2. Artist’s Spotify streams divided by total Spotify streams
This calculates an artist’s popularity on the service, their
“market share.” Dividing an artist’s streams by the total
streams on Spotify determines the percentage of our total
pay-outs that should be paid for that artist’s rights.
This sucks bigtime. Why should artists that I never listen to receive any of my monthly subscription? This portion of the calculation should be done on a per subscription basis.
Some analysis on the comparison of the two systems.
This system chooses all streams (one play by one user of one track) equally, rather than by treating all subscriptions (one user) equally.
When you treat all streams equally, you effectively make a system where users who listen to the most streams have their weight felt the most.
For instance, consider this world with two users.
[User 1 listens to artist A once a month]
[User 2 listens to artist B five times a month]
Under the Spotify stream-based system, artist B is awarded five times as many royalties as artist A. This might be seen as "fair" in that royalties are paid out depending on the proportion of value/happiness/"utility" created, assuming each stream produces the same unit of value.
Under an alternative user-based system, artist A and artist B are awarded equally. This might be seen as "fair" in that royalties map as directly as possible to the user fees. It sees users 1 and 2 as equally satisfied through their bare use of the service, regardless of the number of streams they've consumed.
People generally want to see systems where value creation is proportionately awarded. Under that goal, which model is "better" depends on how you map value to streams and users. That's the core problem... also consider some complicating factors like users gaming the system by spam-streaming artists they want to support or the value associated with stream consumption by free users (a large part of Spotify's user base).
Say I pay $10/month and only listen to 2 artists, each with .001% market share. IMO, each one should get $3.50 from my pool of subscription money. Using their current scheme, each would get $0.00007 while Lady GaGa would get $0.30 when I didn't even listen to Poker Face.
I see your point, but I think your reasoning is flawed. If each one got 3.50$ from you then they would get nothing from the other subscribers, whereas, as it currently is they receive .00007 from everyone... Even if Spotify followed your subscription model the end result would be the same small amount of money.
Your method would only increase complexity for Spotify and serve no real purpose in terms of getting your favorite artists more money.
If tweens listen to 10 hrs/day of Justin Bieber on repeat using free accounts, Bieber shouldn't get money from adults who listen a few hours a day to anything but him using premium accounts.
I think it's quite reasonable to assume that differences in listening and paying habits between users would NOT average out, and a per user split would be fairer.
I agree that a per person split would be might be better (for how I think royalties should be distributed) -- but one would also take into account the ad revenue those tweens (and other "free" users) generate.
I don't know how Spotify is doing financially, but hopefully they make money off free users as well as paid users (but I'd not be surprised if they end up a bit like Opera did -- making money from paid users and licensing/bundling deals -- and just using the commercial breaks/ads as stick to guide users towards the paid service).
Either way I'd much prefer being able to pay for lossless records that I get to keep -- I gave up on Spotify quite early as it ended up a little like youtube -- come back to a playlist after a few months and half the songs were gone. I know they're better now, but that experience just underlined the idea that paying for licensing content in a way that leaves you vulnerable to that content disappearing is a very bad deal for me as a listener/consumer.
Its interesting that you mention their ad breaks. For 6 months, I had no idea that spotify operated a premium service whatsoever, until they imposed the 2.5hr limit per week on my account and hit me with more ads. I subscribed immediately.
I think by broadening the pool they are effectively bringing more music to Spotify. This is good even if it's of the Bieber variety because then users like me can discover artists I wouldn't normally listen to or discover new CDs from artists that used to be in fashion and are no longer.
Imagine you have 10 people who listen to artist A once every month and one person who listens to artist B 90 times.
Under the current spotify system, artist B would get 90% of the revenue and artist A would get 10%. Under the parent's proposal, artist A would get 89% of the revenue and artist B would get 11%.
In any case, I'm willing to bet that the parent's proposal would actually backfire. I would wager that people who listen to obscure music also listen to more music so they're actually receiving more under the original spotify proposal.
Yes, because buying merch and donations are one-time deals. No one is buying the same shirt from their favorite band every month but they are sure listening to their music every month. And if the the lionshare of my own subscription is going to Lady Gaga just because she released a new album...well that makes no sense in any business.
Yes. But the millions of people who don't listen to those two artists are also paying them $0.00007. If you assume everyone listens to the same amount of music, then the two system are equivalent. If you listen to more music than the average person, then your 2 artists get paid more using the current system. If you listen to less music than the average person, then your 2 artists get paid less using the current system.
As long as you use spotify to at least the amount of an average spotify user, your two artists will get the equivalent amount of money, since all of the other spotify listeners will be paying that sliver for your artists.
From an artist's perspective, the decision to sign up for Spotify basically comes down to one question:
Does the benefit in music discovery make up for the cost of losing album/download sales?
