The other issue that has been raised a number of times is that I believe the revenue is still divided up based on overall listens across the platform, which means it's dominated by artists who feature heavily in popular playlists and automation-driver radio/recommendations.

These listens are often passive and should not be considered as valuable as those of someone listens only a few hours a month, but spends all of that time listening to specific artists (indie or otherwise).

Basically the argument is that if I spend $9.99 on a Spotify sub, the entirety of that cash (minus Spotify's cut) should be directed to the actual artists I listened to.

Here's an articulation of this perspective from 2014: https://medium.com/cuepoint/how-to-make-streaming-royalties-...

If there's two groups of people, one who listens to pop music for 10 hours a month and one who listens who indie artists for 20 hours a month (which could easily be the case if their recommended playlists are mostly indie artists as mine is), then switching to the model you propose would reduce how much money indie artists make, right?

It seems more fair to me that artists would get a set amount of money per stream, rather than the amount they get being dependent on how many other songs their listeners listened to.

I think the argument is not that it's the 10 vs 20 hour listeners, but rather the 10-40 hour listeners (workday) vs the 700 hour listeners (radio playing in the corner of a breakroom somewhere).

The 700 hour listener has no audience and no human curation, and yet it's got 10x the influence on who gets paid by Spotify.

The article I linked also makes a really good point about this model motivating artists to encourage listeners/fans to listen to whole albums rather than single songs— right now if you're playing Spotify for revenue, the only thing that matters is landing your songs on popular playlists where they will get hundreds of thousands of passive plays. The incentives there are not really aligned to what anyone wants long term out of music.

It’s not inherently clear to me that one play should be paid more than another play.

In fact, having a different fee per play dependent on the user is a pretty opaque and unclear metric. “How much are 1 million plays worth? It depends on your audience - if your listeners listen to lots of music you get paid less per play, and you will just have to trust us to compute an appropriate fee for you correctly.”

No matter how you cut these things there are pluses and minuses . You seem to think that Universal should get paid less for a single Katy Perry stream played from a global playlist compared to a small indie song streamed in a private playlist, and while that’s a valid opinion it’s not intrinsic or even self evident to me.

I don't think it would be that hard to show a small histogram which had your plays bucketed across different levels of commitment, engagement or some similar metric. Or instead of doing it that way, you'd bucket the users, so say "for song X, you had 30k unique users listen to it in the royalty time period, and here's a histogram of how engaged they were, with the left side being a small number of users who listened to it a lot or didn't listen to much else, and the right side being the large pool of users who heard it once and listened to many other things.

The reason this is good for artists is that it rewards them for what many want to be cultivating anyway: a relationship with enthusiastic fans, rather than faceless "plays" which could have come from anywhere.

In any case, obviously the conversation is DOA if you start it from a foundational assumption that "a play is a play", but if we acknowledge upfront that artist compensation has always been weird and often quite unfair (see radio royalties, for example— they sure as hell aren't getting compensated by how many people were tuned into a radio station playing their song), then there's room to discuss better options in a new medium that sits somewhere in between the old roles filled by FM radio and private collections of physical media.

> In any case, obviously the conversation is DOA if you start it from a foundational assumption that "a play is a play"

That's not my foundational assumption, but I've also not heard a compelling argument that one play / minute streamed should be inherently more valuable than another play / minute streamed.

It's also not clear how this value would be calculated, because I also don't think it's clear that proportion of listen time by user necessarily captures the difference in 'value' of one play against another.

It also feels like it will be very complex to negotiate with rights providers. Spotify's current negotiations with rights providers are really simple and easy to understand - we divvy up our revenue by play count, you are responsible for $x revenue by this metric, so let's negotiate the %. Moving all providers to another weird metric would be an extremely tough thing to do, particularly because having half on one and half on another would likely happen for a crossover, and would result in a loss of profit during the transition period.

I don't think it'll necessarily happen via Spotify renegotiating it. It'll happen via a competitor offering this scheme, and indie labels pulling their content from Spotify.

