The new update allows users to sort their weekly music recommendations according to different genres, discover new artists and see suggestions from editors, which represents a new era in combining technology with human input.
Spotify has updated one of its most used personalised playlists, introducing customisation controls to Release Radar that give listeners a more active role in shaping the music they see each Friday. The changes, rolling out globally across mobile and desktop, reflect a broader rethinking at the streaming platform about the balance between algorithmic recommendations and human editorial judgment.
What Release Radar Is and What Is Changing
Release Radar is a playlist that refreshes every Friday, surfacing new music from artists a listener follows or has shown interest in. Until now, the playlist was generated entirely by Spotify’s algorithm with no user input. The new update introduces filters that let listeners choose up to five customisation options, which can include genre-specific categories such as Pop, a Discover New Artists option for those wanting to broaden their listening, and Editor Picks curated by Spotify’s in-house editorial team.
Alongside the filters, Spotify has refined the underlying recommendation algorithm to better reflect individual listening preferences, and added a visual refresh to the playlist with updated cover art and redesigned headers.
Why Spotify Is Making This Move
The timing is not accidental. Over the past year, Spotify has faced sustained criticism that its increased reliance on algorithmic curation has made music discovery feel increasingly impersonal and homogeneous. This year the company introduced editorially chosen release recommendations as a first step toward reintroducing human judgment into the process. The Release Radar update extends that thinking by giving listeners themselves the ability to steer their recommendations rather than passively receiving whatever the algorithm decides.
The problem addressed by the update is something the music industry has been dealing with: although automated processes are good at making music recommendations, they look uninspired, while human recommendation is more justified but leaves much room for improvement.
