Spotify Takes the Personal Mixtape to the Masses

Every Monday, Spotify creates 100 million personalized playlists. The key? Big data algorithms.

This program was produced by the Marketing Department of WIRED and Ars in collaboration with CA Technologies.

Spotify is a darling of the online music scene, and Discover Weekly is one of its most beloved features. For those not in the know, Discover Weekly is a list of songs, delivered every Monday to Spotify's 100 million active users, which listeners probably aren't familiar with—but that Spotify thinks they will want to hear.

Every user's discoveries are unique, tailored to his or her individual tastes. How does it work? Many have suspected that there must be a human hand guiding the process, but Spotify Engineering Manager Edward Newett, who is the lead engineer on Discover Weekly, says that's not the case. It's all based on advanced machine learning, with thanks to the big data set generated by more than 100 million users with billions of collective playlists.

How does Discover Weekly find unheard songs you'll likely love? Newett walks us through the tech behind the scenes:

Derek Korte
By Derek Korte | November 08, 2016