Retailers Have Just 30 Seconds To Hook Consumers Before They Look Elsewhere

A new online shopping experience is only ever a click away, so loyalty online always seems a somewhat ephemeral experience.  Indeed, research from the Rotman School of Management shows that e-commerce retailers have just 30 seconds to impress customers before they look elsewhere.

The researchers have also got some help at hand, however, as they developed algorithms to optimize the display of products that managed to boost customer engagement by around 30% over traditional ranking mechanisms that use past popularity to determine the order in which products are displayed.

“We were surprised by the magnitude of the results,” they explain. “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.”

Hedonic browsing

The researchers focused their attention on something known as “hedonic browsing”, which is the digital equivalent of window shopping.  It’s when we’re not really looking for anything in particular but are just somewhat aimlessly browsing a website for things that might catch our eye.  The retailer’s job, therefore, is to ensure that something does, which will hook the customer to browse more.

This is often easier said than done, as around two-thirds of online shoppers fail to make it past the first page of the product listings.  The researchers worked with the online retailer Wayfair to see if they could improve matters.

They used historical sales data from the company to develop a range of algorithms to adjust and optimize the product rankings in real-time so that they could better adjust to particular shopping events, such as Black Friday and Halloween.  The algorithms were designed to not only rely on traditional product popularity but also to ensure there was diversity of items to broaden customer appeal.

The researchers managed to hook up to 30% more customers than the previous algorithm.  What’s more, the researchers also believe that their algorithms were designed to be super-efficient and therefore not require large quantities of data to function effectively.

They argue that while this particular study involved an e-commerce store, their findings would be equally applicable to any environment whereby choices are ranked, such as in the ordering of news media 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,” they conclude. “If a customer gets hooked today, the platform may get to keep them as a repeat customer over and over again.”

Facebooktwitterredditpinterestlinkedinmail