Zalando launches algorithm to suggest outfits to customers

Zalando will use machine learning to inspire its customers with outfit recommendations. The German e-tailer is launching Algorithm Fashion Companion (AFC), an algorithm which suggests outfits matching the most recent items customers have bought.

The algorithm is able to identify clothing items on Zalando’s marketplace and match them into outfits that include an item recently purchased by a customer. But how to make sure those outfits are good? According to a statement published today on the e-tailer’s website, AFC has taught itself what makes a good outfit based on more than 200,000 outfits created by stylists working for Zalon, Zalando’s personal styling subscription service.

“We did surveys containing outfits created by stylists and outfits created by the AFC to see if they would resonate with customers. Even our team was really surprised with the outcome”, said Zalando’s AFC Product Manager, Marta Skassa, in the statement.

The new algorithm is set to roll out this week in all countries where Zalando is present: Germany, Austria, Switzerland, France, Belgium, Holland, Italy, Spain, Poland, Sweden, Finland, Norway, Ireland, Czech Republic, United Kingdom, and Denmark.

Photo: courtesy of Zalando