My Take on the Spotify Debate

July 18, 2013

This debate about Spotify payouts isn’t new, but it has flared up again due to Thom Yorke and Nigel Godrich pulling some of his catalog and getting vocal about it. I want to respond to some of the critics and frame the information and sticky points because I don’t think some of the criticism is fair or viewed from the correct perspective. I was greeted with this the other day, with people re-tweeting these posts in particular:




To be completely honest, I agree with their point. Spotify massively favors major labels in their payouts and independent artists get hardly anything. I know, as I represent a few independent artists, along with my own catalog. The payout is pretty insignificant. Godrich also linked this article on Pitchfork, where Damon Krukowski of Galaxie 500 and Damon & Naomi does a full breakdown on Spotify’s (and Pandora’s) payouts to independents.

We took it, which means for 5,960 plays of “Tugboat”, Spotify theoretically owes our record label $29.80.
I say theoretically, because in practice Spotify’s $0.004611 rate turns out to have a lot of small, invisible print attached to it… The payments actually received by our label from Spotify for streams of “Tugboat” in that same quarter, as best I can figure: $9.18.

I don’t want to dispute anything these guys are saying because they are correct and I don’t entirely disagree. But I want to provide some perspective on a few points. That meaning, I still think having your music on Spotify is a good idea and I think this criticism is a little unfairly skewed.

Plays vs Sales

I’ve read a number of articles containing numbers similar to what Krukowski is saying above. I recall an article showing the pittance Lady Gaga made from streams (a major label player). They love to throw up big numbers to shock and awe. 6,000 plays but only $9.18? What does this really mean? $0.00154 per play. How does this compare to a single download purchase from iTunes or Amazon?

On this blog post by David Touve, he explores the comparison much more cohesively than I could:

When you purchase a download you are paying in advance for all of your subsequent listens of that track.

When you listen to a track through a streaming music service, you are paying as you go for each and subsequent listens of that track.

In order to analyze the situation fairly, we need to understand the distinction between Plays and Sales and their long term effects. Per-per-play accumulates over time, pay-to-download is a one time payment for infinite lifetime plays. That said:

…let’s make a simple assumption: A single track purchased from iTunes will be listened to 250 times, on average, over the lifetime of the person who acquired the single. This assumption may not seem that crazy—over twenty five years will you listen to the “average” purchased song in your library about 10 times per year? Or twenty five times per year over the first ten years?

…now we can just divide the amount music rights holders would earn from a sale on iTunes to estimate a “per listen” fee—the effective rate paid over the lifetime of the track’s enjoyment for each listen of the song.

$0.70 -> the amount a label receives for the sale of an iTunes single (US)
250 -> the number of assumed listens to the track over ownership life
$0.70 / 250 = $0.0028 per listen

OK. So some quick math has led us to the conclusion that our “per listen” payment for iTunes singles over the lifetime of our enjoyment of these tracks is just short of 1/3 of a penny.

I have no idea if 250 is a good number. Based on my stats, my 8 year median for songs seems to be 40-50 plays, but I have a tendency to listen to an extremely wide selection of music, so I would assume normal listeners fair a bit higher. Pushing it up to a 25 year total 250 seems like a totally reasonable assumption.  Anyway, this gets the per-play payment from Spotify MUCH closer to iTunes than is typically represented. Yes, it is true, the difference is still big. In fact, iTunes pays double! But, it isn’t the pittance we’re led to believe when you put it in the right context. Every time someone plays a song on Spotify, the payout is the same. Every time someone plays their iTunes download, the per-play value shrinks. It takes a LOT of shrinking to get to Spotify payout levels, but let’s at least acknowledge what is happening here.

Of course, many people pay for music they download that they never listen to, so that skews the data favorably toward iTunes, but leveraging that as part of anti-Spotify argument is unfair and just feels patently greedy.

Main point here is: When looking at the theoretical pay-per-play value of a typical iTunes download, the pay-out disparity between iTunes and Spotify isn’t nearly as dramatic as presented by Spotify’s detractors. Historically, artists and labels are looking for an immediate return on investment that CD/Vinyl/Paid Downloads provide. With pay-per-listen, the income comes in more slowly, but more steadily, assuming your music maintains a fan-base and relevance over time. This stops being a question about how much one pays vs. the other, its when it pays and the business model upon which bands and labels design release strategies on. One thing I’ve heard some artists and labels do is put their music on Spotify, but only after the release has been available for purchase for a month or so. Keep this idea in mind!


Spotify IS good for music

To the next point I object to in Krukowski’s article:

As businesses, Pandora and Spotify are divorced from music. To me, it’s a short logical step to observe that they are doing nothing for the business of music– except undermining the simple cottage industry of pressing ideas onto vinyl, and selling them for more than they cost to manufacture.

This is unfairly cynical, in my opinion. Even if Spotify were bad for music business, I think it is wonderful for music, though I contest the point that it is bad for business, too. Spotify and Pandora are businesses and the nature of businesses is for them to look out for their own best interest. There is no separation between Spotify, Pandora or Krukowski in this respect, since the whole point of his article is to preserve the value of his own product, presumably at the expense of Spotify and/or Pandora. I don’t listen to Pandora but I put a couple albums up there when it was a very, very young project. What I took away from the experience, at the time, was that the guys at Pandora were music enthusiasts at heart: They found a cool methodology to introduce people to new music. Sure, they built a business around it, but I don’t question for a second that their intention was to improve the world and the way people listen to music. As someone who loves sharing music, I noted that as an important part of the process.

