That’s the way you do it
You play the guitar on the MTV
That ain’t workin’, that’s the way you do it
Money for nothing and your chicks for free

So said Dire Straits in 1985. When I was a kid, I legitimately thought that the lyrics were “money for nothing and your checks for free”. As a ten year-old it just made more sense — if you were a rock star your bank didn’t make you pay for new checks.

About 20 years later, I did get a check for playing the guitar on the MTV. It was $265 for a 10-second clip of a guitar solo of mine shown on the MTV reality show Sorority Life in an episode entitled “Birthday Bitches”. The song was called “The Ballad of Disco Stu”. Somewhere, Mark Knopfler of Dire Straits was shaking his head at the demise of rock n’ roll even back then. By then (early 2000s) the MTV era was very much over, though there was still money to be made for small independent music makers.

Fast forward another 20 or so years. If there was a song entitled “Money For Nothing” now I’d assume it to be about AI or crypto. I’m still doing music though. In fact this post is stealth promotion for a new album that I’ve just put out with Josh Silverbauer, the second volume in his analytics rock opera “User Journey Vol 2: The Custom Dimension“.

This new album won’t make any money. I don’t mean that album revenue won’t cover the cost of production, which it surely won’t. I mean that it will make a negligible amount of money period.

It was a super fun project, I’m proud of the work we did and I think it’s quality music — but the economics of streaming is stacked against small artists. We won’t even come close to that $265 check I got in the early 2000s.

We made about $25 in royalties from the first volume of the rock opera. To be precise: $.0052 per stream, with the majority of that money coming from Apple Music. Hopefully we can match that number with Volume 2… or maybe even best it, so Josh can buy a celebratory cookie to go with the sandwich and coffee that he bought with the revenue from the first album. For an album with many hundreds of listeners, intuitively that seems like a pretty low number.

Why is this? The first answer is that there’s simply a lot less money in recorded music now than there was around the turn of the century at the height of CD sales. The RIAA’s US music revenue database is a good source of data here, showing total inflation-adjusted 2023 revenue down 36% from the lofty heights of 1999.

Still, the RIAA’s numbers make 2023 look normal enough when compared to eras of music other than the peak CD era. However I would maintain that pre-digital eras had a lot more untracked sales and are thus probably underreported. I know most of my physical media purchases back then never got tracked by SoundScan and the RIAA.

There is still definitely money out there of course. Spotify says that they paid out $9B in royalties last year: their highest payout total ever. This era may in fact be pretty good for those at the top, but many artists claim it’s harder than ever to make a living as a working musician. I expect that AI will make this even harder.

Partly as a preventative defense against AI-created tracks, in April 2024 Spotify changed their rules such that only tracks which get more than 1,000 streams per year are eligible for royalties. This is why much of the revenue from User Journey: Vol 1 came from Apple Music, since none of the individual tracks hit that threshold on Spotify after April 1.

Thankfully, Spotify isn’t keeping that money for themselves. They are still distributing it, just to artists that are higher up on the ladder. Their logic is that this helps professional artists that are trying to make a living at music. They also correctly point out that tiny payouts like $1 or $5 don’t meet the payout thresholds that the music distributors have, so artists can’t get their $2.83 out of DistroKid or whoever they use to actually upload their music to the streaming services. This isn’t a terrible argument, but it does take away an important first step on that ladder.

Back when people bought physical media, a band selling their own records was one of the best ways to fund band activities. At that small scale, nobody is making a living doing music — but recorded music could ideally fund making more music. In the era of physical media, bands might make a limited number of CDs and sell them at shows, to friends, on consignment at record stores, etc. A short run CD might cost $3 per unit and be sold at $10, netting $7 a sale for the band, assuming that they were all DIY and had no label. The big disadvantage was your band needed to somehow scrape together enough to actually press those CDs, but the upside and margins were pretty good.

Let’s take a look at my old band The Adjusters, active from 1995-2003 and solidly in the “peak CD” era. We never sold a huge amount of records, but the thousands that we did sell paid for a lot of what we did as a band. With the same amount of listeners in the streaming era we likely wouldn’t have been able to exist as a band.

