Columnists

CES: Analyzing the iTunes usage statistics

Front-and-center in the debate over to what degree the Apple juggernaut is sustainable in the future is the question of how many FairPlay DRM-soaked tracks the average iPod user buys.

By Brian Dipert, Senior Technical Editor -- EDN, 1/10/2007 12:04:00 PM

Macworld Expo and the Consumer Electronics Show, both in progress right now and dueling for mindshare, are aptly symbolic of the heated battle also underway—for consumers' ears, eyes, and wallets—between Apple's iPod product line and....well....everyone else that participates in the portable audio and A/V player space.

Companies such as iRiver, SanDisk, and (at least seemingly) several hundred Chinese, Taiwanese, and other Asian vendors here at CES (but not, notably, Creative Labs) are rolling out new player hardware.

Rhapsody is attempting to move that hardware and us, its users, away from Microsoft's future-in-some-doubt PlaysForSure DRM scheme and toward Real's own proprietary content-protection approach. During Bill Gates' keynote Sunday night, Microsoft president Robbie Bach attempted to straddle both sides of the DRM debate. He simultaneously touted the PlaysForSure-incompatible next-generation Zune and reassured PlaysForSure backers by noting that the company is still committed to the year-old URGE music service "and others like it from the 350 partners who are helping us deliver a platform for music around the world."

Opposing them all is Apple's latest iPod masterpiece. Feature-rich and eye-catching—but also eyebrow-raising given its significant price tag—the iPhone won't be available until June. That fact that has some folks in the blogosphere wondering if Apple has just Osborned its iPod sales. I'm not one of them; I don't see much overlap (though there is some, particularly when you consider the iPhone's large, video-viewing LCD). At 4 and 8 Gbytes of capacity, the iPhone's limited capacity is dwarfed by 30- and 80-Gbyte iPods, while the iPhone's $499-599 ASP dwarfs the prices of 4- and 8-Gbyte iPod nanos.

Front-and-center in the debate over to what degree the Apple juggernaut is sustainable in the future is the question of how many FairPlay DRM-soaked tracks the average iPod user buys. The AAC audio format that iTunes defaults to when ripping CDs (along with, for that matter, the MP3 format that's the other ripping option) is an industry-standard codec that any company can support after paying requisite fees to technology rights-holders. Microsoft's support of AAC in the Xbox 360 and Zune exemplifies that reality.

FairPlay, however, is a completely different matter. Apple continues to hold its DRM scheme close to the chest, with very limited third-party licensing to date. Therefore, the bigger the investment a consumer has made in iTunes Store purchases, the less motivated that consumer will be to re-purchase that same content for use on a non-FairPlay-supporting player. This argument formed the foundation of my mid-2005 Apple-versus-the-world cover story.

iTunes music sales are so critical to Apple's long-term iPod success that Steve Jobs addressed them at the very beginning of yesterday's keynote speech. Refuting mid-December analyst claims that iTunes sales were collapsing, Jobs commented, "What we see is iTunes sales were really up this year. We doubled the number of songs we sold in 2006. We are selling over 5 million songs a day. Isn't that unbelievable? 58 songs every second." Jobs' comments were the latest in a long line of Apple iTunes reports which, for reasons I think you'll soon understand, focused exclusively on absolute sales numbers. What Apple conveniently ignores is the relative aspect of sales statistics, specifically the degree to which increases in iTunes track purchases pace (or don't pace) increases in iPod purchases.

What's the iTunes purchase pattern of a typical iPod owner? That's the all-important-question that, in the absence of hard numbers from Apple, a recently published series of blog posts from Microsoft Zune team member David Caulton attempts to address. You might be tempted to knee-jerk dismiss Caulton's thoughts given his employer, but I'd encourage you to give the posts that follow a perusal before you form an opinion. I've personally found Caulton to be a notable straight-shooter in my past conversations with him. If you read his latest post from the Macworld Expo, where he's doing competitive analysis, I think you'll see what I mean.

Check out these chronologically-ordered writeups from Caulton:

• Virgin Digital Music Shuts down gracefully
• Music sales in the clear?
• Subscription service finance 101
• I'm involved peripherally in a dust-up....
• One comment made me realize I should....
• Update to songs per month per ipod
• Trends in digital downloads
• I'm cramming to get packed for my trip....
• Boiling it down - songs per ipod
• Macworld Keynote numbers

The key conclusion of Caulton's number-crunching analysis is his belief that iPod owners "initially buy 30 tracks in the first three months, but then drop off to one track per month or less on average." Two key variables, if Caulton's guessed wrong on them, would affect the result, but I personally think he's got his assumptions dialed in pretty close.

If he's underestimated the "replacement market" (the average number of iPods that each owner has purchased), he'll therefore underestimate the number of iTunes tracks each owner has also purchased. I won't tell you how many iPods are under my roof, but I'm also an atypical tech enthusiast (and, of course, I justify owning them all for EDN research purposes…ahem…).

His analysis does not assume the existence of a statistically significant number of iTunes customers who don't own iPods (folks who just use the iTunes Store as a replacement for traditional bricks-and-mortar purchases, burning the downloaded tracks to CD but never transferring them to an iPod).

