Using Data and Algorithms to Recommend Fashion Products

Source: TechCrunch, Aug 2013

Chicago-based StyleSeek has raised $750,000 more in pre-Series A funding for its e-commerce recommendations service for lifestyle products, which is today primarily focused on men’s and women’s fashion. Though the service competes with a number of product aggregators on the market, like Wanelo, Fancy, Want, Fresh, TheFind and others, StyleSeek instead is betting on algorithms and data over social signals. The site walks new users through a “style game” in order to customize their preferences, turning the online store into a unique experience personalized to every visitor.

Spalding describes his early vision as one of building “a Pandora for e-commerce products.” The goal was to figure out why people chose this shirt over that one, for instance – could a machine be taught to understand an individual’s style? The team spent all of 2011 running tests: paper tests, online tests, surveys, focus groups and more, and added a fourth co-founder, Frank Yang.

Some 200,000 early testers went through StyleSeek’s “Style Game,” which is essentially a tool that trains the algorithms to understand your personal style profile.

StyleSeek Screenshot - StyleGame

After stepping through the game, the service then pulls in items based on your actual style preferences – not your Facebook “likes” or profile data, not what your friends are buying, or what’s trending or popular with a majority of users – just things that match your own interests. “We can give you Pinterest-quality recommendations in 30 seconds – you don’t need to follow anyone, you don’t need to interact with anyone,” Spalding explains.

The company has relationships with over 150 online retailers, including big names like Macy’s, Barney’s, Saks, and Nordstrom, for example, as well as online shops like Nasty Gal, Bonobos, Warby Parker, and others. For now, StyleSeek generates revenue when users click through to purchase items off its site at these stores. And to encourage repeat visits, it sends out weekly emails, personalized to each user.

Advertisements
This entry was posted in Big Data, Decision Making, Relationship, Retail, Startup. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s