I did the calculations once and I forget the details, but the gist is that without factoring in music discovery benefit, an indie artist probably only hopes someone chooses to stream instead of download if the listener is going to stream the song around 1,000 times or more. Otherwise, you get more financial upside if they buy the download.
So the question is if any discovery benefit will shrink that number enough. This would include the people who stream and don't buy but would never have bought in the first place, as well as the people who stream and then buy.
The answer is difficult to prove because you get into weird counterfactual questions, and there really aren't any good studies out there from unbiased experimenters. But it's pretty reasonable for a rational musician to decide that any discovery benefit will not make up for the cost in losing album or download sales.
I think that the real value to artists with spotify is that they show upcoming touring dates for the artist inline with the artist profile[0]. I think that spotify has a lot more that they can do to facilitate direct artist interaction with the fans though.
I love that feature but just listing concerts in my country isn't very useful in the US. I regularly get shown that someone is playing in Portland or Austin and I live nowhere near either of those places. Given it hooks into Songkick (awesome site btw) which knows locations of all these events, maybe they could find a way to track where you are and give more relevant suggestions.
I'd like to see a comparison of royalty models between streaming services (Spotify, Pandora, Rdio, iTunes Match/Radio) as well as how it compares to buying a physical CD.
There are a ton of comparison charts out there... one thing to keep in mind is that a service like Pandora is completely different than a service like Spotify. Different royalty rates, different implications. It comes down to the fact that on Pandora, you can't request particular songs.
For that reason, artists don't tend to mind Pandora too much, since any royalty income they get from Pandora is basically gravy. It doesn't really rob album sales. You could argue that it robs radio play (since radio is less popular now), leading to lower royalties that way, but it's a different beast than Spotify. For Spotify, a person can request a certain artist/song and listen to the stream, without buying the track/album. It robs album/download sales, and it's debatable (to put it charitably) whether the "music discovery" benefit makes up for that loss.
One of my favorite things about the Spotify service is music discovery. In my experience I've found that they do a pretty decent job of helping to show me new bands based on my current preferences and listening habits, as well as those of my friends. If that continues to improve, I think they will also be able to do a much better job of helping smaller bands get additional exposure and therefore royalties.
Looks like the system can be gamed by an artist than listens to his 4 min track 24/7. If it pays about $0.007 per stream, listening to a 4 min song would pay out $75 - $10 of the subscription fee = $65/month/subscription. That wouldn't happen if the royalties redistribution was done on a per subscription basis.
Minus the cost of a dedicated server. Minus the risk of getting caught. Minus the probability spotify has a trivial cap in their software to prevent it. Gets you pretty close to $0.
Youtube was gamed for years. It might still be possible on spotify. Not just for the money, but also to get higher in their rankings and get more exposure.
I don't have much personal investment in the story, I'm a significant consumer of music but haven't user Spotify (despite being in Canada I was only vaguely aware the service isn't available here).
They've taken some flak in the past for their payments to artists and while I'm not following the discussions closely enough to suggest those criticisms are unjustified it does seem that they're not the chief culprit. 30% retained doesn't on the face of it seem unreasonably high, does anyone know how this compares to equivalent services?
Ultimately I don't think disclosure is ever a bad thing and hope maybe this is one more small step advancing the general 'royalty model' conversation.
70% royalty share was imposed by the majors to Spotify. Just because this was the model used by physical stores and it's how the industry works. They didn't change the revenue split model.
One of the problems here is that there are many middlemen involved between the artist and Spotify. The label, the manager, the aggregator. Everyone gets his piece.
Also, there is problem regarding how much it takes for an artist to get the same from Spotify than they used to get from a song sold from a service like iTunes. Some artists claim that a song has to be listened up to 60 times in a streaming service in order to the same as a song sold in a digital music store. And here there comes the big issue. Have you ever listened 60 times to a song you buy in iTunes.
Spotify has been trying to claim that 70% is too high, and were trying to push for a reduction in royalties paid to artists via labels and lobbying congress. Trying to point out that they were not profitable, even though they're a new business and expecting profitability in this area so early isn't realistic by most measures.
But paying 70% of revenue for inputs isn't that big of a deal, especially for a business that can scale without having to make significant long-term capital investments.
Here's the problem: 70% of what? Revenue? What if the revenue is lousy?
Also, indie musicians have lousy royalty terms compared to the major labels that have partnered up with Spotify.
There's also the problem that services like this and Pandora are apparently unsustainable.
It's really part of a long pattern starting with Napster. Start a kick-ass service that gives music away for "free", ignoring pesky problems such as illegality or unsustainable royalty costs. Get the consumers excited about it by lying to them, and lose money for a while as you grow. Then pit the users against the musicians by having everyone complain to Congress about the royalty rates and try to convince the musicians that they shouldn't expect to make money if they aren't off sleeping on Motel 6 floors and making $12/night trying to sell t-shirts in dive bars. (Oops, did some bitterness sneak in there?)