We had about 6,000 listeners last year on Spotify, which earned us about $60. Obviously not all of those listeners would have bought a CD, but even if 5% (300) of those did we’d be talking about $2,100 in CD profit vs. that $60 in streaming profit.

It’s kind of amazing to me that The Adjusters continue to get any revenue at all, since the band hasn’t released a record in over 20 years and never had any hits. One big benefit to older bands like The Adjusters is that the streaming revenue has a long tail over many years, vs. CDs where all the revenue was up front. However as a young band starting out, we needed that record sales revenue simply to keep the band going.

Selling a couple hundred CDs isn’t going to allow anyone to quit their day job… but it might help pay for needed gear, software, transportation, practice space, etc. Having a band itself be self-sufficient in terms of funding is really positive. For example, it can help avoid scenarios where some members are having to carry more of the expenses than others (definitely not speaking from experience, why would you think that?).

The popularity of music follows a power-law distribution. We’re all familiar with this kind of distribution in the form of “the long tail” that is part of so many discussions about the internet. Here’s a simple power law fit based upon the artists with the most number of all-time streams on Spotify.

So yea, definitely a power law distribution. At the very top we have the megastars, like #1 Taylor Swift with an eye-popping 94B streams and #2 Drake about 13B behind with 81B streams. Then once we get out to artists near the end of that top 1,000 it has started to flatten out some… we have Willie Nelson (#995) with 2,280,239,600 and then Moby (#997) with 2,276,496,283. That seems pretty flat, but when we’re talking about millions of total artists the plays & revenue trail off pretty quickly, leaving only a small amount of artists that actually make substantial revenue.

Spotify reports 66,000 artists making $10k+, and they identify 225,000 as the number of “emerging or professional recording act globally” they want to support. In other words, they are saying there’s 225k pro or wanna-be pro recording artists in the world. Except that less than 1/3 of those making $10,000/year on the biggest platform out there. So 2/3rds of those are definitely on the “wanna-be” side of making a living. Spotify suggests that their service makes up about 25% of most artists’ royalties so we should multiply that $10k by 4 to get total revenue. Meaning if you made $10k from Spotify you made $40k total.

Whether it’s 66,000 or 225,000 — that’s a tiny percentage of people compared to the 10M+ that have uploaded music to Spotify. Spotify compares this situation to professional soccer, where there are millions who say they play the game but only about 130k who actually make money doing it. This sort of winner-take-all marketplace means that recorded music really is not a viable career path for all but a lucky few. In some ways it’s even worse than pro sports, because there are older acts like The Beatles, The Stones, et al. parked indefinitely at the fat head of the distribution.

Let’s also not forget that those royalties may be split between multiple people.

This means if you’re a “working professional musician” these days, you better find some sources of revenue other than your recorded music. The US Bureau of Labor Statistics reports 35,000 musicians employed in the US in 2023. This is down from 52,000 in 2000, and considering the overall growth of the workforce this is even a bigger drop, down well over 40% in terms of percentage of the workforce.

In these stats “musician” was never a very populous profession overall, but when limited to those listed by the BLS as “Independent Artists, Writers, and Performers” it becomes a microscopic, with only 690 identified as such. That’s such a tiny number that as former independent musician myself, you’d think I’d know all 690. As very non-scientific proof of that, in fact I did recognize one of the names in the New Yorker article I linked at the top of this article (one of his bands and mine had toured together in the ’90s).

For context, there’s 193,000 data scientists and 1.7M software developers. There’s twice as many database administrators as musicians. Sadly I was unable to find the number of telephone sanitizers.

Assuming we’d like to turn this around, how might that happen? Should we go back to physical media?

Despite the incredible resurgence of vinyl in the last few years (now accounting for 7.9% of recorded music revenue and over 2.5x the revenue of CDs), that’s not gonna happen. Convincing people to support local & independent bands is a great idea, but a lot of this is in the hands of the algorithms. Spotify does a pretty good job at recommending things you’d like based on your taste profile, but it doesn’t usually dig too far down on that power law graph. Why would it? The stuff at the top is proven to be popular with people, and music services more focused on discovery haven’t done as well.

“Stealth” promotion, remember?

Share this post: on Twitter on Facebook on LinkedIn