Now worth around 25 billion dollars, Spotify is currently the world’s most valuable music company. Spotify completely shifted the music industry, not only in terms of making streaming music the current standard but also in how customers interact with their platform via the use of big data collection. With over 100 million paying subscribers, they have a lot of data that they can derive meaning from.
Machine Learning Creating Custom Content
Spotify’s Discover feature has evolved since it was first introduced in 2012. It began as a playlist of tracks released by users’ favorite artists. From there, the streaming giant introduced recommended songs at the end of playlists that were similar to the playlists content. Fast forward to today, the Discover Weekly is Spotify’s bread and butter proving to be one of their most popular features.
Discover Weekly creates a custom playlist, unique to each users activity. This is curated entirely by a machine learning algorithm that analyzes other users playlists finding commonalities between tracks to create a playlist just for you. Additionally, each user has a ‘taste profile’ comprised of microgenres that help further customize the playlists. Quartz, a global economic news source that focuses on the changing media landscape, developed a helpful visual of this process:
Spotify is paying attention to not only what you listen to, but how you listen, and what you do next. If you play a song and change it within the first 30 seconds, spotify recognizes this as a ‘thumbs down’ and won’t use it when calculating playlists. However, adding a song to a playlist or your library and listening to it in full helps the algorithm develop your taste profile further.
Digitization of Your Taste
Your taste profile is also used in a section called Daily Mixes. These playlists are separated by the genres of music you typically gravitate towards ( I personally have one dedicated to 80’s). The Daily Mix is comprised of songs you have saved or added to playlists, tracks by the same artists, and new music you may not know. These playlists are bottomless and ever-changing. They tend to have more familiar content than the Discover Weekly playlists, but still sprinkle in some interesting tracks you may not know for variety.
Who’s On Your Radar?
Release Radar is a weekly playlist comprised of new releases from artists you follow, similar the original Discover playlist. To get the most out of this playlist, you need to follow artists you enjoy as you listen to them. Following artists helps the algorithm develop a more accurate playlist of new songs from an artist. The algorithm also will add a few extra new songs to keep your mix fresh.
Using Data to Market
With all of this data based on musical preferences available, Spotify is also able to strategically market to their audiences. Innovation Enterprise highlights an ad run in Williamsburg, NY, a known hipster area: “‘Sorry, Not Sorry Williamsburg, Bieber’s hit trended highest in this zip code was displayed on a huge street advertisement, a piece of marketing made engaging and funny only by virtue of its localization.” This seems especially comical due to the insane popularity of Justin Bieber in an area with a notoriously high concentration of ‘music snobs.’ The thought is, if a marketing campaign worked in one area where a particular music preference was shown, a similar campaign may work somewhere else with similar musical interest based on data.
More Freedom, But At A Cost
On April 25th, the music giant announced their free users would no longer be forced to shuffle music, but have the freedom to explore 15 of their most popular playlists including RapCaviar and Discover Weekly. This is a win-win for everyone. Free users get to experience some of Spotify’s top content, and now Spotify can utilize data from as many as 90 million more users on these playlists! New meaning can possibly be derived as it opens up new variables of free vs. paying customers and their listening habits. It will also help in further developing purely algorithmic playlists like Discover Weekly
The music industry is on the largest in the world. Since the shift towards streaming music over purchasing began, the industry has had to rely more on collecting data in order to decipher how the public responds to music. Additionally, they have new insight into markets they haven’t previously tapped into as deeply. Spotify will continue to be one of largest music data sources for some time. Using data to make better business decisions is the future of most industries. Take the time to make yourself ready for that future.