Missing the boat in music

When Spotify and Pandora were starting their streaming services many were quick to point out that Apple was about to be disrupted. The future, they said, was streaming because (young) people could not be bothered with ownership of music and the limitations of a personal collection. Who would want to pay for a few hundred songs when they could listen to millions for free?

This perception continued and became more vocal over the years. Seven years in fact. Spotify collected 20 million paying subscribers while Apple did nothing. Pandora grabbed 80 million active listeners and possibly 4 million paying subscribers while Apple did nothing. The boat had sailed and Apple was not only not on it but oblivious that there was a boat in the first place.

At first Apple launched a half-hearted streaming service and then a paid service finally showed up with Apple Music in mid 2015. Since then the company managed to add 15 million subscribers. A tiny number compared to the 900 million iTunes accounts it had reported a year earlier. Pathetic. The number of music subscriptions relative to iCloud accounts, iTunes accounts and active devices is shown the the graph below.

Screen Shot 2016-06-20 at 3.30.38 PM

It may be paltry compared to the count of users Apple may have in total, but how does a 15 million user base in 1 year compare with the growth rate for the incumbents Spotify and Pandora?

The following graph shows the ramps for Spotify, Pandora and Apple Music since their moments of market entry. The accumulation of users by Apple looks to be the fastest yet.

Screen Shot 2016-06-20 at 3.31.02 PM

This is, of course, due to a maturing use case. Apple did not have to educate people to the notion of music as a subscription. It could just announce it and users would discover it and just sign up, especially if they were already iCloud subscribers and had a credit card attached to their iTunes account.

But that’s the whole point. Apple did not have to move first in music subscriptions. It did not even have to move second or third. When it did move it could just skim the market and add to its already healthy Services revenue (orange line in the first graph above.) Missing the boat in music in this case meant capturing all the value quickly and with minimal expense.

Fundamentally, Apple’s entry into music subscriptions was a sustaining effort. Streaming sustained Apple rather than disrupting it. The difference may seem merely one of semantics, but it is also the difference between life and death for a challenger. Meaning matters.

This is a cautionary tale for those who would pronounce every new idea as “disruptive” to Apple or anyone else on the basis of novelty alone. The tests for disruptiveness are easy enough and it behooves the analyst to apply them before dropping the d-bomb.

  • jwan584

    To quote directly from Christensen’s website:
    “Characteristics of disruptive businesses, at least in their initial stages, can include: lower gross margins, smaller target markets, and simpler products and services that may not appear as attractive as existing solutions when compared against traditional performance metrics.”

    Let’s see how this applies to streaming music vs downloads:
    – Streaming music has lower margins than digital downloads, as evidenced through Spotify’s financials vs. Apple’s content category.
    – Streaming music revenue / TAM is smaller than digital downloads
    – Streaming music is simpler than downloads. There’s no need to manage storage or transfer files.
    – Streaming music appears unattractive to those accustomed to owning music and has inferior selection (eg. lacking Taylor Swift).
    And to quote your favorite attribute of disruption—it has a different business model—it’s subscription based as opposed to pay per download. Further it competes on asymmetrical merit.

    So going by the most classic definition of disruption, streaming music absolutely qualifies as a disruptive technology vs. downloads.

    Apple’s response is even classic. Its first effort, as you mentioned, was half hearted. It wasn’t until Apple’s music revenues started *declining* that the company did something about it. The competencies / motivations clearly could not be aligned internally since it was *disruptive* to the existing downloads business. So Tim Cook acquired Beats, an external solution (DNA graft) to address a disruption.

    If you recall Christensen’s example of the 8-inch to 5.25-inch HDD transition—out of four companies, one company—Micropolis—managed to make the leap, but it required a “Herculean” effort.

    I would argue that Apple botching its first streaming effort then taking drastic measures to acquire Beats is exactly that—a Herculean effort to move the company from downloads to streaming. And it worked, as it did for Micropolis.

    Apple—like Micropolis—is the exception that proved the rule—streaming *is* disruptive.


    • charlesarthur

      “going by the most classic definition of disruption, streaming music absolutely qualifies as a disruptive technology vs. downloads.”

      Welllll… it’s frequently forgotten that streaming (music) was available before Apple made iTunes downloads a viable business in competition with CDs. There were services such as PressPlay and Real and Napster, and of course Pandora. But they completely failed to compete successfully with iTunes (and other download services such as Amazon) because the technological prerequisites weren’t in place. Network coverage was spotty. Music player software wouldn’t play nicely with both your PC and your music player (and that music player wasn’t an iPod, which had much of the market). Phones didn’t have the capacity, and see above about network coverage. There were often rumours that Apple would introduce streaming, but it didn’t.

      What streaming needed was a number of technology elements to come together, and Spotify’s very slow start (it was one of the last “desktop first” apps, in 2009) points to the important preconditions that had to be in place for streaming music services really to take off: powerful smartphones, good network coverage, sufficient phone storage. (This assumes licensing from labels, but they have done that for years.) Streaming couldn’t disrupt downloads for years because the surrounding preconditions didn’t exist; only a small number of people with exceptional coverage (Swedes) or willing to put up with some inconvenience (early subscribers) could benefit.

