Spotify is an online and app-based music listening platform that generates revenue from monthly subscriptions for unlimited use. The company is also branching into original content in podcasts and in-studio recordings with artists. Artists are paid by Spotify per listen in exchange for the license to include their music in Spotify’s library.
Spotify targets music listeners, especially mobile listeners using their phones.
To grow its company and revenue, Spotify aims to grow their monthly subscribers and retain existing customers.
Spotify can help users discover new music that they will enjoy listening to, both recently released and music that was released in the past but might be new to the user.
Spotify currently creates a “Discover weekly” playlist issued every Monday and customized to each user based on a user’s past listening preferences. Spotify also makes a “Recently released” playlist issued on Fridays also customized for each user. These types of recommendations are created by Spotify using collaborative filtering, comparing users who have similar historical playlists and then recommending new music to the other user where there are unique songs. As described by Adam Pasick on Quartz (link below): “In the simplest terms, if Spotify notices that two of your favorite songs tend to appear on playlists along with a third song you haven’t heard before, it will suggest the new song to you.”
To improve the quality of its recommendations, Spotify should incorporate length of play into its algorithm so that songs a user plays for only a few seconds and then abandons would be stricken from the algorithm. A user skipping forward through a song is an indication they did not like the song, and the algorithm should learn from that action.