Fast company logo
|
advertisement

The marketplace is using big data–and a questionnaire–to help sort through the 1.1 billion items its merchants offer.

eBay’s new feature finds the products you might actually buy

[Photo: courtesy of eBay]

BY Yasmin Gagne3 minute read

eBay, says Bradford Shellhamer, the company’s head of engagement, wants to be the “heavy metal-loving, yogi, OSU fan’s favorite store.” It also wants to be yours. And with a new feature called Interests, it aims to be better at helping you find the products that match your passions.

With over 1.1 billion items in almost every imaginable category–offered by everybody from Target, Barnes & Noble, and Neiman Marcus to individual sellers–shopping on the venerable marketplace can be overwhelming, an issue the company acknowledges and has long sought to fix. Interests is designed to take advantage of the site’s unique inventory from smaller merchants, such as antiques, niche collectibles, and designer clothing items that are unavailable on rival sites such as Amazon.

To use the feature, shoppers fill out a short questionnaire covering a few categories, clicking on favorite topics in categories such as hobbies, sports teams, and styles. eBay then combines that information with data from previous searches to generate tailored product recommendations specific to each customer, cutting across shopping categories.

[Images: courtesy of eBay]
With this new initiative, eBay aims to identify patterns and trends that go beyond the standard personalization features it’s long offered. “We want to do more than look at what you buy and alert you when something you need is back in stock, or let you know that what you were looking at before is available at a lower price or from a different seller,” says Mohan Patt, VP of buyer experience, who has led buyer experience for 13 years.

Leading the effort is Shellhammer, the founder of ill-fated e-commerce startups Fab.com and Bezar and eBay’s personalization honcho since March 2016.

As he imagined what eBay could become, Shellhammer looked beyond the retail world to the music industry. For instance, Spotify’s “Discover Weekly” personalized playlists served as an inspiration for Interests. The Interests questionnaire was influenced by the questions Apple Music asks users in order to provide them with the best possible recommendations. “We want you to tell us what you like, let us look at your behavior, and get you into your own personal store,” he says.

[Image: courtesy of eBay]
When I fed my own interests into eBay’s new feature–such as running, alternative music, and baseball–the site built my personal store. Most selections were spot on: a pair of sky blue loafers (apparently found based on my “denim” interest), trail running shoes, and a Houston Astros hat were all items I would consider purchasing. However, a few selections–like a Taylor Swift vinyl album and an off-brand imitation of a Cult Gaia bag–were a little off. Shellhammer explained that the algorithm would refine its suggestions over time based on factors such as what I search for.

“Everyone’s a little weird”

Asked what he learned working on Fab.com, a design-focused e-commerce site that sold in 2015 for a fraction of its peak value, Shellhammer says that he “realized that taste doesn’t scale–Fab was based on tastemakers making decisions for you, but I realized that it shouldn’t be about me. What we sell on eBay is not a founder’s vision . . . It’s the user’s vision.”

advertisement

[Images: courtesy of eBay]
eBay started out as an auction site selling all types of goods to all kinds of people, and historically it was less strategic about using data to power customer experience than Amazon, which began as a bookseller, strategically planned which verticals to grow into, and harnessed data from customers at every step of their journey. After spinning outPayPalthree years ago, eBay has doubled down on its personalization efforts, investing heavily inAI and image recognitiontechnology to improve the site’s shopping experience and attract new customers.

The company has already rolled out an image search tool and an AI personal-shopping concierge for Facebook Messenger. Recommendations started out relying heavily on looking at what users had recently bought; now previous purchases are just one piece of the puzzle as eBay factors in other information, such as location, and tailors product discovery more specifically to individual shoppers.

“We think it’s a question of getting people to walk into the right store,” says Shellhammer. “Everyone’s a little weird, and we want a store that caters to that.”

Recognize your company's culture of innovation by applying to this year's Best Workplaces for Innovators Awards before the extended deadline, April 12.

PluggedIn Newsletter logo
Sign up for our weekly tech digest.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Privacy Policy

Explore Topics