Hacker News Reader

Comment by boomlinde | Hacker News Reader

> That's literally what charging for the copies does, though. People just want something for nothing.

I pay about 10€/mo for Spotify. A significant portion of my use of the service is devoted to artists that are dead. They can't put time and effort into producing music. They can't be reimbursed for the time and effort they've already put into it. I could bury the gold with them, but they wouldn't have any use for it because they're dead.

The difference is especially important because options are cropping up for actually funding artists and other creatives by paying for their work, not copies of the results. Kickstarter, Patreon etc.

View on HN

> I pay about 10€/mo for Spotify.

Consider not; the majority of that money just goes to Spotify. If, every month, you paid 10€ to a random artist you liked (or split 10€ between all of them), the artists would receive probably more than two orders of magnitude more money. (If I remember correctly, anyway.) Of course, this assumes you can get the music elsewhere.

If Spotify's too convenient to give up, there are several Spotify download tools you can use to get the music onto a device that doesn't support Spotify (e.g. MP3 players). Lots of them seem to actually grab from YouTube, but perhaps if you use the Spotify API, you might be able to do it?

A quick google says that spotify pays at least $0.003 per stream to an artist. So your $10 pays for 3333 streams per month, or just over 100 a day. I probably don't average 100 streams per day, but I definitely average over 10. I'm not sure its as bad a deal for the artist as you may think. It's not two orders of magnitude bad, at least.

The problem with this math is you're assuming a roughly even distribution for paying out to artists, when in fact the Spotify pay model wildly favors the popular artists in terms of paying out money. Basically if you aren't already famous, Spotify isn't going to pay you any meaningful money.

https://qz.com/1660465/the-way-spotify-and-apple-music-pays-...

I should probably mention I haven't really looked into Spotify's business model in about a year, so if something has changed and my information is outdated someone please let me know.

You also have to consider the cut for developers, lawyers, servers and everything else that goes behind getting music to you.

> the majority of that money just goes to Spotify

Last I checked, Spotify's cut is 30%. That's a lot, but even with a significant error it's nowhere near the "majority."

> If, every month, you paid 10€ to a random artist you liked (or split 10€ between all of them), the artists would receive probably more than two orders of magnitude more money.

That's a function of how popular your favorite artists are and how much you listen to them. No matter how unknown they are, if you listen to an average number of streams per month then they'll get at least 70% of your 10€/mo -- a far cry from two orders of magnitude. If your favorite artists are much above average popularity then they'll make more with the Spotify arrangement than if you gave them cash even if you don't listen to them at all.

There's still a kernel of truth in what you're pointing out -- at some point in the fairly recent past you had to buy the music outright no matter how often you would listen to it, so if there exists a long tail of musicians who aren't listened to much but whose music would have been paid for in a record/tape/CD/itunes market (and I suspect there does) then Spotify is going to be much worse for them.

> L ast I checked, Spotify's cut is 30%. That's a lot, but even with a significant error it's nowhere near the "majority."

It is in fact way way more than the majority because the remaining is split among artists while the cut is kept in totally by Spotify. Spotify is taking 3€ while any given artist would get a few cents at best.

A majority is more than half.

Kind of, but that's an artifact of not holding enough variables constant. If you were going to distribute $7 in cash to N artists then that's functionally equivalent to distributing $10 to Spotify and watching those N artists equally up to an average number of streams per month. The few main reasons that model isn't quite perfect are:

- You probably watch more artists via Spotify than you would via other methods, so each individual artist is paid less. Note this is less of an issue than it seems because on average there will be proportionally more people watching each of those artists.

- Artists may have cut other deals so they're getting much less than the 70% being distributed to "not Spotify." Note this isn't much different from pre-spotify, but it's a real effect.

- Something I left out completely in my original comment is that you very well might be paying less per month to Spotify than you would for other forms of media. I vaguely recall a stat saying that the opposite has happened mildly (but we pay a lot less for merchandise), but my memory is fuzzy on that point, so somebody who knows more should step in.

