I mean, a more fair way would be to split each subscriber's bill by the artists _they_ listened to. That would favor small / local artists more.

There were multiple responses questioning how this was different, so I'll respond here:

- 2 users, A and B, each pay $10/month

- User A listens to 20 Taylor Swift songs

- User B listens to 5 Radiohead songs

- Spotify gets a 30% cut

Currently:

- Spotify gets $6

- Taylor Swift gets $11.20

- Radiohead gets $2.80

If each subscriber's bill was split separately:

- Spotify gets $6

- Taylor Swift gets $7

- Radiohead gets $7

So this change would benefit artists that less active Spotify users listen to.

Right now the less active users are paying to support the listening of the most active users.

Personally this would make me feel like my money is more directly supporting the artists I care about.

Why is everyone ignoring the user side? It's not that there are equal users and a Taylor Swift person just listens more, there are millions of more users listening to Taylor Swift. Radiohead could actually do better with the much larger user base Swift brings. Without more data, it's hard to know.

I was going to make an example of what if restaurants used the Spotify model and how people would hate it because they'd be paying for other people's tables but I guess that's buffets and all you can eat sushi places are like. Not exactly shining examples of quality for the end user though.

> - Spotify gets a 30% cut

This is where you're wrong..

1. The article literally says ‘ “Spotify already pays nearly 70% of every dollar it generates from music to the record labels and publishers that own the rights for music, and represent and pay artists and songwriters," it continues.’

2. It’s a simplified example for explanatory purposes

Sure, that's the general narrative of spotify.. But it's false. It's more like half of that 70%.

There's a reason Spotify is ending their service in Uruguay, and it's not because they're enough.

I don’t think that would necessarily favor small artists. That would just favor artists who are listened to by people who don’t use Spotify a lot.

Right now someone who only streams a few songs gets a very small “vote” (assuming pay is per stream). That would make it so that everyone had the same “voting” power. But I doubt there’s much correlation between people who use Spotify less and small artists. In fact that’s probably a negative correlation if anything, and this could end up hurting local artists.

I'm not sure about spotify, but I have seen this problem with Netflix: I'm an adult paying full for my Netflix subscription and for that price I would love to have a few great and expensive movies every month for my age group.

But what I usually see is lots of movies made for teens who are binge watching Netflix even though they are paying the same. Netflix has some public presentations on their algorithm and from the presentations it looks like they are optimizing for watches without weighting by subscription revenue per watch.

Sorry, I’m failing to see the difference between this and what actually happens. They currently take 70% of their revenue and split it across artists by stream, and what I’m reading is that you’re proposing is that we take each subscriber’s revenue (presumably less some profit factor - say, 30%) and split it by the artists they stream. What am I missing?

Fwiw, the actual Spotify math is a bit more complicated than just splitting by streams - I.e., https://www.theverge.com/2023/11/21/23971616/spotify-royalti...

It's different because if you imagine Spotify only had 2 users, Alice and Bob, and Alice listens to Indie Band X 10 times a month, and Bob listens to Megastar Y 10,000 times in a month, Y would receive almost all of the artist revenue in the current model, but in the model that GP is proposing, both artists would make the same.

Essentially A's money is going towards Y, even though she never listed to Y.

(I'm not expressing an opinion on which of these I prefer)

It can be a huge difference. Say this month I only listen to a completely unknown artist. I am the only person listening (to make the calculation easy). I've listened to 10 of his songs and this is everything I've heard this month.

Now if my 7$ are distributed to all artists I've listened to this artist will now receive 7$. If my 7$ is thrown in a global pool and split by global stream counts this artist gets almost nothing.

Of course this is a extreme example but it should illustrate why this can matter.

> If my 7$ is thrown in a global pool and split by global stream counts this artist gets almost nothing.

Not by definition. If you generate more streams this month than the average user, your streams are a larger fraction of the stream counts than your 7$ is of the revenue pool, and this artist will get paid more than if he only got your 7$.