The obvious player to offer this competing service is Bandcamp. They haven't yet, though an awful lot of the pieces are in place— they have a mobile app that lets you trivially stream anything you've ever purchased and listen on trial to things you haven't, and it even has genre playlists and recommendations:

https://www.npr.org/2020/08/19/903547253/a-tale-of-two-ecosy...

It's tough, because indie labels will likely find it hard to both remain popular/relevant while pulling all their songs off the worlds most popular listening platforms.

It's a bit of a catch 22 - these platforms are now where people discover new bands, so if you are not on them people don't discover you so can't buy your album, and if you are on them then people can listen to your album without paying.

Unfortunately I think the truth is that in the future bands will have to think of slightly different monetisation strategies - maybe a mix of subscriber content (see Pomplamoose Patreon as the creators of this concept), fan engagement, merchandise, shows, premium albums that aren't on streaming platforms e.t.c. rather than pulling from Spotify & Apple Music entirely.

> I've also not heard a compelling argument that one play / minute streamed should be inherently more valuable than another play / minute streamed.

Here's a compelling irrefutable argument.

Let's start the discussion from a couple of scenarios. These scenarios are not the definitive and complete set of use-cases, but are enough to prove a point.

Scenario 1: User manually searches for a song and clicks play. Again the following day. Again and again for a couple of weeks/months. The user forgets about the song after that period.

Scenario 2: User creates a playlist and clicks play. Then uses that same playlist again, and again and again for years. Alterantive scenario: users searches for a published album/compilation and plays that playlist again and again for years.

Scenario 3: Streaming service generates a new generic infinite playlist every day, based on a biased algorithm that upvotes some artists and downvotes some other artists. E.g. Playlist called - Discover Daily/Weekly.

Scenario 4. Streaming service generates a new generic infinite playlist every day, but personalized to the users taste and interests, again based on a biased algorithm that upvotes some artists and downvotes some other artists. E.g. Playlist called - Recommended for you today.

Scenario 5. User hits random play and a truly random algorithm decides what will be played next, no matter of musical style, language, culture or any other aspect. Truly random, no biases whatsoever.

So I will argue why not all streams are equal or should not be treated equal.

The difference is the specificity of intent. Whether the user wished to listen to some specific song, or if the user was force-fed some song. For the initial discussion I am assuming force-feeding, later I will discuss alternative scenarios (clicking skip).

In scenario 1 and 2 there is intent from the user what should be streamed, and this means that the user is interested in the music to be streamed. It can not be refuted that the authors of that specific song should be paid for the stream.

In scenario 3 and 4, the user just wishes to have something making noise, not anything specific, but not random. This is fine, if the user is satisfied, but the problem is all algorithms have biases. If the algorithm favors the most popular authors, and the user could be satisfied with anything in the same style and tempo without preferences, than the streaming service is not fair to the authors who are less popular and never gives them a standing chance to compete. This creates a boosting effect so that the most popular become force-fed to the audience involuntarily and they become even more popular. This is a well-known side-effect in all platforms that use popularity as ranking or rating metric. So here, not all streams can be considered equal, since some had higher probability to be present in the playlist due to inherent algorithm biases.

Scenario 5 is the only scenario where the user is force-fed music, but in a manner that is irrefutably unbiased and irrefutably fair to all authors and songs. In such a scenario all get a chance and all such streams are equal. Unfortunately, users that would prefer such a scenario do not exist. Nobody would endure the torture of a truly random playlist for more than 3 songs.

So we have use cases where the users know what they want, and use-cases where the users don't know what they want, but know what they don't want and the streaming service tries to guess what it is that they want next, without truly knowing.

How do the streaming services know what a user does not want? By measuring the % of the stream that went through. If the user managed to suffer only <5 secs before clicking skip, then that stream should not be counted at all towards payment to the musician. If it was 60 secs than it's a different case. If the song went through entirely it's even better. If the user streamed the entire song, and was actively using the app during the whole time than it is an indicator that the user appreciates the song even more than if a passive encounter occurred.

