TORONTO, Ontario — Remember when you had to leave your house to buy a new sweater or the latest video game console? All that feels like ancient history and nowadays there’s no shortage of online shops to choose from for all of your shopping needs. How, then, can a new fledgling online store convince shoppers to stay on their site? Researchers from the University of Toronto report variety may just be the key to online retail success.
Most online stores use a mechanism that arranges their products in order according to popularity, with the shop’s biggest sellers appearing at the top of the page. For this research project, however, scientists decided to try something different.
They developed a new algorithm that accounts for product popularity while also adding a “dynamic blend” of variety. To the research team’s surprise, customer engagement increased by as much as 30 percent.
“We were surprised by the magnitude of the results,” says Shreyas Sekar, an assistant professor of operations management at the University of Toronto Scarborough and the Rotman School of Management, in a university release. “We thought that maybe we’d get a 5 percent improvement, which is not bad because we’re dealing in the order of hundreds of thousands of customers, if not millions on some of these platforms. But this was amazing.”
Creating a new ‘hook’ for online customers
Study authors focused on the concept of “hedonic browsing,” which is really just a fancy way of saying window shopping. Many people can relate to absent-mindedly browsing around a local mall with no real purchase in mind. The same concept applies to online shopping. Consumers look around their favorite online websites from time to time, hoping something that might catch their eye.
The real challenge for the online retailer is to ensure the shopper is “hooked,” or decides to stay on the site for longer than a few seconds and continues browsing. It’s a much harder task than it sounds, as most statistics show the majority of online shoppers never make it past the first page of a store website and usually click out to another page within 30 seconds.
So, in collaboration with WayFair, the team at UT developed a new algorithm for WayFair’s website. In more concrete terms, the algorithm helped to arrange the ideal set of product rankings for any specific event or occasion — Black Friday or Halloween, for example. Product popularity remained a major determining factor, but the algorithm also prioritized diversity and variety among displayed products. According to the research team, this produced much stronger customer interest.
Re-ordering everything from shopping lists to news stories
Across six different events, the new program produced anywhere from five to 30 percent upticks in customer traffic. Even better, the algorithm is adept at continually learning, so it doesn’t need a constant stream of new customer information. Study authors add that the algorithm can help with far more than just online shop displays. Theoretically it should produce similar results for email marketing lists or news website top stories.
“We tend to think that revenue is the only objective that online retailers care about, but increasingly platforms care about growth and retaining their user base,” Prof. Sekar concludes. “If a customer gets hooked today, the platform may get to keep them as a repeat customer over and over again.”
The study appears in the journal Management Science.