Splitting the revenue per-user instead of per-stream can certainly make a huge difference, but switching to per-user instead of per-stream benefits artists with listeners that generate a below-average amount of streams per month, regardless of their absolute stream numbers.

Their current split favors big artists. 90% of the music I listen to is jazz and classic, and if I was on Spotify (I use Qobuz instead), 90% of my subscription (minus Spotify's share) wouldn't go to these artists, instead most of it would go to the most popular artists on Spotify. That's not fair, I am de facto subsidizing the biggest artists, even if I don't listen to them, while the artists I like receive less.

And there's the fact that the payment per stream is low on Spotify compared to Qobuz or Tidal.

It’s the weighting. If you listen to very little music, but 1 artist, your $10 (in that scenario) would go to them.

Instead, your listening is not weighted in any special way so it goes into the global pot of money and hours listened and gets diluted by the people listening more.

If I'm listening to only Primus for a month basically none of my subscription goes to Primus.

Alice pays 10$ per month and play (only) a thousand times a song of Madonna each month Bob pays 10$ per month and play (only) a hundred times a song of Queen each month

Currently Madonna will get 1000/1100 of artist's share => 10/11 x 70% x 20 = 12,73$ While Queen will get 100/1100 => 1/11 x 70% x 20 = 1.27$

With a per subscriber split => Queen and Madonna will get 7$

Right now there is an incentive for artists to create shorter songs so they get more stream = more revenue

Not sure how this favors small/local artists. Most people I know who listen to small/local artists listen to much more music than those who listen to big pop artists. That means the current system actually favors the local artists since their listeners have an outsize “vote” when all streams are bundled together.

Is there any reason to believe big artist listeners on average stream more songs per month?

A better way would be for streams to interoperate with a copyright clearance mechanism where, when an artist opts out of the regular stream users still can validate they own a copy and get access through the stream. I have a library of saved songs I never hear on Pandora, and often my favorite artists and songs get dropped leaving a rather bland experience over the long haul.

But in terms of unit economics, the bigger artists/labels (or the artists that bring more users) are responsible for funding the entire platform, no?

In other words: A small artist can _bring_ users to the platform, fair enough, but the platform that gives access to other artists and enables that needs to take the bigger share.

>are responsible for funding the entire platform, no?

No, they all are. I wouldn't be a subscriber and I doubt many others would be if they only had the "top" artists that "are responsible for funding the entire platform".

Go through your playlist for a minute and find how many artists aren't "the big ones" and ask yourself if you'd use the platform if the small artists weren't included. The whole point of spotify is ALL music, so ALL music is important, surely?

Articulating a bit more in depth: Spotify it’s a multi sided marketplace where on the artist side will have a very big power law where the big labels (that has a catalog power) brings most of the users.

What sustain the economics on the Spotify is its catalog that is a composition of the labels catalog + long tail artists and their podcasts also.

When users goes to the Spotify, they are paying not to support their favourite artist but to be able to access the platform, that occasionally will have their long tail artists.

> The whole point of spotify is ALL music, so ALL music is important, surely?

Not necessarily. Catalog + Network effects + Marketplace is important.

What you’re describing in your argument is that “since you’re in the platform due to the artist, it’s fair that this artist receives the 70% of your money”. The flaw here is that this artist is benefiting of a marketplace and all its infrastructure and distribution without paying and the whole point of the platform is to give access to its catalog.

Think on the Spotify price as some kind of “optionality fee”.

No, users are funding the platform not artists. If someone doesn’t listen to major artists then revenue from them shouldn’t go to said creators.

Spotify however wants to maximize its cut and so has incentives to shift money from less popular artists if it can entice major labels to its platform. This has absolutely nothing to do with what’s fair, it’s simply about profit maximization and reducing the risks of competing platforms.

> If someone doesn’t listen to major artists then revenue from them shouldn’t go to said creators.

Well I do not have the numbers here, but I would argue that this is definitely not the case in reality.