> 2. Artist’s Spotify streams divided by total Spotify streams
this is quite surprising. does this mean that hypothetically, if there is only 1 band in the spotify universe, it will receive half the revenues it once did if another band joins the service?
How I read it is only if that other band take half of Spotify's streams. If no one likes them and they get no plays, the original band still receives all of the revenue.
"if we could build a service which was better than piracy, then we could convince people to stop illegal file-sharing, and start consuming music legally again"
this is exactly the right attitude. i've been 'selling' this for years as my reason for piracy - its just easier. spotify absolutely killed that for me. it really is more convenient...
i do wish existing businesses in that area would look and learn instead of lobbying and abusing legislation to try and keep hold of a business model that died years ago... (or publishers trying to convince their artists that Spotify is ripping them off rather than /them/)
In both cases, there's no further incentive for a consumer to pay out to get better access to the music. AFAICT at current royalty rates, artists get about the same amount of money they get from piracy.
Except with piracy, people know the way they obtained the music is questionable/unsustainable, so in some people it probably creates a drive towards future patronage of some kind.
Meanwhile, Spotify gives the illusion that you're involved in a legitimate transaction and have more or less done your duty, while what's basically happening behind the scenes is we have a "disruptor" whose main innovation so far appears to be replacing recording revenues with the fractional broadcast revenue.
There is a catch: If the service is priced low to attract pirates and then non-pirates also switch to it, total industry revenue actually goes down. I don't think Spotify is there yet since few people spend over $120/year on music, but it is a long-term concern.
If you read all the way to the bottom you can find the main reason why I support Spotify, "Spotify’s impact on piracy"
>Spotify has been successful in convincing this younger generation to abandon piracy and begin using and paying for a legal service. In fact, over 50% of Spotify’s paying subscribers are under the age of 29.
This is the reason that I love Spotify and its ilk.
1. Setup a short track on one of the linked providers.
2. Setup a bot to play the track over and over.
Do they pay out equally for tracks from paid vs. unpaid spotify accounts? If so, you could setup a bunch of anonymous accounts to increase stream plays.
I have a gut feeling Spotify would pick up on the unusual listening activity. If you have a botnet lying around collecting dust, you could always try it and see what happens. My guess is that they have algorithms to detect people trying to game the system.
This section is highly misleading. US terrestrial radio mainly pays song writers not "artists", and it pays them decently well. This section should be regarded as a solid lie.
At KiteBit we recently pivoted to the music market. Selling music, concert tickets & merch.
Music artists obviously have the pain of not getting enough from streaming service. Also physical and regular digital downloads are falling... So there is a cash flow problem in the bands not being in relations with the very opaque majors.
So now KiteBit offers a service for artists/bands who already have a loyal fan base who is willing to pay for the contents they produce. Not just for the sake of the contents put on sale but appealing to the emotional relationship they have with the artists, and using pricing models like the "pay what you want" that improves revenues by 12% goo.gl/N0oOeD
All in all, KiteBit it's like a bandcamp for Pros that can be used to sell music, concert tickets, or link their merch to digital downloads to retain customer loyalty. The trick is that the service wants to be transparent to the audience so that there's no feeling of any middlemen in the buying experience. that's why you won't see any marketplace in the KiteBit's site.
I love Spotify. It actually revived my joy in doing anything really. I've kind of grown bored of life, of coding, of anything. Music is the one thing that I am and will always remain passionate about.
I'm Canadian, so getting Spotify running on all my mobile devices and my mac was tricky, especially with Premium, but I got it done.
The ability to support artists, while discovering new ones, and not acquiring the material illegally is crazy awesome. I am discovering new artists day in and day out and it keeps life very exciting for me personally. I even got back into coding.
It's great to listen legally, but keep in mind that the support artists receive is very small (all but the most popular musicians receive less than 40 hours of minimum wage from Spotify). If you like a few groups consider buying merchandise or making a donation.
Wait, are you saying that artists rely just on Spotify for income? That would be foolhardy. I would think that Spotify would be just one small chunk in a wide portfolio of distribution channels.
Also, artists have generally never been very richly paid on music sales. Touring and merch are where the money's always been.
> Wait, are you saying that artists rely just on Spotify for income? That would be foolhardy.
Well, yes. Since the entire premise of Spotify seems to be that we should pay broadcast royalty rates for a service that replaces recordings entirely, it would indeed be pretty foolish to rely on that.
And in fact, some artists think it's so foolish they've realized its to their advantage not to participate.
> I would think that Spotify would be just one small chunk in a wide portfolio of distribution channels.
