Overstock.com Inc. wants to show consumers it knows them.
With an Internet Retailer-estimated 6 million SKUs on its site, Overstock is working on ways to show the right product to the right consumer at the right time. And it’s doing so through personalization.
“We want to improve the customer experience, which started our journey toward personalization,” says Amit Goyal, senior vice president of software engineering at Overstock.
Although Overstock has incorporated personalization into its website for years starting with basic product-based personalization, such as organizing product pages based on price, color and material, the retailer is now more focused on personalizing content and page layouts.
Overstock pulls content from an in-house built asset data platform, which is a system that stores all of the retailer’s content and product assets, and shows tailored content to consumers that visit its site. On the home page, product recommendations will be different shopper to shopper based on past searches, in-session behavior and purchase history.
For example, if a consumer searched for and purchased a couch during her last visit, and she’s now searching for ottomans, the platform will adjust and show her ottomans rather than couches as she navigates through the site and changes her search behavior, Goyal says. Based on this behavior, Overstock’s platform knows she is redesigning her living room and can populate the home page she sees with relevant products in the shopper’s preferred color and style, along with relevant articles and content.
“We believe that if we can personalize a consumer’s page to her style, with the colors she likes, with items within her budget, Overstock will be top of mind for her the next time she’s looking to make a purchase,” J.P. Knab, senior vice president of marketing at Overstock, previously told Internet Retailer.
But the challenge for Overstock is generating user data. “It’s hard to personalize if we don’t have enough data,” Goyal says. “As we see more repeat behaviors, the user will have a more personalized experience.”
46.3% of Overstock’s shoppers were return shoppers in 2017, while 53.7% were new shoppers in 2017, according to data from Top500Guide.com.
Despite this challenge, Overstock has seen improvements in key performance indicators, Goyal says. For example, shoppers who see personalized recommendations convert at a greater rate than those who do not see personalize recommendations. Shoppers who see personalized content also visit more pages, Goyal says, without providing specific figures.
“E-commerce is behind the curve when it comes to personalization,” Knab said. “If you look at the Googles and Facebooks of the world, you see that every aspect of the experience is personalized.”
That means the retailers that offer shoppers a tailored experience—in their marketing messages, on their sites and across devices—can gain an edge, he says. Overstock ensures its e-commerce site remains personalized whether a consumer is shopping on desktop, its mobile website or Overstock’s app.
“The only thing that changes with a device is the page layout,” Goyal says. “The personalization angle works the same way; the content and product works independently no matter the device.”
Overstock uses algorithms in its asset data platform to improve in-session personalization that works behind product recommendations—for example, a shopper who bought a blue area rug eventually bought a matching pillow or someone who searched for lamps also clicked on lampshades. The algorithms adapt to shopper behavior.
But Overstock still has a long way to go, Goyal says.
“We’re working on a new personalization algorithm with our machine-learning team,” Goyal says. “We need to better identify the next step in the user journey. On top of that, we need to do all that very fast so it will become even more engaging.”
Overstock is No. 30 in the Internet Retailer 2017 Top 500.Favorite