Anyone who has ordered groceries online in the past 15 months likely has a "What were they thinking?" item substitution story. Whether it was an item from an entirely different category or one with a wholly different price point or nutrition profile than the product the customer wanted, an out-of-sync sub can frustrate customers and ultimately diminish their affinity for or loyalty to the brand.
Walmart wanted to get smarter about that substitution process, and the AI-powered algorithms it built to do so has boosted customers' acceptance of suggested substitutes to more than 95%, according to the retailer.
"The tech we built uses deep learning AI to consider hundreds of variables—size, type, brand, price, aggregate shopper data, individual customer preference, current inventory and more—in real time to determine the best next available item," Walmart Global Tech EVP Srini Venkatesan wrote in a blog post. Every acceptance or rejection of a suggested substitution helps the algorithms become more intelligent about customers' personal preferences, he noted.
"The decision on how to substitute is complex and highly personal to each customer," Venkatesan wrote. "If the wrong choice is made, it can negatively impact customer satisfaction and increase costs."
The tech also reduces potential frustration for store associates, he added. Previously, order pickers used a manual process to select and manage substitutions. That process could be time-consuming and of varying effectiveness, Venkatesan noted. Now, Walmart's home-built tech suggests personalized alternatives and identifies where in the store they're located.
"Our goal is never to be out-of-stock and never to have substitutions," he wrote. "But, when it happens, the technology we’ve built helps ensure customers get the next best thing."
WANT BREAKING NEWS AT YOUR FINGERTIPS?
Get today’s need-to-know grocery industry intelligence. Sign up to receive texts from Winsight Grocery Business.