So they try to guess what to play next. With a biased recommender algorithm that tries to recommend what would be a good fit to play next, and as I said the algorithm might favor some musicians over others.

This is a hard problem and is always discussed at research conferences, both by researchers from the streaming services, and by independent researchers. The topics of bias, randomness, satisfaction are always present. Some authors are even using personality traits to discover the type of music that should be streamed next, which is a very very deep breach in user privacy. I imagine that some services might consider eaves dropping on their customers to evaluate if there is true interest and appreciation of the served content - whether the customer is eating, dancing, smiling, chatting, ..., having sex, during the stream. To gather as much data as possible on what is the context, to offer a better aligned next song (if you are a paying customer) of better aligned ads (if you are a non-paying customer).

This is why all streams must not be treated equal, and the streaming services must find a way to be as fair as possible towards musicians (having in mind that no-one likes a random playlist).

As this is a hard problem, I suspect that more and more services will divert their research towards automatically generated music, so nobody will care if the algorigths are fair and if the authors are paid. Music is most susceptible to this, as it is easy to generate music that sounds good enough to a fairly large audience without realizing it is all auto-generated.

Advertisers pay for a radio and tv service, not users, right?

Don't compare Spotify against god, compare it to the alternative. Everyone has a "recommendation algorithm", it just doesn't have to be spotify's. It could be the radio, or what they see talked about in pop culture (e.g. WAP), or what their friends are listening to, etc. Spotify's algorithm has the potential to be radically more fair than any of the available alternatives. A fair algorithm is one that always shows people the songs they're most likely to appreciate, and the incentives are correctly aligned for Spotify to make a good effort to get as close as possible to that ideal (Spotify's top-tier "discover weekly" playlists are why I stick with it over any other platform). Music I listen to through the recommender is actually more valuable to me than music I picked out myself - if I only wanted to listen to music I already know about, I would just pirate it or buy it on itunes and save myself $10/month.

As an aside, I also disagree with "In scenario 3 and 4, the user just wishes to have something making noise, not anything specific, but not random". I don't just want something making noise, I want something making the kind of noise that I like to listen to, which includes a mix of old songs/artists and new ones I've never heard before.

> Here's a compelling irrefutable argument.

That argument isn't irrefutable at all - let's break it down.

> Scenario 1 & 2 it can't be refuted that artists should be paid for the stream.

Agreed.

> [Scenario 3 & 4] So here, not all streams can be considered equal, since some had higher probability to be present in the playlist due to inherent algorithm biases.

Disagree on the implication here - I don't see why it irrefutably follows that an artist should be paid less for these listens just because its in an automatic playlist (assigned because it matches the same profile as a record I previously liked). What you have here is an opinion, not an irrefutable argument.

My personal opinion is that if you have a streaming service, a minute paid for a minute listened is about as fair as a metric you can get.

Why are my listens, which are mainly from automatic playlists based on my historical listening, less valuable than someone else's listens, just because they listen to cool obscure bands not on the playlists?

Right, but artists don't care whether you're listening to their music actively or passively. Whether you stream 10-20 hours or 700 hours, 1 stream is 1 stream. I don't see why this should affect the payout

Doesn't Spotify say "don't play at work"? Seems it shouldn't be on in the break room

I agree and I think this is really the key demand that they should have organized around. Splitting up the revenues based on what you actually listen to is really how things should work from a fairness perspective to artists. It also is likely more compatible with heavier listeners than their demand of 1c per stream. To put that number into context, I listen to ~1000 songs per month on Spotify and I pay $2.50 for Spotify each month using a family plan. I'm not sure how the economics of 1c per stream are supposed to work out there.

Soundcloud has already gone in the direction of paying out based on what you actually listen to, and it would be great to see Spotify and other platforms follow suit. See http://fanpoweredroyalties.com/ for more details on Soundcloud's implementation.

Aren't your listens to an indie artist still driving up their listens across the platform? It's an interesting point that Spotify's algorithms probably drive more listens to popular artists, but I think overall listen-count it still the most equitable metric.