> Spotify however wants to maximize its cut and so has incentives to shift money from less popular artists if it can entice major labels to its platform

It’s the other way around, because the platform has competing interests: Spotify wants to have the labels to sell an “optionality” to the users (I.e. get the money from subs) but in terms of reproduction/execution Spotify wants to reproduce the maximum of podcasts and/or long tail artists because in those 2 cases Spotify does not spend a single penny.

The labels has a very high bargaining power over Spotify due to its catalog, so as expected this is the biggest expenditure.

That’s one of the reasons why Spotify is desperately pushing podcasts like crazy.

> this is definitely not the case in reality.

Agreed.

Major labels have leverage which they use to get a larger slice of that 70% than they would otherwise make in a ‘fair’ split.

This is what happens already, isn't it?

No. If you listen to only a single artist for a month, your payment is still split the same way everyone else’s payment is, by total streamed numbers.

It sounds better if they artist got your entire monthly payment (after payment processors and Spotify’s fee), but probably a nightmare for accountants.

I don't see how that would be different.

Do artists have a contract with Spotify or with their labels? If it's with the labels, Spotify pays the labels and they give to the artists what they agreed upon, possibly an unfairly low amount of money but contracts are contracts.

The Uruguay parliament should regulate the contracts between their local artists and the labels. Some of them will ban Uruguay artists but some of them, maybe local ones, will intercept the money no matter what.

Or Spotify makes deals with individual artists, but there is a long and thin tail of them.

Small indie artists will have contracts with Spotify. Larger artists with labels will have contracts with their labels, who will have contracts with Spotify. As you suspected, the labels typically take a ridiculous cut of streaming revenue.

> Small indie artists will have contracts with Spotify.

They won't. They will have contracts with distributors: https://artists.spotify.com/providers

My subscription money ends up going to subsidise super popular artists I never listen to, rather that the artists I actually do listen to.

I suppose the argument wouldn't be about how their revenue is distributed, rather that they should be charging more for their subscription (which is, in my opinion, insanely cheap). They're able to keep it that cheap because they're underpaying artists.

They put all the streams in one pool and then split from there. Popular artists get almost all of it.

When you spend $10 or whatever on streaming, you'd expect $7 to be split between the artists you listen to, but it's not. Instead almost all of it goes to the big artists anyway because in the big streams pool the small artists you listen to have a tiny proportion of listens, even if you exclusively listened to them.

This gets worse when you consider all those shops/cafes/etc with Spotify's top charts playing on repeat. They may be a small percentage of users, but they're an outsized proportion of streams, and almost exclusively big artists.

This is the way it is because big artists have big name labels, and those labels are the only ones who can negotiate, so the deal ends up favoring the big artists/labels.

This is definitely not how most consumers would expect it to all work, and therefore "unfair" seems like a reasonable term to use.

The "pool all streams" model doesn't necessarily favor big artists, it favors artists that are listened to by people that stream a lot of music.

For example, consider a bimodal population of 1000 people that listen 10 times to a popular artist (for a total of 10,000 streams) and 10 people that listen 1000 times to a small artist (for an identical total of 10,000 streams). In the "pool all streams" model, both artists get paid the same, while in a "split per user" model, the popular artist gets 100x the payout of the small artist.

An interesting question then is whether big or small artists are disproportionaly listened to by people that stream a lot. I haven't seen any data on it, but intuitively it wouldn't surprise me if it's the latter case.

You're right that it also favours artists that are listened to by people that stream a lot of music, but it does favour big artists too. Big artists can make up their numbers by lots of streams from few users, few streams from lots of users, or both.

Small artists tend to get a lot of streams from very few users, and lose out, so this suggests that earnings are more correlated to unique listeners than to average listens per user.

If with lose out you mean that it's not financially sustainable for them, then it's not at all clear to me that's caused by the way Spotify splits revenue, instead of Spotify just being too cheap for users to make streaming viable. Spotify takes in less than five bucks per Premium user per month, so after subtracting costs you need to quite a bit of listeners to make a living, even if Spotify shares it payout per user.