The problem being that once something's available on Spotify (or something like it), the structure of the service eliminates any access-incentive to look for it anywhere else.
> Also, artists have generally never been very richly paid on music sales. Touring and merch are where the money's always been.
This is pretty different from the history I'm familiar with -- touring supported/drove album sales for a long time, except for certain acts (huge like the Stones, jam-oriented like the Dead or Phish, or nostalgia focused). And people made a ton of money from recordings (though it wasn't always the artists because of gatekeepers).
The reverse is often true now mostly because we entered a transition to a new format with the uptake getting waaay out ahead of the vendors and legit recording sales collapsed (and then, right as we started to see vendors come on board, we had the biggest economic contraction since the great depression).
This isn't an inevitable trend, and revenue from recording sales this year is increasing again this year based largely digital sales increasing faster than physical sales are falling. And total revenue from recordings sales, while much smaller than what it was at its peak around 2000, is actually 50% more than it was at the beginning of the 1990s.
We could yet see another golden age of recorded music. This time better in that digital albums are cheaper and, with a shorter distribution chain, people who make the music are often keeping more.
At least, that's the potential. Unless we decide to go with buffet-streaming services that remove incentives to buy it.
But maybe that's the next step for our culture: if art can't be supported by t-shirts and tchotchkes, it's not art that we should be economically encouraging, right?
This is one of those zombie beliefs that never die. I can't blame any particular person for holding it, but it's really not accurate. Touring and merch is just as much of a grind as music sales.
How long will it be until artists start using product placement and full blown advertising within the actual content of their songs to get around these things..?
If you listen to any hip-hop track made in the last 5 years you will find that is already the case with a good percentage of said tracks. I met a guy recently who's only job was to find brands to partner with this young but fairly well known hip-hop artist. He charges around $500,000 per mention. It should be possible to calculate rough revenues by going through lyrics on Rap Genius.
They paid around $500 million in 2013. With 24 million users, that's $20.883 per user per year.
To check this: According to Spotify, their 24 million users listened to 4.5 billion hours of music in 2013. This gives us 166.667 hours per user per year. If you take an average song length of 3.5 minutes, this results in 2857 plays per year, which, if you assume Spotify's number of $0.006 to $0.0084 (let's take 0.0072 USD) per stream, this results in 20.57 USD per user per year. Wow, very close, it indeed looks like a user generates around $20 per year.
If the entire US population would use Spotify, this results in roughly $6.5 billion per year.
Of course, that's a bit rubbish, let's use the amount of households in the US. Around 120 million. So that's more like $2.5 billion.
If we assume an average household income of $45'000, this would support around 55'000 individual household incomes in the music industry (artists, studios, engineers, right holders and so on).
Of course, since most money flows to the mega stars. It would support much fewer people. So let's take 31'400 people. Why? Because it goes well with the US population figure of 314 million. It's exactly 0.0001% of the US population.
I'm going to assume the revenue numbers you mentioned are US numbers.
The entire population isn't trying to be musicians - you're only competing with other musicians for that money. This website[1] offers various estimates for the number of active musicians, but the numbers are quite scattered due to the subjective and numerous definitions of "musician" (orchestra player doesn't need royalties, rock band member does, music teacher doesn't, etc.).
I'm going to use the number of albums produced to determine the number of active musical acts who would get paid by Spotify, which according to the source was 76,875 in 2011, and one album per active musical act seems reasonable. Assuming that is a US number for conservatism (not a global number) then we get 76,875 active musical acts competing for your $2.5B. Since most musicians won't assume that they will become superstars, using the 80/20 rule we'll chop off the top 20% of active musical acts and the top 80% of the revenue (which imo is probably most artists that anyone knows) to get 61,500 musical acts competing for $500M. Now lets assume 3 musicians collecting royalties per musical act (seems reasonable, lots of 4-5 member bands, lots of solo artists, conservative) for a total of 184,500 musicians competing for that $500M. $500M supports 11,111 people under your $45,000/year income assumption, meaning 6% of the bottom 80% of active musicians would get a livable income from Spotify alone. As other posters have pointed out, musicians have many other sources of income such as other streaming services, touring, record sales, merch, etc.
So yes, since you have a 6% chance of earning a livable income from a singular source of music streaming while being in the bottom 80% of active musicians royalty-wise, I'd say it's definitely worth it to keep doing what you love. This also ignores the many non-royalty-based careers for musicians, such as teachers and session artists.
Music streaming (be it Spotify or otherwise) is not the only source of revenue for musicians though. Touring, Merch and so forth also generate revenue.
That said, most musicians obviously cannot live off their music.
If Spotify is worth $4B, that means the labels have made an "extra" ~$600M in addition to the 70% royalty. Over its lifetime so far Spotify has paid out around $1B in royalties, so an extra $600M is pretty significant.