That isn't to say that Spotify is better for all artists (it's not), better for most artists (given typical pareto distributions my gut says that's probably also false), or that it's a "net positive" in some ill-defined way (maybe, maybe not, that's not an argument I want to try to make) -- just that Spotify's payment model isn't super complicated or super unfair at a macroscopic level, and it's super easy to construct reasonable scenarios where your dollars would have more positive impact on artists via Spotify than if you just paid those artists cash.

Ouch. So 30% per subscriber, not 30% per artist which is what we would usually mean by a cut. A classic instance of "two for you two, and two for me, too!".

> No matter how unknown they are, if you listen to an average number of streams per month then they'll get at least 70% of your 10€/mo

From what I understand the number of listens that month is pooled globally/in buckets for the purpose of distribution. So if you listen to your friend's band once and nothing else, and someone plays a 31 second song on repeat 83612x over 30 days nonstop, your friend's band gets {{payout rate}}*1. And the other artist gets {{payout rate}}*83612. Rather than both artists getting ~70% of their listener's $10 subscription.

And then the payout rate is a bit complicated because it comes from a variety of buckets. Some regions earn more/less than others, some account types (free trial, free with ads, family plan) earn less than others. Some specific labels or artists even get their own special buckets. 30% take overall is still correct. If you check their 2020 Q4 "Cost of Revenue" was about 73.5%, which is almost entirely royalties.

In practice though, I don't think it matters too much. I don't think many people listen to one artist exclusively. Or if they do, it's probably a pop or children's music artist getting many streams already. But in the rare case of some person listen to one specific relatively unknown artist as most of their streaming, I'm pretty certain that artist does not receive an equal majority of that subscriber's money.

> From what I understand the number of listens that month is pooled globally/in buckets for the purpose of distribution.

Yep, exactly. Views are pooled per stream rather than per artist, which is its own can of worms (I mentioned that somewhere in this thread), but if you watch an "average" amount that already accounts for such an artifact. At least 70% of your contributions go to the artists you listen to in such a scenario (that money has to come from somewhere; many artists _don't_ get paid much).

> And then the payout rate is a bit complicated because it comes from a variety of buckets.

Yep, that's why I simplified it as much as possible in the context of the net effect of your personal actions. Your paid listens will impact one bucket and not the others, and that allows them to be safely ignored when evaluating the difference in an artist's pay with you watching them on Spotify or just giving them cash.

> In practice though, I don't think it matters too much. I don't think many people listen to one artist exclusively.

This is a point I'm seeing crop up a lot, not just from you. I think we might be talking past each other; I'll lay out why I don't think that matters, and I'm curious to know what you think.

In particular, if you're comparing spending X per month on Spotify or spending X per month on other forms of music, in both scenarios you would be splitting that between the artists you actually care about. I'm pretty sure people spend more per month on music than they used to (still haven't found the original source, so maybe that idea is wrong), so IMO it doesn't make sense when comparing Spotify to alternatives to point out that in _both_ instances your money doesn't all go to one artist.

There is a point to be made that you'll probably listen to a broader variety of artists through Spotify than otherwise (why not? that's one of their core value propositions), so your individual X payment is split further than it would have been. However, everyone else is listening to more people too, and on average assuming everyone listens to the same distribution of music as in a pre-Spotify world those effects actually cancel out.

There is still an open question of whether that distribution is the same -- does there exist a pool of artists being listened to via Spotify who would rarely have their music purchased otherwise? My gut says "yes," but I don't actually know. Is it a bad thing for those artists to be paid?

Spotify pays a lot more than radio plays, and radio plays reach a lot more people. From this POV, artists aren't that hard done by.

Spotify plays a lot less per play, on average, than most albums sold. From this POV, artists are pretty hard done by.

If you're a fan, buy the album (i.e. pay even for the music you don't like as much). If you only listen to the music ambiently, stream and don't feel too guilty about it.

Reply on news.ycombinator.com