      It’s an interesting question whether it’s a disruptive innovation to the existing download business. Certainly it competes for spending, and functions in the same space. But it isn’t cheaper for the average music buyer (though much cheaper for the eager musophile) who would only buy 1 CD/album per month.

    • victor

      hi james,
      i would argue that both you and horace are right. disruptive is a relative term so A may be disruptive to B but sustaining to C. Streaming is clearly disruptive to music download (iTunes) but it is sustaining to the Hardware business (iPhone, iPad, etc.) allowing Apple to charge more on its devices.

    • Sacto_Joe

      Horace never said streaming wasn’t disrupting. He said Apple wasn’t disrupted. It’s similar to the way a larger form factor smartphone failed to disrupt Apple.

      Ironically, it may have been Apple’s much maligned purchase of Beats that innoculated Apple against disruption….

    • Disruptive to whom?

      • jwan584

        To Apple’s music downloads business.

      • Yes, but that’s not Apple. The company always ran the music business as a break-even business sustaining its core and the transition to streaming is more of the same.

      • jwan584

        Okay. In that case Apple Music should be made exclusive to iPhone.

  • Tim W.

    Referring to the Clayton Christensen article you linked to at the end, I think that whether you can be qualified as disruptive or not, also depends on how you define the “market” one competes in.

    Personally, I don’t agree with the assumption that Uber competes in the “taxi” market. Of course it does, and it’s winning handily because it offers a better product. But both also compete on a larger scale in the “transportation” market, or the “getting-people-from-point-A-to-point-B” market. Here, taxis and Uber compete on the high end while public transit competes on the low end.

    Compared to public transit, they have the unique advantage of offering a true door-to-door solution (= integrated), while they are also capable of joining the modular transportation game by solving the “first mile / last mile” problem of getting travelers from their doorstep to public transportation hail points.

    So while they certainly win from a convenience point of view, the only thing that’s keeping them from truly competing with car ownership (integrated) and public transportation (modular), is price. And I think Uber is uniquely incentivized to improve its operational excellence to drive prices down. Maybe autonomous cars will give them the solution, maybe “suggested pickup points” will work for them, maybe something else. But once they find out how to get travelers and commuters cheaply to their destination, public transit may be screwed.

    So my question is: why is there no such thing as “high-end disruption” described in Christensen’s model, where the better product simply wins by virtue of becoming cheaper over time?

    • Sacto_Joe

      If all things are equal, cheaper wins. That’s pretty obvious. It’s not at all obvious to me that, in the case of Uber vs it’s competition, all things are equal.

      That’s very obviously true with Apple; all smartphones are not equal, ergo commoditization hasn’t yet occurred.

      • The idea is not all things are equal, it’s that all things are good enough. At that point, better is not a driving force.

    • There can be no high end disruption because it’s not how disruption is defined. Disruption is defined as the process of customer defection to a product that has a lower performance measure than the prevailing solution. If you draw a trajectory diagram (P or performance vs. t, time) disruption is the process that occurs when solutions improve more quickly than the rate at which improvements are absorbed. It can be proven mathematically that customers have to jump down to a “worse” product when these conditions exist.

      • Rodrigo Fernandes Moreira

        Horace, when you say “proven mathematically”, are you talking about the research from Ron Adner?

      • No. I am referring to my research.

      • Rodrigo Fernandes Moreira

        That’s awesome! Can you point me to where it is?

  • fstein

    Excellent points and data to support. Thanks.
    Yes “disruption” is overused. Reminds me when people used “PC wars” until that was proven wrong by Apple’s relentless increase in Mac market share.
    Both ideas are related. We rarely see true disruptions, where the new virtually destroys the old. Nor do we decisive victories in the market. iOS and Android are not in a death match contest. What matters is that Apple provides compelling improvements to all it’s products (and to it’s ecosystem partners). The more shallow members of the press and blogosphere need stories. For them ‘incremental’ makes no sense.
    Apple’s reciprocal value with its ecosystem merits more attention. For example, top musical talent have used Apple’s customers and brand to stand out – both benefits. The App Store, growing at 50% p.a. has decades of growth for both App developers and Apple – so much for Apple not being a growth company.

    • Sacto_Joe

      These are important points. The Apple ecosystem (which I define as it’s complete stable of integrated computer-based products) is a differentiator, staving off commoditization. So are many other aspects of Apple (Apple stores with Genius Bars, highest quality products, cutting edge updates, and yes, Apple Music, to name a few), all of which make Apple products in general extremely “sticky”. The latest fixation is over Apple’s 2 year upgrade cycle supposedly deteriorating. Turns out it’s hogwash ( ) but even if it weren’t, so what? If an iPhone is held longer, it will still eventually be replaced with an iPhone, simply because Apple’s sticky-ness is still operative.

      It’s amazing to me that this incontrovertible evidence of recurring revenue can continue to be so thoroughly discounted by so-called analysts. The minute it suddenly becomes blindingly obvious (hopefully any day now), the price of AAPL should just explode, especially considering it’s present ridiculously low valuation.