If with lose out you mean that Spotify pays out less per stream to small artists as opposed to big artists, then that's not the claim made in the parent comment nor one I've been able to find any evidence for.

My understanding is that the amount of money spent on music by consumers has dramatically increased in the streaming age. It's not clear to me that Spotify only passing on ~half after fees is any different to a CD bought in a shop that pays rent and needs manufacturing, physical distribution, etc.

I'm fairly sure this is not about Spotify being too cheap, it's that the revenue distribution skewed in the transition from CDs to streaming in favour of popular artists.

> "They put all the streams in one pool and then split from there. Popular artists get almost all of it."

Like they should, mostly because they drive Spotify subscriptions. It's fair because there is a strong correlation between popularity and work, coupled with talent.

> When you spend $10 or whatever on streaming, you'd expect $7 to be split between the artists you listen to, but it's not.

Even if they'd distribute your subscription to your favorite artists, the total revenue for those artists and their share would still be mostly the same. Only in some degenerate scenarios, where you listen to only one obscure artist that nobody listens to, would those numbers be any different, and not by much.

> This is the way it is because big artists have big name labels.

Economies of scale probably apply. Like they should. ;-)

Talented artists can often sign with big labels. And isn't it wonderful that in 2023 you can publish your work, on Spotify and elsewhere, without having to sign with a big label?

> they drive Spotify subscriptions

There is no reason to believe this is true - I can download all the main artists in 30 minutes. Its the niche ones that I enjoy discpvering and wouldnt bother to download

> there is a strong correlation between popularity and work, coupled with talent.

Even stronger correlation between marketing spending and popularity

> Only in some degenerate scenarios, where you listen to only one obscure artist that nobody listens to,

This is waay too disrespectfull

I think he’s using degenerate in the mathematical sense.

> They put all the streams in one pool and then split from there. Popular artists get almost all of it.

... but that's because popular artists account for almost all the streams, right? I don't think a per-user split is any fairer, it's just different, and there's no clear evidence on whether there'd be less of a gap between big and small artists with a user-centric split.

Tidal (spotify competitor) takes 2 euro and gives 'em to the artist you've listened to the most. But the rest of the money is probably all going to the big ones as you explained.

Just saying this because it is so weird. Like, hell yeah I sure hope my money goes to the people that I listen to why is this a thing??

> and gives 'em to the artist you've listened to the most

This seems like a way to sound popular without actually solving the problem. IT sounds like it would help "yay my favourite artist gets lots of money!" but when you model it out I'd expect quite similar outcomes – most of your money is still distributed in the way Spotify do, and often your top artist might still be one of the popular ones, and small artists aren't going to be the top of many people's listening.

What am I missing?

You'd think it should still be fair because money is supposed to be fungible.

So if Artist 1 gets 99% plays, and artist 2 gets 1% plays, then if you split the 70% of revenue at end of month by 99/1 , that should be fair, right?

I'm probably missing something important. Is there a source for the actual formula or so?

Users U1 and U2 have a the same subscription, and $10 of the fee they pay will be distributed to artists.

User U1 listens to tracks from artist A1 ten times.

User U2 listens to tracks from artist A2 190 times.

The $20 that U1 and U2 have paid are distributed to artists according to number of songs streamed. Artist A1 gets $1, and artist A2 gets $19.

It's counter intuitive that U1 effectively pays $9 to A2, without having listened to their songs even once.

The counter is also true that A1 gets money from people who have no idea A1 even exists. It's hard to say if one method is better or worse for the smaller artist. Basically, is it better to get a bigger piece of a much smaller pie or smaller piece of a much bigger pie. Without the data it's hard to know.

EDIT

I'll also add that the big name artists have value beyond the number of streams. A platform without Taylor Swift will have less overall users, less money for the royalty pool, less money for platform development, etc...

An interesting thought idea would be to create a streaming platform which only has artists outside the top..50? and see how it does.

