What Will it Take for Music Streaming to Become a Good Deal for Artists?

It’s no secret that recorded music sales have dwindled dramatically since the dawn of the Napster age, more than 15 years ago. When adjusted for inflation and population growth, U.S. recorded music sales in fact dropped by an astonishing 64% between 2000 and 2009, and they have been slow to begin digging their way back out of that deep hole ever since.

Although professional musicians have found new ways to make a living in the 21st century between licensing, crowdfunding, merch sales and ever-rising ticket prices, there are indeed fewer of them working professionally today than there were 15 years ago. It is likely that this has something to do with fact that nothing to date has quite replaced the revenues lost to music piracy.

Some musicians today make the mistake of vilifying the new wave of legal music streaming services. But they are not the cause of the problem at all. If anything, they are part of the solution. The success that services like Spotify, Pandora, Apple Music and Tidal have had with consumers mark a small step in the right direction: They are part of a wave of re-monetization in music.

A chart of recorded music revenues adjusted for inflation and population growth.

Unlike pirate websites, these legal services are based on consent, meaning that unlike pirate websites, no one is forced to make their music available on these services if they don’t like the terms of the deal.

However, this is not to say that the deals currently being offered by streaming services are good ones. In my opinion, they still pay far too little, by a factor of as much as two to three times.

So what would a good deal look like? And how would you go about getting one? To put this all into perspective, it’s time to crunch some numbers.

“Records” of Account

To answer the question of whether a streaming service offers a deal at least as good as the album-sales model of the past, it is useful to ask: “Just how much did musicians make off of album sales?”

Obviously, some of them made millions, and some of them never paid back the debts incurred by recording their magnum opuses. However, we can still break down the essential structure of the deal, using some very basic math that even us musicy-types can understand.

To do this, we’re going to have to make some assumptions (and leave out some small details for the moment.) But we can revisit these and refine our numbers once we’ve established the essential framework.

Crunching the Numbers on Record Sales

First, we’ve got to make some key assumptions: What’s a reasonable number of record sales to plug in to our equation?

To keep this simple, realistic and relatable, let’s assume we’re thinking of a new artist making their debut on a small but respected indie label.

When I go through this exercise with students in the college courses I teach, the answers vary, but most students say they’d be pretty impressed and pleased to have sold about 10,000 copies of a debut album over the course of a year at the indie stage.

This is a nice round number, and although many aspiring professionals may not ever do this well, it’s not too ambitious of a number either. With the right coaching, some talent, persistence and hard work, this is definitely an achievable goal.

Next, let’s assume that we are thinking of a 10 song album that costs $10 to buy. Now we can get started:

10,000 records x $10 = $100,000 gross revenue

Not bad. Then, lets’ assume that the artist is selling these exclusively through iTunes and Amazon download, which take a 30% cut of sales. (For distribution in brick & mortar stores, you might assume 40% of sales.)

$100,000 gross revenue – 30% = $70,000

Still pretty good. Now, let’s assume that this artist had the support of an independent label to help get this kind of reach, and that they had a 50/50 deal with that indie label.

(For a major label, you could assume the label would take 80%-90% instead, but on a potentially higher total number of sales. For self-released albums, the same artist would keep 100% on a potentially smaller number of sales.)

$70,000 – 50% = $35,000

Alright: By selling 10,000 records, this new upstart indie band has made $35,000 to split among themselves.

(Minus any production and promotions costs they took on themselves, and before the tax bill comes and eats up to another third of the total, that is!)

Not bad at all for a new band. But you’d hope it’s not their sole source of income.

Add on some touring revenue and merch sales, a crowdfunding campaign, a little bit of music licensing and maybe a day job or two, and there’s a chance they could justify continuing on in their hopes of making music for a living.

Of course, this simplified calculation leaves out a few variables. Mechanical royalties, taxes, the costs of production and promotion, and how the revenue is shared between band members and songwriters, will all affect what percentage of this revenue will ultimately end up being profit in the pocket of the artists.

We could certainly improve this calculation further by adding more details, or by making even better assumptions. Still, this example provides a pretty decent quick and dirty approximation of the kinds of gross figures that are really at play. Now, lets do the same thing for streaming services.

