We study the drivers of playlist demand on Spotify. We base our analysis on a unique panel data set that combines aggregate listener data for 12,000+ popular playlists with data on how prominently these playlists are featured in the Spotify app. We decompose demand for playlists into three major drivers: (i) time-invariant user preferences for playlists, (ii) users’ responsiveness to featured playlists, and (iii) the popularity of tracks on a playlist. Our results demonstrate that users have strong preferences for playlists curated by Spotify, underscoring the importance of platform-curated playlists for stimulating consumption. Users also value featured playlists in the app, illustrating how Spotify can effectively guide attention to a selected set of playlists. In comparison, users respond weakly to the addition of popular tracks to playlists, pointing to a limited role of popular major label content in attracting listeners. Our demand estimates reveal that revenues of content suppliers asymmetrically depend on Spotify’s decisions about playlist compositions and playlist featuring. We provide novel evidence for a much lower power of the major labels, and thus music producers more broadly, compared to Spotify. We conclude with recommendations on how power asymmetries between digital platforms and their suppliers could be mitigated.