Spotify's Algorithm Measures Its Own Success

The Guardian asked readers for their favorite albums of 2026 halfway through the year. The responses came back full of language that reads like algorithm poetry. The language was propulsive, joyful, relatable, and queer joy.

The picks feel coherent, almost curated—they feel right. Yet they feel like something else entirely. They are not a window into what people are actually replaying at midnight, but evidence of something narrower and stranger.

The assumption underneath reader favorites is simple and never examined—when someone says they love an album, that statement reflects their genuine listening behavior. In the streaming era, this is almost certainly false.

Spotify's algorithmic recommendations and editorial playlists don't operate neutrally. They promote artists with strong social narratives—particularly those framed around queer identity and community joy—because those stories convert to playlist adds and social media amplification. An artist like Muna, whose identity and politics are inseparable from their music, gets surfaced relentlessly. A listener encounters the album repeatedly through discovery playlists, sees it praised on platforms where their friends are also seeing it, and forms a sense that the album is culturally important.

They report liking it—and Spotify's recommendation engine has just measured its own success. This isn't a conspiracy. It's how discovery works in a world where attention is the scarce resource. But it means reader taste aggregates something closer to "what Spotify's editorial team prioritized in Q1 2026" than "what people most want to listen to when nobody's watching."