy, and it’s deeply suboptimal. Popular products are popular on average, but no shopper is average. The serum that tops the chart may be wrong for a given customer’s skin, budget, or concern — so the recommendation is ignored, and the sale is lost.
Yves Rocher replaced that one-size-fits-all approach with real-time personalization, and the difference was dramatic: an 11x lift in purchase rate on recommended products. Crucially, it worked even for first-time, anonymous visitors — the system builds a profile from the moment a shopper views a product, personalizing the experience within the same session.
How real-time AI recommendations work#
The key word is real-time. Rather than waiting for purchase history to accumulate, the engine starts learning from the shopper’s very first action and adapts continuously through the session.
Three capabilities drove Yves Rocher’s result. The system personalizes from the first interaction, creating an anonymous profile the moment a visitor views a product, so even new shoppers get relevant suggestions. It updates in real time, reshaping recommendations as the shopper browses rather than relying on stale, pre-computed lists. And it optimizes for purchase, not just clicks — surfacing products a shopper is genuinely likely to buy, which is why the purchase-rate lift was even more valuable than the click lift.
Why the anonymous-visitor angle matters most#
The most strategically important detail is that the 11x lift extended to anonymous, first-time visitors. For most stores, the majority of traffic is exactly that — people with no account and no history. If your personalization only kicks in for logged-in repeat customers, you’re leaving your largest segment on a generic experience.
Yves Rocher’s in-session personalization captures intent from behavior alone, no login required. That means the bulk of traffic — the new and anonymous visitors other stores treat as strangers — gets recommendations as relevant as a returning customer’s. Multiply an 11x purchase-rate lift across that volume and the revenue impact is hard to overstate.
What this means for your store#
The lesson generalizes to any store still leaning on best-seller rows:
- Replace static top-seller lists with real-time personalized recommendations tuned to each shopper.
- Personalize from the first interaction so anonymous, first-time visitors get relevance too — not just logged-in regulars.
- Optimize recommendations for purchase likelihood, not just clicks, to capture the full revenue lift.
Popularity is a weak proxy for relevance. The stores that win replace “what’s popular” with “what’s right for you.”
Bring AI recommendations to your store with CartAmplify#
CartAmplify brings the same real-time personalized recommendations that gave Yves Rocher an 11x purchase-rate lift to any store — Shopify, dropshipping, or marketplace. Relevance from the first click, for every shopper — even the anonymous ones.
Related reading#
- How Ozon’s AI Search Helped Drive a 64% Revenue Surge
- How Zalando Cut Content Costs 90% with Generative AI
- How Adlibris Grew Sales 49% with AI-Powered Search
Figures cited from the publicly reported Yves Rocher / Bloomreach case study. Results vary by catalog, traffic, and implementation.
