The way we shop for fashion is different from how we buy cameras—especially online. With fashion, reviews and specs are less important; fashion shopping is about discovering something that fits your taste and feels right. The web works well for buying cameras and other hard goods but for soft goods, such as clothing and accessories, it’s not the same as shopping in a store.


What’s more, the market for soft goods online is growing tremendously. A year and half ago, our team (which at the time was part of Like.com) started to wonder if we could create a better experience for people to shop online. Our team consists of PhDs in computer science with an emphasis on machine learning and computer vision, along with fashion designers and stylists—we jokingly called ourselves the computer nerds and fashion nerds (and a few of us were both). So, we set out to create a new way to browse, discover and shop for soft goods online.



Today, we’re excited to share with you our first step towards realizing this goal. It’s called Boutiques.com: a personalized shopping experience that lets you find and discover fashion goods, by creating your own curated boutique or through a collection of boutiques curated by taste-makers—celebrities, stylists, designers and fashion bloggers. Boutiques uses computer vision and machine learning technology to visually analyze your taste and match it to items you would like.

































In fashion, there are lots of choices. If there are, say, 500,000 items in a store, that means there are literally billions of different combinations of outfits you can make with those items. How do you sort through all of this? This site had to be a collaboration.



First we partnered with taste-makers of all types. We asked them not just to curate 10-50 great items they loved, but also to teach our site their style and taste. They did this by telling us what colors, patterns, brands and silhouettes they loved and they hated. They took a visual quiz that taught the site to understand their style genre: Classic, Boho, Edgy, etc. Our machine learning algorithms use this information to enable you to shop all of the inventory in the style of that taste-maker, on top of the 50 items they’ve hand-curated.



These days, bloggers, stylists and everyday fashionistas are expressing their sense of style online. We invited them to create boutiques so people could shop their diverse styles. But you have a unique and independent style too, so Boutiques also lets you build your own personalized boutique and get recommendations of products that match your taste.



In addition to all this, Boutiques offers a variety of features to search and discover merchandise including:



Advanced search filters - Filter by genre, silhouette, pattern, color families and sizes.






















































Inspiration photos - Try a search for [yellow pumps] and you’ll see matching outfit ideas to the right of the search results. We feature images from streetstyle sites, and collage and styling sites to provide you with the online equivalent of styled mannequins to give you inspiration.



Complete the Look - Ever wonder what to pair with that dress? Our fashion designers wrote hundreds of style rules—like “heavily patterned handbags don’t tend to go with heavily patterned dresses”—that we used to develop a tool to suggest items that match.



Visual search - Sometimes you love an item but not in a particular color. We analyze the photograph of an item for its color, shape and pattern and try to help you find visually similar items.



Boutiques on your tablet - Download our iPad application, lean back and move through inventory as if you were flipping through clothes on a rack at the store.



You can start shopping now at Boutiques.com. At this time, Boutiques is only available in the U.S. and only for women’s fashion, but we plan to expand in the future. Tell us what you think on our feedback form. And if you’re a designer, stylist, celebrity or retailer and want to participate on Boutiques.com, drop us a line.



Cross-posted from Official Google Blog: Munjal Shah, Product Management Director