Client Success Stories | Criteo

Increase in AOV YoY


Increase in ROAS YoY


US-Mattress is the largest online mattress retailer in the world, with more than 500,000 customers in the United States alone. Until 2015, the company was managing its Google Shopping bids in-house, structuring bids down to the brand model level with SKU-level bids only for top sellers in the group. US-Mattress was looking for a way to cost-effectively scale volume and increase revenue for its Google Shopping program while improving performance for its top-selling mattress models. Criteo conducted an audit for the company that showed that an automated, end-to-end approach could increase revenue by at least 20%.


US-Mattress brought Criteo Predictive Search on board to automate and optimize its entire Google Shopping program, using machine learning to maximize revenue and efficiency. Criteo transformed US-Mattress’s remarketing lists (RLSA) into a robust remarketing program to value each user individually and to set a bid to match each user’s likelihood of making a purchase. They saw a +21% AOV driven by returning users, while still gaining significant growth from new users. Criteo also helped US-Mattress ensure that every product in the catalog received its own bid, drawing on over 40 signals to accurately predict the value.


Criteo Predictive Search nearly tripled YoY revenue and delivered:

  • 21% increase in AOV YoY
  • 51% increase in ROAS YoY

“We are benefiting from spend efficiency by leveraging Criteo Predictive Search — and significantly more revenue. Criteo’s solution automates bidding at the SKU level and leverages precision targeting so we can better understand which customers will make purchases and can focus on shoppers with higher average order value.”

– Bohan Huang, Marketing Manager, US-Mattress

We use Cookies to enhance your experience on our website. By continuing your navigation, you accept the placement and use of Cookies. To learn more about Cookies or opt-out from these services please see our privacy policy more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.