eBay launches Interests, personalized product recommendations that cut across shopping categories, using big data and a questionnaire
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.
Context & Ripple Effects
Interests is the productization of a build-out eBay has been assembling since its acquisition of predictive-analytics firm SalesPredict for machine learning and natural-language expertise in 2016. It arrives just months after an unflattering portrait of the company — 171M users and 1.1B listings but stagnant and struggling for relevancy under competitive pressure — making personalization the visible answer to that critique.
The framing from head of engagement Bradford Shellhamer is telling: rather than a store for everyone, eBay wants to be the favorite store of specific, multi-interest identities like the 'heavy metal-loving, yogi, OSU fan.' That is a deliberate pivot from category-based browsing toward identity-based curation.
First-order effects
- Shoppers get a cross-category recommendation feed built from big data plus a questionnaire, so discovery no longer depends on knowing which eBay category to search — and eBay's engagement team gains declared interest data it can layer onto behavioral signals.
- Sellers' listings become eligible for algorithmic surfacing outside their own categories, changing how inventory on a marketplace of 1.1B listings gets found.
Second-order effects
- Rival marketplaces face pressure to match feed-style personalization or cede discovery to whoever curates better — a race Etsy and eBay are still running years later as they emulate tactics popularized by Instagram, TikTok, and Pinterest.
- The questionnaire-plus-data approach shifts merchandising power toward eBay itself: the platform, not individual sellers, decides which products reach which identities, tightening control over commercial-intent density on the site.
Third-order effects
- If the pattern holds, marketplace commerce structurally converges with social feeds — search gives way to algorithmic curation as the default entry point, which is the same logic behind eBay's later experiments like the eBay Live interactive shopping beta built around rare trading cards.
- The long arc runs through CEO Jamie Iannone's eventual refocus on passionate users in high-resale-demand categories like trading cards and sneakers — Interests was an early bet that relevance comes from serving intense niches, not being everything to everyone.
The trend: Marketplaces are rebuilding discovery around social-media-style algorithmic personalization, trading their catalog breadth for curated identity-driven feeds to win back relevancy.