The point is that it's about valuing the content which actually makes or breaks people's decision to subscribe to the service. Spotify is convenient, but it wouldn't survive on top-40 content because people would just go back to their MP3s or listen to it on YouTube.

I'm pretty sure that Netflix understands this, by the way— they know that deep-pocketed competitors like Amazon are going to eat their lunch if they try to do mass-appeal prestige shows starring A-list celebs, so they only do a few of those (enough to basically fill the homepage), while leaning hard into extremely niche content. Like, have you seen how many weird cooking shows there are on there? Or Hallmark Christmas movies? This is stuff that's harder to find elsewhere, and there are dedicated audiences on Netflix who are mostly or exclusively consuming it. If Netflix's internal accounting worked like how Spotify pays royalties, those shows would be instantly canceled in favour of new primetime-style sitcoms.

But they aren't, and I think it's because there are pools of their subscribers who will cancel the service if there isn't a steady stream of the specific content they're interested in. Thus, their subscriber $$$ go directly to funding that content, rather than into a general pot that's broken up by raw viewership numbers.

Netflix understands this as they have become content creators.

Should is a big word, but it feels like Spotify could make us indie-lovers happy by rewarding the intentional listens with slightly more money. Perhaps giving them a cut of the next few recommendations. Even though I know that famous artists would get more of this than the indies, taking value from the long mid-catalog.

I think that's risky because then there's some process that puts you in one of the two buckets, and there's probably going to be a valley of suck between them, where you're big enough that you don't qualify for the indie special treatment, but not so big that you're raking it in like Taylor Swift.

The nice thing about the proposal from the article is that in theory it should be something where everyone's incentives are aligned, since it rewards artists and genres for having fans who listen to their stuff a lot.

Yeah, that's what I mean by the mid-catalog. People you didn't quite go intending to play, and not the headliners. To the degree that it's a fairness discussion though it always comes down to sorting into buckets (even if absolutely identical buckets) and arguing for some distribution.

Good point. That's the sort of thing that warms the cockles of my artist-supporting heart.

I agree with this argument, and this is THE reason why I don't pay for a streaming subscription. I have no desire to see my money to go to the likes of Justin Bieber (who I don't listen to), whereas I would much prefer it goes to the smaller ambient and indie artists that I like to listen to while working.

Spotify should charge by the minute so you turn it off when you're not actually listening.

They do kick you off after a while if you're a free account, but the marginal cost of the bandwidth for them is basically nothing. Anyway, it's 2021— who would accept a pay-per-time-listened Spotify in an age where Netflix, Hulu, Prime, and Disney+ are all all-you-can-eat for a similar price?

I feel like you should take Spotify's audited financial statements about how much they make and how much they return to rights holders (~75% of revenue) more seriously than your back of the envelope estimates: https://investors.spotify.com/financials/default.aspx

Agreed. The above calculation for example does not account for the fact that Spotify pricing varies significantly between countries and can be as little as 1.60$ per month.

https://edm.com/industry/spotify-premium-prices-by-country

Indeed! I have a family account for 6 people here in Brazil and I pay around 0.77 USD per person.

The problem is that most artists do not own the distribution rights to their own music, these are owned by various record labels. Many top tier artists create their own record labels and shoulder the cost of advertising/promotion while also dramatically increasing their cuts.

The fact that individuals aren't able to break out via self-publishing music is an oddity in the internet age. Why aren't there more avenues to hear "random new artist that matches my tastes"? or is it self-perpetuating that labels sign all successful artists who can go mainstream?

>Why aren't there more avenues to hear "random new artist that matches my tastes"?

Because Bandcamp is perfect at this and nobody can beat them.

I say that with tongue in cheek, but they have been the platform for average musicians to self-publish on for a while. They've even totally foregone their cut periodically during the pandemic to help artists. And unlike Spotify, artists are able to apply their own branding to their pages. I've never met a musician with a bad thing to say about them or their platform.

> Why aren't there more avenues to hear "random new artist that matches my tastes"?

