Case Study: GlassesUSA.com achieves 88% revenue uplift and 68% increase in purchases with Dynamic Yield

GlassesUSA.com Deploys a Deep Learning Algorithm to Adapt Its Recommendations to Each Shopper

GlassesUSA.com, the world’s largest online eyewear retailer, faced the challenge of helping shoppers discover the right pair among thousands of styles across private-label and 60+ designer brands. The eCommerce team wanted homepage recommendations that could self-train quickly, use both historical and in-session behavior, and continuously improve so visitors would be matched with the most relevant products immediately upon entry.

To solve this, GlassesUSA.com tested Dynamic Yield’s deep learning recommendation engine against a traditional collaborative-filtering approach on its homepage widget. The deep learning model automatically tailored suggestions per user and session, driving a 45% increase in add-to-cart rate, a 68% uplift in purchases, and an 88% increase in revenue, after which the retailer adopted the strategy broadly (including as the primary mobile approach).

GlassesUSA.com

Nadav Yekutiel

Head of Data