Other online retailers are looking for ways to solve the opposite problem: Customers often begin shopping with only a vague idea of what they're looking for. Yet most sites are equipped to help shoppers compare products based only on price, not other important factors such as color and shape.
One French shopping site says it's designed for the way women actually tend to browse -- hunting for items based on style and color. "You cannot use the same attributes to navigate through dresses as you use for digital cameras," says Chief Executive Fabrice Berger Duquene.
"We compute an 'image DNA' -- a descriptor of the visual content of the image, such as shapes and colors," he says. Then that data can be easily compared with information from other pictures to identify, for example, the palette of a clothing collection.
Shopoon's visual-search site first asks shoppers to identify what they are looking for in basic terms, such as shirts or pants. Users then browse through pictures for an item that catches their eye, and the site shows them products that complement it.
Mr. Duquene says that users of his site stay two to three times longer than they do on a typical shopping site, and are 25% to 35% more likely to click on a product, then buy it from the merchant. In the near future, he says, shoppers will be able to grab a photo from a magazine or snap a picture in the real world and the site will match it to clothing for sale online, by analyzing image DNA on the fly.
One French shopping site says it's designed for the way women actually tend to browse -- hunting for items based on style and color. "You cannot use the same attributes to navigate through dresses as you use for digital cameras," says Chief Executive Fabrice Berger Duquene.
"We compute an 'image DNA' -- a descriptor of the visual content of the image, such as shapes and colors," he says. Then that data can be easily compared with information from other pictures to identify, for example, the palette of a clothing collection.
Shopoon's visual-search site first asks shoppers to identify what they are looking for in basic terms, such as shirts or pants. Users then browse through pictures for an item that catches their eye, and the site shows them products that complement it.
Mr. Duquene says that users of his site stay two to three times longer than they do on a typical shopping site, and are 25% to 35% more likely to click on a product, then buy it from the merchant. In the near future, he says, shoppers will be able to grab a photo from a magazine or snap a picture in the real world and the site will match it to clothing for sale online, by analyzing image DNA on the fly.
Posts
0 comments:
Post a Comment