I know a bit less about Spotify’s origins. But to suggest it “does nothing for the music business” is absurd. Before Spotify, I ripped and downloaded all my music to the harddrive of my computer and had to upload/download to external players or burn CDs for my car, etc. Now, there is a perfect seamlessness between my desktop and mobile. Music I own that isn’t in Spotify’s catalog, I just flip a switch in Spotify to make it available offline, and then flip a switch on my mobile to download it. No cables, no sync programs, a totally seamless environment. The social elements are a tad inferior to’s but are at the direct finger tips of my music client. Between watching friends, getting recommendations, using the Apps inside Spotify and now their new “Discover” function (which is usually wonky but yields some interesting recommendations), I am perpetually discovering new music.

Spotify also curbs piracy. This was the most compelling argument for Spotify in its early days, for me. Someone pays a reasonable monthly fee, has open access to a near limitless music library. Sure, artists get less money from Spotify streams than they would if the same person had purchased their song from iTunes. But, this is the same flawed argument musicians love to throw around when crying about their music being pirated. There is no data to suggest that someone who listened to your song on Spotify (or downloaded it via BitTorrent) would have purchased your song anyway. Studies show music piracy doesn’t hurt sales, and even helps it in the case of digital purchases. In fact, using Spotify’s various tools, I discover new music all the time I wouldn’t otherwise. Hell, I read Pitchfork on their Spotify app, I wouldn’t otherwise. Consequently, I discover a fair amount of gems there, because I can read their review and with one click listen to the whole album.

Speaking as a music consumer (vs. a music publisher, which I also am), I am helping many artists earn $0.00154 that they otherwise would not have earned. Plus, now their music is on my radar and I’m effectively a fan. I may stop in and see them play if they tour through locally. If they release a really cool box set, I may pick it up. Worst case scenario, I play their album 3 or 4 times, they make $0.06 from me and I recommend their album to people I may think will like it, and the cycle continues. And for that matter, do you know how many albums I listen to on Spotify day-to-day that I actually own physical copies of? You’re double dipping about 20%-30% of my library! I can’t possibly be alone here. Spotify gives me access to a music catalog that prevents me from having to continue ripping CDs and uploading them to Google Music or something else.

The value in having your music on Spotify is branding. A lot of independent businesses resist any sort of marketing where the ROI (Return on Investment) is really low or difficult to measure, and this is a bad idea. This is really what you can’t track via Spotify (and Pandora!): It does more than pay X per listen, it is a great tool for music discovery and sharing, and puts your music in the hands of people who otherwise would never have heard of you. Sure, in some cases you may lose an iTunes sale because someone can bookmark your album in a Spotify playlist. But, what is often loss in immediate sales

For an approximate 50% reduction in profit, as an artist, it promotes your brand by increasing your fan base, provides value to your fans by way of convenience, and creates a trickle of revenue that, in all likelihood, would otherwise be lost. Also, you’re helping build a new business model, an all-you-can-eat, single fee music repository. Does it need improvement? Absolutely. But we have to start somewhere, making this an interesting point because of all the great things I list about Spotify, it has a lot of flaws and I see room for a competitor to come along and do things better. Beyond all that, I believe this business model is here to stay, we’re all better of trying to improve than fighting it.

Where I agree

All that said, I still agree with the point Yorke and Krukowski make: Spotify needs to step up and pay artists a bit more, particularly independents. If not that, be a little more transparent with how pay-outs are distrubuted. It would do wonders if we knew more about how their payout algorithms work. I think it is well played for Yorke to remove his catalog from Spotify, even if it means turning away some fans. I wouldn’t recommend this to any true independent acts, but why is it ok for Yorke? Because he doesn’t lose as much as we would, and his statement resonates much more loudly. I’ve always admired Yorke’s stance against corporate greed and feel it is a fight worth fighting. But then, even more admirable than Yorke removing his catalog from Spotify is Public Enemy releasing a single directly on BitTorrent. How much do they get per play? Zero, plus donations. That’s sticking it to corporate greed.

I also agree that, on the whole, Spotify’s pay-out to independent artists is too low. I don’t think it is dramatically too low, but it is a bit low and I feel it is a discussion worth having.

Lingering Questions

  1. What is a realistic average number of lifetime listens for a digitally purchased song?
  2. Recent studies indicate that music piracy contributes a marginal boost to digital sales. What is the net effect of all-you-can-eat streaming on direct music sales? Would like to know this for both physical and digital sales. Note, this doesn’t mean much, depending on the results. With some minor pay-out adjustments, pay-per-play could be comparable to pay to download when measured over long periods of time. It is also possible we’d see that Spotify actually causes more harm than piracy, but I can’t fairly speculate here.
  3. What is the average net affect on a band’s audience from being on Spotify or Pandora? How many new fans to do they get? I’m guessing this varies radically depending on the band’s popularity prior to their catalog getting on the services in question, but perhaps there is a scaling ratio?
  4. Is there a better pricing model for Spotify? Would it help solve problems for them to boost their fee to $15/month, and would the raised rate compensate for the number of people who would unsubscribe due to the price increase? My only current reservation for a Spotify price increase is precisely because artists like Yorke are pulling their catalog: Without Spotify being a near complete experience, its price has to remain low. But, I’d love to see some numbers predicting growth results from pricing structure adjustments. Personally, I’d pay $30/month for access to a library that was near ubiquitous, but I don’t see that ever happening, nor do I think many people would concede to that sort of monthly fee.
, , , ,