In recent years, the way people discover new music has evolved significantly, moving from methods like browsing record stores to algorithm-driven recommendations. Prior to the 2010s, many sought new music through physical experiences like visiting record stores or through recommendations from friends and family. The advent of algorithmic recommendation systems began in the 2000s, notably with Pandora’s Music Genome Project, which analyzed songs based on quantifiable traits and offered recommendations based on these traits. However, Pandora faced challenges, including a smaller song library that limited variety.
Spotify entered the U.S. market in 2011 with a catalog of 15 million tracks, quickly gaining popularity through a focus on algorithms. Its “Discover Weekly” feature, introduced in 2015, provides personalized playlists by analyzing user data and preferences alongside music data from various sources. This has positioned Spotify as the leading music streaming service globally.
However, concerns have arisen about the impact of algorithmic recommendations on music discovery. Recent studies indicate that younger listeners, particularly those aged 16-24, are less likely to discover new favorite artists compared to older generations, often preferring to consume music snippets on social media platforms without exploring artists further. This has led to conversations around “algorithm fatigue,” prompting some services, like Apple Music, to emphasize human curation.
Moreover, the rise of vinyl records and college radio suggests a potential rejection of algorithm-driven music consumption in favor of more traditional and personal methods. As the music industry grapples with these shifts, streaming services are beginning to introduce features aimed at human curation. However, it remains to be seen whether these changes will significantly alter the current course of music recommendation systems, which prioritize user engagement over genuine music discovery.
Source: https://www.theverge.com/column/815744/music-recommendation-algorithms