Give his data and his analysis methodology a close perusal and let me know what you think.

While I have your attention, I'd also like to toss a few other long-brewing thoughts on digital music your way. A month-plus back, I pointed out some interesting data on ISP bandwidth trends, as they related to digital distribution of multimedia content, from an AES Convention session I attended back in early October. What I didn't share then, but will now, are the comments offered by the other Microsoft panelist, Elliot Omiya. Omiya is not from the Zune organization, I'll point out up front, but from the development team of the Windows Vista audio software stack.

Had Omiya been a Zune team member, he frankly might not have a job now (although, come to think of it, his boss seems to feel the same way), because he delivered a quite scathing indictment of the "state of the digital music business." And ironically, his critique predated Universal's successful $1-per-Zune extortion "negotiation" with Microsoft. A few of the many verbal grenades that Omiya lobbed at the audience:

• The music download business can't survive without a multitier strategy that incorporates numerous music-sourcing options.
• Apple makes no money on downloads, only on hardware, on purpose. This is both so no one else can easily compete, and because if Apple tried to earn a profit on its iTunes Store sales, the all-powerful music labels would simply raise prices and transfer that incremental profit to themselves, a phenomenon he referred to as "transfer pricing power."
• The current distribution model leaves profit in too few places to allow audio to thrive.
• The only realistic way to fix this unhealthy consolidation of power is to create a new distribution channel, i.e. to reduce the amount of content controlled by today's few, large labels.
• Alteration of the current distribution structure will require some sort of "catalytic event," although Omiya doesn't know what it will be.

In the Q&A that followed the panel presentations, I suggested to Omiya that the "cataclysmic event" might be MySpace (for any of you wondering why on earth Rupert Murdoch's News Corp, paid $580 million for it, keep reading) as an excellent article from the November 2005 issue of Wired also proposes. He agreed that community-cultivated buzz about new artists had the potential to displace at least some of the corporation-generated hype that's currently pervasive, thereby bypassing the music labels and putting money directly in artists' pockets (along with some in Murdoch's). However, he pointed out that this alternative channel for promotion and distribution, along with others like it, is currently too immature to draw definitive conclusions about its impact. In Omiya's words, "monetization hasn't been figured out yet."

Finally, I'd like to say a few words about Allofmp3, a Russia-based digital-music distribution service that I've mentioned on several past occasions. The RIAA and its member companies have long viewed Allofmp3 and its questionable legal status as an annoying thorn in their side. They've recently succeeded in engaging the US government as a partner to their claims, and they just filed a roughly $1.65 trillion lawsuit. I'm not going to comment on the legal aspects of the Allofmp3-vs-RIAA lawsuit; I'm no lawyer. But what I do find interesting about Allofmp3 is its service model, which I'd argue has directly translated to its mind-boggling business success.

Some analyses peg Allofmp3 as the second-largest digital-music service worldwide, behind the iTunes Store. One might wonder at this point just how much larger Allofmp3, or a service like it, could be with the "Is it legal?" specter lifted. Allofmp3 delivers music to its customers in a diversity of formats, both lossless and lossy, the latter also in both constant- and variable-bit-rate encoding schemes and at a variety of bit rates. DRM is nonexistent. Downloads are speedy, and payments are easy (or at least they used to be). And Allofmp3 prices are cheap....roughly 10 cents per track for the most common format/bit-rate options.

Here's my contention: The record labels are currently charging about as much for a digitally downloaded album as for a CD purchased at a bricks-and-mortar store. The prior sentence's statement is also only true when the download is bought as a complete album; the digitally downloaded album is much more expensive when acquired as a collection of tracks. This is because the labels don't want to encourage the single-track cherry-picking behavior that digital distribution enables, as that would substantially and negatively impact their business model if it became pervasive at the lower per-track prices.

The parity between CD and digital music sales exists in spite of the fact that with digital music delivery, there is no physical disc to manufacture, no physical inventory cost (neither at the manufacturer nor the end retailer), and no physical transportation cost from manufacturer to retailer via intermediary stops. Instead, the labels and their distribution partners like the iTunes Store are paying for much cheaper hard-disk space at the server, along with low-cost (and getting ever-lower) bandwidth to the purchaser.

Consumers aren't stupid; they know they're getting ripped off, and they respond with piracy—piracy that today's high track and album prices subsidize, just as those folks with health-care insurance subsidize the medical expenses of those who don't have insurance and don't, for a variety of reasons, end up paying their bills out-of-pocket instead.

At 10 cents or so per track, on the other hand, I'd claim there's near-zero incentive for most folks to violate copyright, either by downloading tracks over P2P and/or by freely offering tracks to others (aside from the lunatic fringe who'll try to "stick it to the Man" regardless of how consumer-friendly the financial and DRM aspects of the legally alternative scheme are).

I've said it before and I'll say it again; I'm not fundamentally opposed to DRM, as long as it preserves reasonable consumer Fair Use, and particularly if the hardware and software suppliers can huddle and squash its numerous bugs. C'mon, record labels, cut your prices. I'm betting your revenue and profits will grow, not shrink, in response. And anyway, if you don't, MySpace and other community sites (or something else I haven't even thought of yet) is sooner or later going to eat you for lunch.