How To Fix Spotify

A year ago next month, we released an article exploring how much Spotify pays independent artists. At the time we published, there was a lot of misinformation circling around, and ours was one of the few reliable stories about the topic on the web. It quickly became one of our most popular, and I still get emails about it nearly every week — mostly from artists and journalists or bloggers working on their own stories.

I wrote it more out of necessity than anything else. We wanted to know the answers, and they seemed to be hard to find. Since then, there’s been a lot more great writing on this subject in other outlets. But once again, the most interesting, relevant and reliable data doesn’t always show up at the top of a search. Here’s an overview of what we’ve learned over the last year, and few ideas on how to make the new system better.

Still Growing

Last week, Spotify announced that since we wrote our article on this subject in September 2011, paying customers have doubled from 2 million to 4 million, and non-paying users have increased 10 million. This has made founder and CEO Daniel Ek millions of dollars, and the company’s is newly estimated to be worth $4 billion.

Although plenty of people at the company are making a ton of money,  Spotify itself has yet to turn a profit. For now, the company’s focus remains on growth and last year, with revenues of about $250 million, they were able to show losses of nearly $60 million.

Spotify expects record-breaking revenue again year, and gross estimates for 2012 are already nearly $900 million. Out of this huge revenue stream, CEO Daniel Ek says that the service has paid “over $200 million to rights-holders” since it began, or a bit less than 1/4 of its 2012 earnings to artists and labels combined.

Is it Fair?

As a user, I’m a Spotify fan. As I write this article, I’m alternately using Spotify to check out albums that I’ve always meant to hear, and using Pandora to stream music from stations based around my favorite artists. It works great.

As a music professional however, it’s hard not to feel that the service has a long way to go. It’s a clear step in the right direction, away from the web-piracy model where self-deluded thieves make millions of dollars by selling advertising on artists’ work. But in order for the Spotify service to make the move from leech to lifesaver, it will have to continue to improve on a few key fronts.

Until then, the service is essentially broken as far as artists are concerned. More about how to fix in a moment.

How Much Does It Pay Now?

When we last approached this topic, anecdotal reports of payouts tended to range from $0.004 to $0.006 per stream, or roughly a half-cent for each play.

Some of the data we presented back then came from a pro-Spotify website called SpotiDJ, which at the time, had very little history. Since then, we’ve watched them develop, and it appears that the site presents some pretty reliable data that mirrors other reports. That data however, is often accompanied by somewhat one-sided editorials that rarely take the musician’s perspective into account.

For instance, when cellist Zoe Keating shared her Spotify royalty rates earlier this year, she offered a full account of all of her revenue over a 6 month period stretching from October 2011 to March 2012. In that span, she earned $46,000 through iTunes, $25,000 Bandcamp and $8,000 from Amazon. At the same time, her earnings via Spotify sat at $290. And not for any lack of interest, either: In total she racked up about 80,000 streams on the service.

The bloggers at SpotiDJ and other pro-Spotify websites were nearly unanimous in their logic-defying advice: Zoe Keating should simply promote Spotify more if she wants to earn more money through the service.

It’s not surprising that to many musicians this instantly registers mind-bogglingly backwards reasoning, and quite simply, terrible advice. Since Keating stands to earn the greatest revenue per listener via Bandcamp and then iTunes, she should do exactly what she is doing already: Promoting her work on those two services, and in that order.

But putting aside a handful of baffling talking points from the big-tech echo-chamber, Zoe Keating’s example is a great thought experiment for other independent artists. If streaming services do eventually displace conventional sales as a primary source of revenue (which is still years away) what kind of streaming rates would be sustainable?

Currently, assuming an average of a half-cent per stream, Spotify pays about $5 for every 1,000 plays. While no one expects streaming revenues to parallel sales figures (we hear the “apples-to-oranges” analogy regularly from the Spotify camp) even with that in mind, this figure clearly does not scale.

If Keating were to attract 3,000 true fans who would listen to an entire 10-song album 12 times each in a year, those 30,000 monthy plays would gross her a mere $160 each month.

Of course, an artist like Keating may be able to do a bit better than this (if she were to foolishly promote Spotify instead of album sales.) Assuming her take on an  iTunes album sale is the normal $6.30 for self-released artists, this means that roughly 6,000 people bought her album on iTunes in the period we’re looking at. But let’s put aside her actual, currently sustainable fan base for a moment and get super-optimistic. Let’s imagine that Keating was somehow able to attract five times as many listeners over the next year. With 30,000 listeners each streaming her entire album 12 times in a year, those 300,000 monthly plays would earn her $1600 a month.

This would be an incredible feat for any independent artist. And unfortunately, it is one that would put her in the poor house with a gross salary of about $19,000 a year. Although it’s certainly possible with the help of food stamps, I wouldn’t like to see her have to try and raise her toddler on that kind of income.

To add insult to injury, this kind of income would leave nothing left over for Keating to to even consider financing and releasing another record. She just wouldn’t have the funds to put out a album that would potentially employ a dozen or more people in recording, mixing, mastering, promotion, accounting, live sound support, art design, catering, booking, babysitting and more. Recording a follow-up  to an album that at least 30,000 dedicated fans already showed they would be excited to hear would change from reality to an impossible dream.

