Coldplay’s decision to keep their latest album, Mylo Xyloto, off all streaming services raises a series of awkward questions about the streaming business model; especially the ad-supported component.  They are questions that have been lurking at the back of the minds of various parts of the music industry for some time, but that have been largely avoided because of the fear of opening up a proverbial can of worms.


Artists are losing faith in streaming

Coldplay’s move was a business decision based upon their preference to have a smaller number of people listening to Mylo Xyloto and buying it, rather than a larger number listening but a smaller number buying.  There are two implicit assumptions at work here:

  • Streaming services don’t generate enough artist income
  • Streaming services cannibalise CD and download sales

There is a growing body of thought, especially within the artist community, that both assumptions have merit.


Why don’t the economics of streaming add up?

Spotify is undoubtedly a great success in terms of user adoption, Deezer is also gaining great traction.  Between them these two services account for 30 million users.  The problem with streaming services is that they generate so little income per activity to artists compared to downloads. In my blog post showing PledgeMusic’s Benji Rogers’ digital income I showed how the average payout per activity for streaming services (premium ones included) is over 300 times smaller than the average pay out per activity on iTunes. Now to be clear, we are not comparing apples with apples here (no pun intended).  An activity on iTunes is a one-off paid download, whilst an activity on a streaming service is one stream and that play could occur multiple times for the same song.  Yet it still leaves a rather large number of plays required before you start catching up with an iTunes pay out.


Where is the money going?

So how can it be that so much scale can appear to deliver so little artist income?  Streaming services already feel that they pay out too much to rights owners: services typically pay out in the region of 80% of their income to rights holders. So increasing their royalty payments would likely put many services out of business, unless of course they hiked their prices. But 9.99 a month is a hard enough sell as it is, let alone anything higher.

There are three possible reasons for why the economics of free don’t yet add up:

  • The long tail is getting mined, and some. One possibility is that users of streaming services are spending their time listening to such a vast diversity of catalogue that any one artist only gets a minimal amount of plays and thus only small pay outs.  However, with discovery features so weak on most services, the opposite is more likely to be true for the majority of users.  Indeed 24/7’s CEO Frank Taubert once stated that a third of 24/7’s catalogue had never been downloaded, not even once.  (24/7, remember, is the service that powers the remarkably successful TDC Play unlimited music service in Denmark)
  • Messy metadata is to blame. Streaming service metadata is a complex beast.  So many different sets of fields from different rights holders have to be blended into one massive dataset by each service, and each time in a slightly different way.  There is always going to be room for error.  This may be causing some proportion – possibly a significant share – of plays not getting reported.  When Benji Rogers decided to test how well Spotify paid out, he left his albums on permanent stream for a month.  Yet his digital income reports for that month not only fell well short of that number of plays, some of the catalogue was listed as not having even been played once.  Given the complexity of rights reporting it is unrealistic not to expect at least some loss of  data quality along the path of point of listening: in-service reporting; in-service data cleansing; data warehousing; distributing data to rights holders; rights holder data analysis; rights holder accounting; rights holder pay outs to artists.
  • Rights holders aren’t distributing all royalties appropriately. The conspiracy theory is that the big bad labels are collecting swathes of digital income from streaming services and then secretly squirreling away the majority of it for themselves.  Though this is less likely than it may seem, there are a number of label practices which can cumulatively contribute to creating the effect.  All artist/label contracts have stipulations about recouping costs – some of which are skewed against artists – and most have different stipulations about digital pay outs.  So there are contractual and accounting reasons why some artists will not see all the income they expect. The notoriously Byzantine accounting practices of major labels are another potential factor.  The Achilles Heel of major label public relations, questionable accounting practices have resulted in many an artist horror story.  The possibility of sums of unpaid royalties, stuck in escrow somewhere until forgotten about is every artist’s nightmare.

The likelihood is that all three scenarios play a role.  I don’t believe that any party, Spotify or the labels included, have intentionally embarked on strategies to cheat artists out of money.  But there is a distinct possibility that not all involved parties are exactly incentivised to plug the holes in their processes to thus bring the increased accuracy and effectiveness which could result in larger artist pay outs.


Are artists left disadvantaged by new record label commercial relationships?

The waters are further muddied by major labels becoming stake holders in some digital services, raising the prospect of portions of income from those services being joint venture income and therefore not subject to reimbursement to artists.  Add to that the issue of the large advances services have to pay labels in anticipation of actual revenues, how much of that is paid to artists, and when, and especially if the service doesn’t ever generate the income guaranteed by its advance.

All these are valid issues that would benefit markedly from an open dialogue across the value chain. Spotify is left looking like the pantomime villain but is likely no more than a cog in a machine that nobody seems to really want to fix other than the artists.

But fixed it must be.  Spotify and YouTube massively outpace most other digital music services in adoption and usage, yet they deliver a tiny fraction of the income.  Artists cannot afford for these services to behave like radio (i.e. the tool to drive sales) when they are also becoming the end product for many music fans.


The case is clear for a transparent and robust dialogue between labels, artists and services.

Coldplay have the benefit of being big enough to dictate terms.  Most other artists don’t have that benefit. Greater transparency, effectiveness and accuracy in revenue reporting and distribution will help drive not only artist trust, but, via increased income, greater support too.  The alternative is that piracy gets another free shot at goal, which is what Coldplay have already likely delivered, driving many Spotify users back to torrents to find Mylo Xyloto for themselves.


This is an edit of a post originally published on Mulligan’s own blog. Mulligan is an independent music industry analyst and frequent midemblog contributor (his last post, on how the music industry business model needs fixing, is here).





About Author


Research Focus Mark serves Consumer Product Strategy professionals. He is a leading expert on music and digital media. Mark came to Forrester through its acquisition of JupiterResearch, where he ran Jupiter's European research and data teams. Prior to joining JupiterResearch, Mulligan spent five years working in media consultancy and market research managing projects and delivering international projects. Previous Work Experience Mark came to Forrester through its acquisition of JupiterResearch. Mark joined Jupiter in May 2000. He regularly appears on TV, radio, and in the press, commenting on the European music industry, including BBC television and radio, ITV, CNN, BusinessWeek, The Economist, the Financial Times, The Guardian, and The Wall Street Journal Europe. He frequently speaks at conferences such as Popkomm, In the City, London Calling, and Midem. Mark was also a recording artist.


  1. Pingback: Spotify Exec Calls Apps ‘Arguably [Spotify's] Biggest Announcement’ Since Launch | Evolver.fm

  2. Pingback: Spotify Is "A Force To Be Reckoned With" Says Exec Chris Maples

Leave A Reply