Skip to navigationSkip to content

How exactly Stitch Fix’s “Tinder for clothes” learns your style

By Quartz

Each customer has an individualized style map, laying out her feelings about peasant blouses, A-line dresses, or pencil skirtsRead full story

Comments

  • Also share to
  • As soon as I saw Stitch Fix's Style Shuffle game, I knew the company had to be collecting reams of data on users' ratings. The game is incredibly addictive--I rated over 9,000 clothing items in about six months. CTO Cathy Polinsky wasn't kidding when she told me they're just scratching the surface with what they can do with all that data.

  • The important piece is knowing what people like, so that they know what you’ll buy next. This means they know how to design clothes that will be a hit, and how many (and even who) will buy them. Cuts out a lot of waste in the business.

    For extra interesting details, check out https://algorithms-tour.stitchfix.com

  • This is not a new idea (Netflix was doing it about a decade ago), but it’s an interesting behind-the-scenes look at it

  • Using ML algorithms to observe and learn from someone’s behavior is much more effective than asking them questions.

    Impressive move by Stitch Fix to leverage ML for clothes selection. If you ask me, this product experience should be copied by all clothing retailers.

  • If nothing else, this article should make it abundantly clear what are possibilities with data (the good, the bad and the ugly). The positive side is never having to worry about how we dress, and whether or not it will fit in, as that is already predetermined by social norms of "likes". On the other side of it, we are going to face some interesting issues about next-generation influencers, as well as the mechanics and processing behind it. It could all be done by software bots liking things that

    If nothing else, this article should make it abundantly clear what are possibilities with data (the good, the bad and the ugly). The positive side is never having to worry about how we dress, and whether or not it will fit in, as that is already predetermined by social norms of "likes". On the other side of it, we are going to face some interesting issues about next-generation influencers, as well as the mechanics and processing behind it. It could all be done by software bots liking things that vendors use to rig the system, and we may just end up following what the big brands impose upon us through social media indoctrination.

  • My style is easy. Blue, grey, brown, black. Proper shoes and ordinary clothes.

  • Brilliant gamification....

Want more conversations like this?

Join the Quartz community for all the intelligence, without the noise.

App Store BadgeGoogle Play Badge
Leaderboard Screenshot

A community of leaders, subject matter experts, and curious minds bringing nuance back to how we talk about the news.

Editors' Picks Screenshot

No content overload: our editors will curate the most notable and discussion-worthy pieces for you every day.

Share Screenshot

Don’t just read the story, tell it: contribute your ideas and experience to the dialogue.