Spotify is letting Premium subscribers directly edit the algorithm that powers their recommendations. The new Taste Profile feature, announced by co-CEO Gustav Söderström at SXSW, gives users full visibility into the data model behind Discover Weekly, Release Radar, and Spotify Wrapped, according to TechCrunch.
The beta launches in New Zealand first.
New Zealand is not a random choice. Spotify used the same market to test its Prompted Playlist feature, which lets users generate playlists through natural language descriptions.
That tool launched in New Zealand, expanded to the US and Canada in late January 2026, and reached Australia, Ireland, Sweden, and the UK by February, according to The Next Web.
It turns recommendations into a conversation between user and machine, where explicit feedback overrides implicit signals. This matters because recommendation fatigue is real. Users who feel "stuck" in a loop of similar suggestions stop engaging.
By giving listeners direct control over the underlying model, Spotify addresses the core complaint without rebuilding the algorithm itself.
But here is the bigger picture. Spotify's latest Loud and Clear report shows that artists from 75 different countries generated at least $500,000 in streaming royalties last year, up from 66 countries the year before.
About half of an average artist's streams now come from outside their home country, per The Hollywood Reporter. As listening habits go global, personalized recommendations become the difference between discovery and noise.
For Premium Subscribers
You can finally prune genres you hate and prioritize what you want. When this rolls out beyond New Zealand, it turns Spotify from a guessing machine into a tool you control. For Developers: Transparent, user-editable AI models are becoming the standard. Build interfaces that let users directly tune algorithms, not just react to them. For Content Creators: Understanding how users shape their Taste Profiles will inform content strategy. Target specific listener archetypes with greater precision. For Founders: User control is the new frontier in recommendation systems. Features that allow explicit algorithmic editing, not just implicit feedback, will drive retention in subscription models.
The Spotify Taste Profile is a new feature that allows Premium users to directly influence their music recommendations. It lets users view and modify the data Spotify uses to personalize playlists and suggestions, addressing issues where recommendations don't align with current tastes. The feature is currently in beta testing, starting with users in New Zealand.
With the Taste Profile, you can view a summary of your listening habits and provide feedback to Spotify. You can request more of a specific genre or actively remove categories you no longer want to hear. The system also understands more complex requests, such as asking for "upbeat music for marathon training" or "news podcasts for a commute."
Spotify often uses New Zealand as an early-stage testing ground for new features, especially those involving AI. For example, the "Prompted Playlist" feature, which generates playlists from natural language descriptions, was first tested in New Zealand before expanding to other countries. This allows Spotify to gather user feedback and refine the feature before a wider release.
The Taste Profile represents a shift towards giving users more control over AI-driven experiences. Instead of algorithms operating as a "black box," users can actively participate in shaping the AI's understanding of their preferences. This approach aligns with a broader trend in the tech industry where companies are empowering users to refine AI through direct feedback.
Currently, the Spotify Taste Profile is in beta and only available to Premium subscribers in New Zealand. Spotify has not yet announced when the feature will be rolled out to other regions or user tiers. The company will likely use the beta period to gather feedback and make improvements before a wider release.
More insights on trending topics and technology
