How Spotify killed off paid downloads

How Spotify killed off paid downloads


norman lebrecht

January 10, 2014

Here’s another of those helpful graphs from The blue bars show the rate of growth of paid downloads in the period before Spotify came on the scene. The black line is the strangling effect that Spotify had on the public’s willingness to pay for music. Chilling.

spotify growth


  • John Newby says:

    As the comments on the original article you link to point out, this graph demonstrates nothing of the kind.

    • Tony says:

      I don’t agree, John. The trend in electronic distribution of recorded music is away from purchased downloads towards subscription based streaming and this graph demonstrates that. The trend also manifests itself with movies where the Netflix model is doing very well.

      The problem Norman is highlighting for music is that revenue from streaming for record companies and musicians is insufficient to enable most new recordings to break even let alone make a profit. Useful for back catalogue as a bit of extra income, but not much more. CDs still sell, especially as concert souvenirs, but electronic delivery will dominate in the end. With video electronic delivery will dominate rapidly, especially as more people will want UHD-4k for their movies, sport and music videos and the BBC will be slow off the mark way behind Netflix and YouTube.

      Electronic delivery threatens the power of broadcast radio and tv as well as the income for record companies and musicians. The BBC will need to pull its socks up and not be so dependent on repeats, Eastenders and Brucie’s Come Dancing if it wants to be able to justify demanding a licence fee. We already watch more Netflix than BBC in our house and are hardly alone in seldom listening to BBC R3.

      • Anon says:

        Sorry, Tony, but the graph does NOT show a trend away from download towards subscription. Indeed, the graph shows a continued modest increase in download activity, even while streaming is growing in number of users (as it happens, telling us little useful about the amount of actual streaming in any case).

        Norman’s assertion that the black line shows some sort of “strangling effect” is not really true either. It shows that the number of Spotify users is increasing, but without more data it’s entirely possible that part of the reason the users are increasing is because subscriptions are on the rise – i.e., people paying for access to music.

        I don’t see anywhere that Norman is highlighting anything to do with “revenue streaming… insufficient… etc.” as you assert, where do you read this?

        (And to address it as a point anyway, can I suggest that it might not be that the public are unwilling to pay for stuff, but more that a huge army of artists and content creators have effectively flooded the market with a vast array of releases in recent years, recordings being rather cheaper to make these days, so perhaps over-sully is one half of the problem?)

        • Tony says:

          Bill Clinton is quoted as saying that “De Nile is not just a river in Egypt” or something along those lines.

          The numerical specifics of the graph display may not be 100pc accurate, but the message most certainly indicates where the distribution of music and home entertainment in general is going. The model of subscription streaming is irreversibly strong and dominant, because it is the ultimate in convenience and remarkable value for money. That does not mean anyone (least of all yours truly) feels like sitting on the beach commanding the tide not to come in. There is no sense in ignoring the reality unless you enjoy an unwelcome dowsing.

          UHD-4k video will accelerate the trend because at present the only way it is going to find its way into most homes is via broadband internet connection, and find its way it will because 4k is fabulous for films, sport as well as music. That Spotify inter alia denies proper remuneration to the owners and makers of recorded music is a separate issue, and plain old fashioned opportunist greed which will possibly bite them at some point in the future – maybe it won’t – not for me to say.

  • Rob Kerr says:

    Where’s the corresponding data on Spotify’s subscription access model? Contrasting decline in downloads with rising numbers of users doesn’t give the whole picture, as it ignores the trend in paid-for streaming services (in which Spotify is the major player/innovator).

  • Tim Benjamin says:

    And you think the two are linked? Maybe they are, but you couldn’t possibly conclude that from this graph.

    From the chart: there is a decline in the rate of growth of paid downloads 2006-2009, as one might expect as the market grows. It couldn’t keep going at 60% growth a year could it!

    Then the market maintains growth until 2013 while Spotify’s user base grows. The two don’t appear linked at all. And even if there was correlation (which there isn’t IMHO), you know perfectly well that doesn’t imply causation.

    Paid downloads were only ever a stopgap between physical media and always-on streaming services. If the streaming services ever get good enough I don’t see a need to own physical media as well. As a listener I get nearly everything I want from Spotify (except ultra high quality audio and proper gapless playback…), in any case, it’s tremendously good value for money.