Meanwhile U2-U99999 pay money towards A1 without having listened to them

I tried to pick a simple scenario to make it easy to see the faulty model, and assume that this version of Spotify has exactly two users and a catalog of exactly two artists.

The imbalance remains even if you scale it up. Artists that are listened to by users that listen to less than the average number of streams are underpaid in relation to the revenue they generate for Spotify.

If I listen to 10 songs in a month all by artist 2 and you listen to 1000 songs by artist 1, should my subscription money all go to artist 2 or should it 'subsidise' your listening to artist 1. That is the core question.

Basically do you put all the money in a pot and then split it, or do you treat each subscriber as a unit and split their subscription fee just among the music they listen to.

If I listen to artist 2 only I'd like all my money to go to artist 2, regardless of how popular artist 1 is.

That's how it worked with CDs: I buy the band I like and they get whatever pennies the mafia leaves them. But the pennies go from me to the band I like and to them only, not some $BIGPOPSTAR instead that I never listened to.

Let's say you pay $10 and I pay $10, you listen to 200 songs by Simply Red, while I listen to 50 songs by A-ha.

Spotify takes $14 (70%) of our money and divides it by 250, this means that Simply Red gets 80% of $20 or $11,20 and A-ha gets $2,80.

The unfair part is that my money got to pay for an artist I never listened to. Why not my share moved to A-ha?

> 80% of $20

Should of course be 80% of $14, $11,20 is still correct though.

Spotify charges a flat monthly fee to their customers, so the income per stream varies widely: Someone who listens to only one album a month is providing more income per stream to Spotify than someone who leaves it running as background 24/7, but the artist that recorded that one album doesn't get a proportional share of the income generated.

This is something twitch-style subscriptions get largely right.

> you'd expect

I don't. I expect that when I pay for a streaming service, as soon as the money is in their account all bets are off and no insight is given.

So such a generalisation of what people expect doesn't hold up for consumer products I'd say. I wouldn't be surprised if most customers don't even think about it at all.

That's false information.. Spotify does NOT distribute 70% of their revenue.

As per their latest financial statement ( https://s29.q4cdn.com/175625835/files/doc_financials/2023/q3... ):

  Revenue:         3357
  Cost of revenue: 2472

> Cost of revenue consists predominantly of royalty and distribution costs related to content streaming.

> Cost of revenue also includes the cost of podcast content assets (both produced and licensed)

> Cost of revenue also includes credit card and payment processing fees for subscription revenue, customer service, certain employee compensation and benefits, cloud computing, streaming, facility, and equipment costs.

So all operational costs is included in that "70%", including: infrastructure, payment fees, etc. etc.

A breakdown:

  217 committed Google Cloud Platform per quarter (based on 4344 in 5 years)
  105 for amortization of content assets (podcasts they bought)
   30 for some content write-off
  102 "minimum royalty" for licensed content (not the regular artists), (based on (2124-83) in 5 years)

  100 my estimate for app store fees, assuming 15% fees for 20% of the users
   67 my estimate for ca. 2% creditcard fees
    ? streaming
    ? facility
    ? support software + employees
    ? equipment, software, etc
    ? refunds etc
    ? free spotify for employees

So a total of AT LEAST 723 is not related to paying out royalties. (2472-723)/3357

They're paying less than 52%. My estimate would be between 35% and 40%

Those revenue and cost of revenue figures blend together different things. Their contracts with music distributors (recordings) + statutory rates for music publishers (compositions) = ~70% of the revenue pool from advertisers and subscribers.

However, podcasts work differently, both on the revenue and cost side, and appears to be mixed in. Additionally, Spotify sells optional marketing programs (ads within Spotify, for example) which I assume are included in revenue. So it's hard to draw the right conclusion just from the summary numbers.

Yep, the unfairness of streaming starts when the money has left Spotify.

They need to at least split it by stream time. It’s non-sensical to have a 2 minute pop song be paid the same as a 14 minute classical song. This has weird off effects like pop stars making shorter and shorter songs so they get played more often, which causes less ability to concentrate for extended periods of time in the young people that listen to it.