Okay Fine, It Is Broken. Now How Do We Fix It?

Fortunately, it won’t take much to fix the streaming system so that it scales in a sustainable way. It may be difficult at first, and could require a fight. However, it can be done, and the proof is in the numbers.

Even if the average rate per play can reach one full cent, or better yet, two cents, on-demand streaming begins to work for both the artist and the distributor .

Where a half cent per stream means $5 for 1,000 plays, one cent means $10, and two cents would generate $20 for 1000 listens. This may not seem like a lot at first glance, but it’s on par with the advertising rates for a targeted niche website like this one. At this rate, an artist who can attract 10,000 no-risk listens in  a month — even partial listens — would generate $200. Not a staggering amount, but a number that scales.

Take these figures and apply them to Ethan Winer, who wrote an article for the June issue of Scientist. He produced a single YouTube video of a song comprised of 37 separate cello parts and went on to generate more than a million listens on YouTube. At a rate like this, his one self-produced song would have generated $10,000 or $20,000 in revenue. This is a worthy sum that almost demands more of the kind of hard work that went into his production.

If we return to our imaginary best-case scenario for Keating, 4 times a paltry $19,000 dollars a year would amount $76,000 — the result of that same 300,000 plays per month at $0.02 per play.

Of course, it’s worth noting that this is still lower than Keating’s current annual earnings through iTunes, and that is would require a far larger potential fanbase than she currently has. Ideally, additional revenue from listeners who just check out her work for a few minutes and then click away would make up for some of the difference, but she would likely need to see a dramatic increase in traffic from listeners who are simply curious.

(Her highest ranked YouTube Video, which was produced by  a 3rd party, has been viewed a total 250,000 times over 3 years. This kind of traffic would be good for $1,600 a year at $0.02 per play, minus whatever the cut may be for the third party.)

But add to all this a crowd-funding campaign like those popularized by Kickstarter and IndieGogo, and then net income from touring after expenses (if any) and a streaming-dominated industry starts to become a somewhat a feasible, although still incredibly challenging model.

But Is It Possible?

A blogger for Music Think Tank has presented one of the more compelling arguments against the possibility of higher streaming rates that we’ve come across so far. The author argues that payouts over a half-cent per stream are unsustainable for distributors. And as Spotify’s user-base and revenue grows, he contends, their streaming costs grow in lockstep, making higher payouts impossible. But although it’s the best argument we’ve seen, it still leaves a bit to be desired, and there are a couple of significant counterarguments built right in to the figures.

First of all: even as a very active Spotify user, I don’t come close to listening to the service for 2.5 hours per day, which is the amount at which the author argues that our ideal payouts are unsustainable. Although the author pays lip-service to the idea, I believe that his figures fail to accurately reflect the “Gym Membership” phenomenon, in which companies end up making significant profits on the majority of listeners, who use the service rarely or in moderation, even while they “lose” money on the heaviest users.

For instance, Netflix is happy to have a guy like me on board. Even if I were to stream two or three movies a month, the service is a great value and a tremendous convenience for me. Amazingly, at the same time, my subscription is almost all profit for Netflix. And fortunately for them, many months go by that I don’t even watch thatmany movies. So their service, just like a gym, can make a killing off the many users just like me, even if a minority of their subscribers use significantly more of the service.  Ironically, it’s the non-aficionados who subsidize the hardcore users. By the same token, it’s worthwhile to remember that hardcore users are more likely to be the earliest adopters. As Spotify becomes increasingly popular, its user-base is likely to include more casual fans who generate significant;y more profit for the service, allowing higher payouts per stream.

The second flaw with the blogger’s argument is that even if this phenomenon was properly taken into account, the author’s numbers don’t support the conclusion that higher payouts are unsustainable. Instead, it gives proof that if anything, only minor tweaks would be necessary to increase payouts to artists — if tweaks are even needed at all. What the author succeeds at showing is that even if his assumed numbers are correct, Spotify would merely have to cap listeners at 1500 minutes a month for their $10 in order to make payouts of $0.01 profitable.

That’s about 50% more minutes than I get with my cell phone plan, and at about 1/6th the price, once you take data usage out of the equation. It’s also far more listening than I’ll ever do on Spotify each month, and I love using the service.

The third and fatal flaw with arguments that higher payouts are impossible is that, if early reports are true, Spotify is already beginning to pay out rates over $0.01 a stream.

On The Rise

I’m hesitant to present the data here, as the last time we wrote an article on this subject it was to help counteract a woefully inaccurate infographic that was floating around the web.

With that said, there’s a new infographic on the loose, created by a digital distributor called Ditto Music. It states that when songs are played by Spotify’s “Premium” users, it generates payouts that average $.015 for their artists. For those of you who hate decimals, this is one-and-one-half cents, or triple that of the widely-quoted half-cent average.