One reason is due to flat rate bundled compulsory licensing. If you run a coffee shop with live music you either restrict all performers to only play original works and still probably get threatened with lawsuits, or you pay up the ASCAP/BMI/SESAC protection money and can play whatever.

So in many contexts weirdo obscure independent artists can't compete on price (e.g. give their music away just to get it heard) because the customer has already paid for a bundled rate for popular music.

I like to listen to jango.com which tries to "push" new artists while you listen to some "station"... I've found a few good "completely unknowns" there, but most stuff they push is really low quality. I think that a platform where you could discover new artists easily using a system similar to this (play mainstream artists, mostly, but throw in a few songs from unknown artists that play similar genres), but then had an option to "upgrade" to get stuff like offline access to artists you like, could become very successful.

This is Pandora, right?

https://en.wikipedia.org/wiki/Pandora_(streaming_service)

> play mainstream artists, mostly, but throw in a few songs from unknown artists that play similar genres

Isn't that just "radio"?

Yes, ignoring all other costs isn't great.

Their revenue per subscriber is also substantively lower - including free users, discounted EDU users, and family memberships. Excluding free users, it's probably ~$5 per user.

That being said, Apple Music pays .75p/stream, Tidal is 1.25p/stream. They can probably figure it out.

If they cannot pay the artists more, a long-term freemium plan might not be a great business model then. If free users are seriously cutting into profits, should they change the free plan to a limited free trial? I was around in the 80s 90s and we didn't really have free cassettes and CDs and radio paid studios & artists with ad revenue. Spotify is well known enough now that it could be argued an unending free tier may no longer be necessary.

Perhaps the dynamics have changed since you had to shop for music in physical stores, and purchase it in hour-long increments.

I certainly play more music now than when I had to change the tape, but without the ritual and the intent, I listen a lot less. This changes my value from streaming over simply not having to store my music library locally.

It doesn't cost an artist more to be played more, but it does cost the streaming service. This may bring more casual listening and help some artists, and "underpay" someone who's more famous and would get the bulk of the first-song play that comes from intent. I don't think it's "wrong" but it may be the wrong fit for any given artist.

Totally agree. Maybe spotify should look like an app store with fees in the 30% range at the top end (shopify/apple/steam). Their role is very similar: distribution and discovery. They probably could wi points giving better payouts per stream to smaller artists too, like steam does. Hopefully competition will force it down too (eg epic vs steam vs apple). We might need open access to back catalogues and not exclusive deals, but its a start

The App Store for music exists many times over and users (including myself) prefer a lower cost streaming model. There is simply too much cheap or free music out there now for me to value it at iTunes singles prices.

Of course, and users would prefer completely free music over cheap music too. That doesn't mean it's sustainable, though.

I mean, short of government legislation the market isn’t going to stop driving the price of music down. The marginal cost of additional “performances” of the song is almost zero and people are constantly making new recordings for free because they love doing so. I don’t see why making a living off digitally recorded music has to be sustainable, and I don’t see how it can be when the supply is so endless.

It very well may not be sustainable. But I guess my point is that Spotify's business model might turn out to not be sustainable if artists decide it's not worth it to have their music on it.

If they don't decide that, then of course, maybe you just can't make a living selling music through Spotify.

You could probably make a living as a musician in other ways, though.

This doesn’t make sense to me, it’s not that they are eating into profits, it’s that they are eating into average margin; they presumably still at least break even on free users. And I would bet dollars to donuts that those free users are not going to convert to paid when piracy, YouTube videos, and free trials of other services exist.

I think Spotify is the first service to tempt me from piracy. Piracy doesn't handle discovery, or access across multiple devices, is a pain, and is actually (who knew?) illegal. Spotify handles all of those and won't get you arrested.

Youtube for music is simply worse than pirating and being caught.

To be fair, back in the day I used last.fm for music discovery, social music stuff, etc. and the more Spotify forces podcasts down my (paying subscription) throat the more I want to go back to piracy + last.fm.

> Yes, ignoring all other costs isn't great.