Crunching the Numbers on Streaming Services

The last time I wrote about Spotify for SonicScoop, this market-leading music streaming service was paying out roughly $0.005 per play on their ad-supported service and $0.015 on the premium service, for an average pay-per-play rate of roughly $0.009.

More recently, Spotify has written that their current average rates range from $0.006 to $0.0084 per play—probably because use of the service has grown faster than its revenues.

So that we’re working with an easy number and a fair assumption, let’s go for something right in the middle of this recent range and assume $0.0075 per play.

Bear in mind that these payout rates come after Spotify has taken its 30% of the pie. So our equation will be a bit simpler this time around.

Once again, let’s assume that we’re dealing with the same artist, and that artist is still capable of “selling” 10,000 copies of their debut album.

But this time, we’ll assume that for each sale, instead of purchasing a physical album or download, fans simply listen to that album on Spotify as many times as they would have if they had bought it instead.

Here’s where things get tricky and we have to start making somewhat shakier assumptions. Specifically: How many times have you listened to the albums you’ve bought, on average?

For some of us, there are albums we’ve bought and listened to hundreds of times. Other times, we’ve purchased records and only listened to them a handful of times.

When I ask my college classes this question, they most commonly determine that something like “10 times” is a pretty fair assumption to make when you average it all out.

This has the added benefit of being a nice and easy number to work with, but if you think you have a better assumption, feel free to fill it in. In the meantime, let’s break it down using this figure:

10,000 fans x 10 songs x 10 listens each = 1,000,000 total plays

Now, let’s fill in our $0.0075 per play figure:

1,000,000 total plays x $0.0075 = $7,500

Then, assuming the artist is still on an indie label with a 50/50 split:

$7,500 – 50% = $3,750.

Ouch. This is almost 10x less than under the album sales model.

But wait! We’re not quite being fair yet. We are leaving far too much out.

To find the 10,000 fans who like your music enough that they’d listen to your album 10 times, how many people would have to a song of yours at least once to begin with?

As great of an artist as you may be, it is quite unimaginable that you are so great that every single person who hears one of your songs one time will run out and purchase or otherwise listen to your album 10 times in row.

If you had that kind of success with 1 in 10 people who heard your single, you’d be doing an amazing job, and would be putting Taylor Swift, Michael Jackson, Adele, Nirvana and The Beatles to shame. In which case, please stop reading and go enjoy being a kajillionaire.

So let’s be both generous about your awesomeness and conservative about your prospects at the same time, by assuming that for every 100 times someone hears your single in full, 1 person buys your album. This means that to find 10,000 fans, you’d have to have your music heard at least 1,000,000 times.

Remember: You still get paid for those 1,000,000 plays that did not result in a “sale”. You even get paid if 90% of those listeners hate-listened to your song, ruthlessly mocking your inability to carry a tune the entire time.

So now, we have to add another 1,000,000 plays into the equation, so it becomes:

2,000,000 total plays x $0.0075 per play = $15,000 gross

$15,000 gross – 50% for label = $7,500 for the artist.

This is getting better. Now we’re only doing 1/5th as well as we would have with the album-sales paradigm.

(In case you were wondering, 1,000,000 plays on the radio would pay about $41 last time I checked, so yes, it is fair to leave this off of our earlier equation.)

But are we really being fair yet?

Admittedly, it’s somewhat difficult to convince 10,000 people to give you an extra $10, even if they do like your music, and even if piracy is not an option.

With that in mind, couldn’t we safely assume that it may be easier to get people to listen to your album 10 times if they don’t have to pay anything extra for the privilege?

OK, agreed. But how much easier? Maybe 2 times easier?

Obviously, this is something that’s difficult to quantify, especially in the absence of any great research to go on.

But at some point, to make any financial projection, we’re going to have to make some assumptions. If you feel you have a better assumption, feel free to fill it in here. In the meantime, I’m going to err on the side of caution add another half million plays to our equation.

But wait! There’s more.

Out of those original 1 million listeners who had to hear your music for you to find 10,000 fans, are there some who may have liked your music enough that they might listen to 1 song 10 times, or the whole album 1 time, or some combination thereof? What about the hardcore fans who might listen 100 times or more? How many of them are there?