We have yet to see corroboration of this figure from other sources, but they appear to be a legitimate distributor, and the numbers do make some sense.  It’s natural to wonder what’s bringing the rates down from this one-and-a-half-cent average to a paltry half-cent. But there’s clear answer in the same chart: While “Premium” plays pay out one and one half cents, the average is brought down by lower paying “Unlimited” users who generate three-quarters of a cent per stream,  by the overwhelming number of ad-supported “Free” users, who generate a half-cent per stream, and by new “Mobile” users who generate just one-tenth of one cent per listen. If these numbers are true, there’s clearly plenty of room to increase the average as Spotify’s initial “Growth” phase begins to come to a close and they begin limiting free trial accounts.

More Ways To Improve The System

Currently, the payouts from Spotify are variable. To simplify things a little, they depend on the total revenue collected by the service divided by the total numbers of plays by users at different tiers. On top of this, some artists and labels are said to have sweeter deals, although non-disclosure agreements prevent us from finding out as much about this as we might like.

Zoe Keating believes that the playing field should be completely level, with all artists getting identical payouts per stream. While I’m inclined to agree that there should be a universal minimum rate, taking away the ability to negotiate is not feasible or beneficial for artists. Simple economics dictates that if an artist or company attracts a disproportionate number of users to the service, they should and will be allowed to negotiate higher fees. In many ways, this is more fair than a completely undifferentiated method.

In fact, I’d love to see artists have additional flexibility in picking how much of their material is available on the service, and how often.

For an artist like Ms. Keating, who has a niche following and doesn’t trade in sheer volume the way a pop artist does, it might be a wise idea to limit her Spotify presence to several key tracks. This way she would have the benefit of extra exposure and revenue via Spotify, without allowing unfettered access to her complete catalog at a price that approaches “free.”

As an alternative option, I’d also love to see artists gain the ability to put a cap on how many times I can stream their tracks or albums in a month, gently encouraging me to buy an album or “tip” them for unlimited access if I become a big fan.

Not only would this encourage listeners to buy the music they really love and provide a shot in the arm to that artist’s sales, it would also knock these tracks out of the pay-per stream equation, thereby raising the per-stream rates for all artists on the service. It’s a real win-win for everyone involved, including Spotify, who would get a small cut of these sales through the service.

A Positive Spin

As we wrote last time, even at roughly a half a cent per play, Spotify is not all doom an gloom.

$0.005 per play amounts to $5,000 per million plays. For a little perspective, this means that if 1 million people listened to your song and hated it so much that they’d never want to listen to it ever again, you’d still get that $5,000. (I’d love/hate to see where this leads when it comes to provocative or misleading song titles. A search of YouTube will be some indication.)

Unfortunately, this also means that if 10,000 people loved your 10-song album so much that they listened to it 10 times, you’d get that same $5,000. Things start to get interesting when you realize that this is almost exactly what you’d earn as an independent artist selling just 800 albums through iTunes.

This is one of the reasons that not all plays should bring in equal revenue. If listeners tune in for less than half or less than a quarter of an entire song, rates should be far lower, perhaps below $0.05. Once again, this would keep these tracks from weighing down the per-stream rate on songs that listeners actually like hearing.

For now, it’s unlikely that you or I will have a direct voice in the backroom negotiations that are shaping Spotify’s payout structure. But your actions do have an impact, especially if they spread.

If you’re music fan and you’ve come to rely on Spotify, consider upgrading to the Premium service. Spotify is known to pay more to artists when premium listeners stream their material, so if you’re using it anyway, you may as well help artists to keep putting out music you love. More importantly, if there are any artists who would leave your world a duller place if they ceased making music, please buy a CD, an album download, a record, a T-Shirt, or some concert tickets. Get behind a crowdfunding campaign, or just drop a few bucks in an an online tip jar. In today’s economy, if you don’t use your dollars to vote for the music you love, it quite simply, doesn’t get made.

If you’re an artist, consider talking to your distributor about limiting the number of your tracks that appear on Spotify while the rates remain low.  Encourage your fans to buy your complete albums on a site like iTunes, Bandcamp, or on your own personal page. These outlets allow you to keep 60%, 90%  and 100% of earnings respectively, and if you like, the latter two options even let you leave it up to listeners to name their own price.This method offers the best of both worlds: You get paid when the curious check you out even for a minute, and you get paid when fans fall in love with your material and want to hear more.

At the same time, if your tracks begin to pop up on sites like YouTube against your will, generating ad revenue for Google and not for you, then be sure to report these illegal uploads and have them taken down. Any legitimate site is required to do so by the DMCA, although they’re not currently required to pay you any back revenue they may have made while hosting your material.

If you want to allow your potential fans free access to your material, do it on your terms. Giving away your art as a gift to your fans is only meaningful when it’s a choice. It happens to be a choice that I recommend you make on a regular basis. But to have your art given away without your say is an insult. Stand up. Have a say. Otherwise, if people like your art, someone will eventually use it to sell their potato chips, whether you like it or not.

This entry was posted in August 2012, Featured Stories, Industry Trends. Bookmark the permalink. Both comments and trackbacks are currently closed.
  • Sign up for our Email Newsletter


  • E-news and Updates

    Subscribe to Trust Me, I'm a Scientist Find me on Facebook Find me on Twitter Email Trust Me, I'm a Scientist