    Obviously as a recording artist you might have a problem with the royalties from Spotify. Or more to the point, with your record company taking a far bigger wedge than they do with CD sales or radio plays. But to try and somehow campaign against streaming services on that basis is to tilt at windmills I’m afraid. That particular Elvis has left the building!

  • robcat2075 says:

    The graph is intentionally misleading. Equating annual year-over-year growth with total accumulated users is misleading.

    The actual numbers there show downloaders (currently about twice the number of Spotify users) have stayed fairly steady since 2010, they haven’t been “killed off.” Spotify was new in this time frame. It had nowhere to go fro zero but up. Downloading is a mature market that could only level off. The graph shows it was already leveling off before Spotify appeared.

  • David H. says:

    Spotify kills the music industry content creators.

    • David H. says:

      More precisely, first it kills the content creators (recording professionals), then the content providers (labels), then itself.

      It’s a typical parasitic non sustainable business model built on nothing but greed.

  • Ok this is not correct. What then accounts for the 20% decline in paid downloads when spotify has no visible signs of growth from ’06 to ’07? And then again another almost 20% decline from ’07 to ’08? By 2009 with what appears to be a 2% share of the market, paid downloads are down what looks like 60%. This just can’t be correct. All my years temping at a management consulting company make me highly suspicious of this.

  • Peter Moody says:

    This is a very odd graph. Percentage change (presumably increase?) against time merely shows that the increase in downloads is decreasing – normal in a maturing market. This rate of decrease actually changes very little after the high rate of increase of Spotify users starts. The change of less than 0% in 2013 is incomprehensible – a percentage change cannot be negative. That is really as far as I can go without using specialist mathematical terms.

  • Anon says:

    Um, not really.

    1. If you think the comparison is a valid one, then the growth rate of downloads was slowing significantly before Spotify began (i.e. pre-2008), and would seem to have stabilised as Spotify has grown, relative to the 2006, 2007, 2008 figures. So it doesn’t show what the headline says!

    2. But actually the comparison is a daft one anyway. Any successful service is likely to have higher percentage growth in early days, and as it expands the percentage growth compared to current size will get smaller, so the “growth of paid downloads” figures might not be so surprising.

    Comparing rate-of-growth of downloads vs. Spotify subscribers tells us very little, and is a meaningless comparison.

    Two rate-of-growth comparisons might be interesting, two number-of-songs-downloaded vs streamed might be interesting…. As it is, this graph doesn’t say a lot.

    I mean, look at that vinyl sales trend posted on this blog the other day – you could replace the Spotify subscribers line on this graph with the vinyl sales trend line and it would look similar in trend, but I doubt we’d say that means vinyl is killing off paid downloads!

  • Sorry for my english, I’m not native.

    To be exact and to avoid a misunderstanding there is NO DIFFERENCE between streaming and downloading. In fact if you are listening a stream is because the stream is being downloaded to your computer in a temp file or a temp memory buffer. And as is a download, anyone can save it to his hard disk using legal software fools proof like audials or replay. The fact is that whatever you are listening or browsing not only can be retrieved later or in real time by you only for joyful purposes, but for forensic ones.

    As a composer in the nineties there was a programm named “Payback for Playback” in the Michael Robertson’s which awarded almost $10.00 for every 300 playback. It gave me the chance to collaborate with Doctors Without Borders during many years as I’ve got more than a million and half of downloads. All my profits were donated to this NGO. This programm was more fair for authors.

    The Spotify model is far, far away from this programm and has killed off the paid to the artists and authors in because anyone can save it for later what he likes. The fact that one do not know how to do it, does not implicate everyone do not knows how to do it legally.

    The streaming service does not compensate the playback. And remember one thing. If Spotify can build its business model with my music among others, why they do not need to pay like than any radio or TV broadcast for every playback?

    Check this to know more about the fair programm:

  • A bizarre misreading. The graph says no such thing. It says that paid download growth was declining precipitously well before Spotify got off the ground, and further that there is no clear relationship between Spotify and paid download growth rate!

  • Clive Paget says:

    Sorry but the graph actually shows that download growth had declined almost to zero before Spotify actually started. It’s more surprising that since 2008 download growth has merely fluctuated to a modest extent suggesting Spotify has had little effect here