We’re getting waist-deep into the murk of assumptions at this point. The less solid evidence we have to suggest that these assumptions are fair ones, the less we can rely on them and the more we have to be suspicious of the number we get in the end.

But once again, to make any kind of reasonable financial projection, you are at some point, going to have to make assumptions.

(In both science and in finance, we generally start with “I have no proof, but the sun will probably rise tomorrow” and then go from there.)

Better data yields better assumptions, which in turn makes for better financial projections. But ultimately, the basic framework is the same.

So, admitting that there’s a chance we’re missing something here, for the time being, let’s plug in these numbers:

500 hardcore fans x 10 songs x 100 listens each = 500,000 plays
5,000 extra fans x 10 songs x 10 listens = 500,000 plays
10,000 core fans x 10 songs x 10 listens = 1,000,000 plays
100,000 casual fans x 1 song x 10 listens = 1,000,000 plays
1,000,000 listens to attract these fans = 1,000,000 plays

Now, we have a total of 4,000,000 plays x $0.0075 = $30,000.

After a 50% take for the indie label, the artist now has $15,000.

Comparing Apples

It’s easy to see that $35,000 is quite a bit bigger than $15,000.

Are our assumptions perfect? As mentioned, probably not. They can be improved with real data on how real people actually behave in the real world. Of course, even then, different artists will have wildly different outcomes, and wildly different levels of success in attracting listeners.

Still, this is the basic framework for finding out how good of a deal streaming is when compared to an album sales model. Provided these assumptions are reasonable ones, it suggests that a per-play rate two to three times greater than the current rate would make streaming be just as a good of a deal for artists as the old record-sales paradigm ever was.

What does this mean to consumers and to the services themselves?

If we are to conclude that an average payout rate of $0.015 – $0.03 would be a good deal for artists, and a streaming service cost $10/month, this suggests that each and every listener would be able to play between 333 and 666 songs per month for their $10.

Assuming an average song length of 3.5 minutes, that’s 20 to 40 hours of listening per week for $10.

But wait, that’s not quite right, it is it?

Assuming that the streaming service will still want its 30% cut to keep the lights on and take home a bit of profit some day, the rate would actually have to be closer $14.50/month. Nearly 50% more than it currently is.

But price hikes aren’t the only option. For instance:

Streaming services could allow listeners closer to 200-400 songs or 15-30 hours each per month and keep the rate closer to $10.

Or, fans who listen less than this amount could end up subsidizing fans who listen more and this way, the rate could stay lower as long as the average number of total plays stayed within this range.

Or, artists could allow a certain number of free listens to their album, past which point the fan must “buy” the album or “tip” the artist to unlock unlimited listening of that album. (Without biting into his or her per-month maximum of course.)

There are a wide variety of creative ways like these to help ensure that artists some day get a truly fair deal from streaming services.

What You (And We) Can Do

In the past, I have encouraged artists to put only a small portion of their work on these services and—for the time being—to think of them more as promotional and discovery vehicles than as revenue-generators.

I am happy to see more artists taking that kind of approach today. Most notably, major artists including Taylor Swift and Adele have used their cachet to hold out for better deals and encourage fans to pay for the music they most want to hear.

The more that artists and their teams crunch these numbers for themselves, and the more freedom that they and streaming services have to craft deals that work for them, the better the outcome will be for the musicians, the fans, and those that help connect them to audiences.

However, this kind of experimentation can only occur in an environment where the artists’ consent is considered and respected at all times.

Musicians don’t need any special favors from government or labels or streaming services or anyone else in order to create the environment for this to happen. All they need is one thing: Respect for their right to decide if, when, and how others profit from their work.

If we can once again agree to respect and enforce basic human rights to intellectual property (that are both reasonable and fair in the context of the 21st century) that is all we’d really need for a healthy and free creative economy to flourish once again.

Justin Colletti is a writer, educator and mastering engineer. He edits SonicScoop and has a passion for music, audio, economics and finance.

Please note: When you buy products through links on this page, we may earn an affiliate commission.