> Their revenue per subscriber is also substantively lower - including free users, discounted EDU users, and family memberships. Excluding free users, it's probably ~$5 per user.

> That being said, Apple Music pays .75p/stream, Tidal is 1.25p/stream. They can probably figure it out.

Their rates are per country, they differ. So these calculations gave to it into account, but everyone pays the 5 or whatever it costs now in the west.

The amount they pay per stream differs per country too, I think.

The nominal figure of what the platform per stream pays artists should not be discussed without taking into consideration the amount of monthly active users using the platform. Tidal and AM have significantly fewer users than Spotify

The $0.0038 figure is probably due in part to the way Spotify calculates royalties, dividing a pool of all subscriber revenue among all streaming activity, rather than dividing the revenue from each subscriber among each subscriber's streaming activity.

Some have pointed to logistical challenges in the latter model, but it's clearly not impossible, as SoundCloud just announced it will be the first service to adopt it: https://press.soundcloud.com/197001-soundcloud-introduces-fa...

Think about a strategic decision like this: Spotify decides to pay-out a $0.005/stream rate to Taylor Swift. That's more than most artists make, but its what Taylor demands, and Taylor holds some cards here because she's popular.

Let's say that Spotify says No. Spotify's argument is, in effect, all artists are sharing a piece of the same pie; if they give Taylor Swift a higher than average rate, it lowers the average payout for every other artist. By saying No, in this hypothetical universe, Spotify is standing up for indie artists.

Taylor (more accurately, her rights holders) does not put her music on Spotify. Customers who come to Spotify for Taylor Swift's music no longer feel that their subscription is worthwhile; they stop paying.

Now, the pie is smaller. Indie artists receive a bigger chunk of a smaller pie. Customers are left without the music they want. Spotify's revenue is lower. Everyone loses.

Its easy to place the blame at Spotify, and I'm sure they deserve some, but we've known for about half a century that the record industry is a hive of shit, and this is just another symptom of this. The disparity in income between the popular artists and average artists is huge, and its mostly because of gatekeeping that the popular artists actively engage in to protect their nine-figure fortunes. Spotify could have stood up for smaller artists, but I'm not convinced they'd even exist today if they had; it was really their willingness to work with the record industry that turned them into the force they are.

I do have some hope that Square's purchase of Tidal could be a catalyst for change in the industry. In other words, if they maintain Tidal's high(est) revenue share with artists, improve the user experience (a lot), leverage their existing industry relationships to keep enough popular artists on the platform, and work with indie artists... it could be very good. I'm optimistic. I think Tidal's core investors being artists themselves and a traditionally low-margin payment processor is actually very healthy, versus Spotify who effectively is just a middle-man that needs to extract enough to survive.

If Spotify is in fact spending 75% of its revenue on royalties, but according to you less than 25% of it arrives at the artist, what does that tell you? Either your calculation is wrong, or somebody who is not Spotify is pocketing the difference.

You think that the developers and every other party involved costs less than 50% here? Devs cost a lot. Bandwidth and servers don't cost that much, but for large scale data transfer like spotify, it does. Running the company costs a lot. Maybe the fair price is double what it currently is. But that might result in the artists making less than they currently do, not more (this is, in fact, my suspicion - if Spotify could increase profits by charging more, they would already be doing so).

How much of Spotify's revenue goes towards songwriters, labels, distributors, publishers, as opposed to the artist?

I tried to look into what radio paid out to these same people as a comparison, but the information I found seemed to be purposefully obtuse:

https://soundcharts.com/blog/radio-royalties

I couldn't actually find out what was paid, but it did look to be that, artists weren't really paid. It was song writers and publishers who were paid. Performing artists, however, only generated word of mouth which generated ticket sales - so far as I could garner.

The quoted $0.0038 is the total of what Spotify pays out to the distributor, who then distributes it between themselves, the artist, songwriters, their labels, etc.

Are only paid plans paying out royalties? I used to pay for spotify, but cut it after a budget rebalance since we just didn't use it enough. I still use it about 50% as much as when I paid.