How do Spotify and Apple Music pay artists? — Media

Have you ever wondered how major streaming services pay artists? You would think that your subscription fee or advertisement revenue goes to the artists that you listen to, right?

This is a question I have been thinking about lately and the answer is probably a little different than you might’ve expected. I’m very passionate about music and Spotify as a company, and this made me want to learn more about how artists benefit from uploading their music to Spotify.

Spotify and Apple Music’s Distribution Model

Note: Spotify and Apple Music keep roughly 30% of the money collected and give the rest to the composers, artists, publishers, producers, etc.

Pro Rata

The way these major companies pay artists is through a distribution model called pro rata. This system pools all the money generated by users and then divides it up amongst all the artists based on their total number of streams. So if Ed Sheeran’s music is 1% of all the streams on Spotify, then 1% of what I’m paying for Spotify goes to Ed Sheeran. My money gets divided proportionally amongst all artists based on their total number of streams and who I listen to does not have a major impact on which artist receives my money.

Pro rata is a Latin term that translates to “in proportion”, which makes sense since the system requires dividing money proportionally to all artists.

User-centric

Now, I know what you’re thinking. Why can’t all my money just go to the artists I listen to? Shouldn’t my money go to them since I listen to their music? Nope, but this model does exist. This type of distribution model is called the user-centric system. This model is centered around the user and ensures that the artists you stream are the artists that receive your money. This is done by splitting up your money amongst the artists based on how many times you stream each artist.

What Distribution Model is Best?

It is widely expressed that the user-centric model is preferred. This is because consumers want to support artists they like. The pro rata system doesn’t seem fair to a lot of people because they aren’t directly supporting the artists they listen to. If I only listen to NF and Khalid, then I want NF and Khalid to receive all the money I am paying to stream their music. Implementing a user-centric model would theoretically decrease the amount of money larger artists receive and would increase the amount smaller artists receive.

While user-centric may seem like the best model, these streaming services actually have a very good reason for not immediately moving over to this model. Will Page, the Director of Economics at Spotify, was the co-author of a paper suggesting that implementing a user-centric model would most likely lead to all artists receiving less money because of how expensive it would be to maintain. I’m currently a college student studying Computer Science and I’m understanding more and more how difficult this would be for computers dealing with hundreds of millions of users. Spotify would now have to keep track of every user’s individual stream in a database in order to divide the user’s money amongst the artists properly. It would be so computationally heavy and expensive for Spotify to maintain that it could very likely lead to all artists receiving less money.

Even though most major streaming services like Spotify and Apple Music use pro rata, some other smaller companies, such as Deezer, are moving into user-centric type systems.

Conclusion

The pro rata system is what most companies use due to its low-cost maintenance while still distributing money proportionally to all the artists. Although the pro rata system may not be the preferred distribution model from the consumer point of view, computers are evolving quickly and maybe soon enough the user-centric model will be more realistic to implement.

Thank you so much for taking the time to read this. I hope you learned more about how major streaming services pay the artists you listen to, and I would love to hear your opinions on the pro rata and user-centric distribution models.