Author’s Note:
I wrote this post in August; it concerns a couple of friends of mine, one who works for iTunes, one who works for Pandora. Immediately after posting it, I got a frantic text message from my Apple friend, saying something along the lines of “OMG REMOVE THIS POST IMMEDIATELY. IT CONTAINS CONFIDENTIAL INFORMATION.” I killed the post, having no real idea why I was doing so. It definitely seemed a little weird.
Then just a few weeks later, Apple introduced Genius, an enhancement to iTunes that essentially copies Pandora’s approach, sequencing songs based on their inherent characteristics. That’s when the reason for my friend’s freakout became clear: Although I hadn’t posted anything about the Genius feature (and in fact didn’t hear anything about it until it came out), my friend obviously thought I had heard something and posted about it. Apple’s security about new products and features is legendary; if my friend had been tied to the leak of information like that, he’d have lost his job. (For the record, he never tells me ANYTHING about new iTunes or iPod features, no matter how many cocktails I ply him with.)
So now that Genius has been released, I present the original post, both as a historical artifact, and because I continue to be very pleased with its title.
1.
I have a good friend who works for iTunes. Apparently you don’t say “the iTunes division of Apple,” simply iTunes. It’s a different wing of the company, I guess; he works in a separate building from One Infinite Loop, although apparently it’s closer to the gym, which he says is nice.
Hanging out with my friend, I’ve come to know a little bit about how iTunes’ marketing machine works. Basically, the primary unit of measure for marketing success is iTunes Music Store user account creation. They believe (and presumably research has shown) that creating an account makes a person much more likely to purchase music eventually, whether or not they purchase anything when first creating the account.
For this reason, the primary iTunes marketing efforts center around giving away free iTunes songs. If you’ve been to a large music festival of note this year, someone may have handed you a card offering you free iTunes songs from the artists participating in the festival— that’s what I’m talking about. So you take the card home, open iTunes, go to the Apple Store, enter the 12-or-so-digit code, and your download begins. But if you haven’t opened an iTMS account yet, well, you have to create one, right? So you do, and the mission is complete: the iTunes folks have their conversion.
There’s something about this marketing plan that feels off-kilter to me. For one thing, it fails to note one important fact of human life in 2008— that if you have an iPod, you use iTunes, and for practical purposes everyone in America has an iPod. At this point in tech history, it’s nowhere near as easy to download free music as it used to be, meaning that 9 times out of 10, if you want to hear “Don’t Stop Believin’”, you’re going to buy it. Probably from the iTMS.
So conversion is important, but I think you could just as easily proceed from the assumption that iTMS sales growth is tied to iPod sales growth, and that’s not a bad horse to be tied to.
iTunes’ real problem, it seems to me, is that it’s not connected to a Web site. (The iTunes Music Store isn’t a Web browser, even though it behaves like one in some respects.) You’d think that the #1 music buying site in the world would somehow protrude into the Internet— that there’d be an itunes.com where you could visit, preview new music, read and post reviews, etc. Something like Pitchfork meets last.fm, but with everything tied to iTunes. Reading and previewing music in a social Internet environment, it seems to me, would make it pretty damned easy to spend $.99 for a song— and that, I would argue, is the real conversion point. Once you get someone in the mindset where they’ll quickly buy a song they like, well, you’ve got ‘em. Right?
2.
I have another good friend who works for Pandora. If you don’t know about Pandora, you should— it’s the commercial outgrowth of something called the Music Genome Project: an attempt to categorize all the world’s music in terms of a finite number of variables. From Wikipedia:
A given song is represented by a vector containing approximately 150 genes. Each gene corresponds to a characteristic of the music, for example, gender of lead vocalist, level of distortion on the electric guitar, type of background vocals, etc. Rock and pop songs have 150 genes, rap songs have 350, and jazz songs have approximately 400. Other genres of music, such as world and classical, have 300-500 genes. The system depends on a sufficient number of genes to render useful results. Each gene is assigned a number between 1 and 5, and fractional values are allowed but are limited to half integers.
What this means to a Pandora user is that Pandora can take information about what you like— an artist, group, or even a single song— and turn it into a streaming, personalized radio station, with songs by bands you’ve never heard of that share qualities with stuff you like. It is amazing technology in practice— pick your favorite song, hit play, and the fun begins. As you go, you can rate things up or down, which lets Pandora understand your interests better. And did I mention that it’s free?
(The friend in question, Kevin Seal, also hosts a tremendously entertaining and educational podcast, The Musicology Show, as part of his Pandora duties.)
Experiencing Pandora for the first time is pretty mindblowing— it’s one of those things that seems too good to be true. And there’s the rub— it is. Information might want to be free on the Internet, but there are a million contrary forces to that basic longing— some of them totally baseless and pernicious, some of them perfectly valid. But in the ongoing battle between the recording industry and the Internet, it’s tough to bet against the old white guys with the deep pockets, and Pandora is, it seems to be, fighting a difficult and losing battle. Kevin has been pointing me to a Web site, SaveNetRadio.org, devoted to fighting the good fight; it is by no means a lost cause, but it looks like a tough road ahead for Web broadcasters, even for reasonably well-funded and popular services like Pandora. In an article about Pandora from the Chronicle of Higher Education, Pandora’s founder suggests they may be close to pulling the plug, as fees for Internet broadcasting rise and rise.
3.
When you mouse over a track that’s playing in your Pandora stream, you get some options:
The left and right buttons let you rate the track you’re hearing. The middle one opens a menu:
As you see, it is technically possible to buy the iTunes track from the Pandora interface. But it’s easy to see, even for a non-UI expert, that actually *buying* the track seems to be a pretty low priority. The Pandora-specific functions, like creating a new station, bookmarking, etc. are much more highly prioritized. And once you do mouse over the Buy… button, you get two options, iTunes and buying the CD on Amazon. But isn’t Pandora the product of a more highly evolved approach to music where the physical CD is almost completely irrelevant?
4.
Apple should buy, license, or exclusively partner with Pandora. That middle button ought to take you right into iTunes to buy the song. I couldn’t tell you how many songs I would have purchased on iTunes if that were the case. If you get into the habit of rating up, rating down, etc, then having that button there just becomes an extra-high rating; man, I like this song so much that I’ll pay the $.99 to play it anytime I want. Not much of a sacrifice— just about the easiest conversion ever, I’d say. And given that Pandora is now a free and immensely popular download for the iPhone, well, there you go.
I’m sure Pandora has lawyers writing briefs on this whole thing already, but it seems to me that making the conversion to a sale easier might take some heat off the “giving away free music” charge— the fact is, there’s not a HUGE difference between the one-time-only, unpredictable play of a song you get with Pandora and the 30-second preview of a song you can already get with iTunes. In fact, a good attorney could probably make a case for considering Pandora not as a “radio broadcaster” but rather as a straight-up social networking/viral marketing tool for the songs themselves. It has always seemed a weird characterization to think of Pandora as “radio,” even though that’s how they market it.
Now think about how the benefits could flow the other way. If iTunes were driven by the Music Genome Project, the ways in which iTunes could recommend music for me to listen to would become ten times as interesting. iTunes itself could incorporate Pandora-like functionality. Ratings within iTunes could feed, and be fed by, Pandora ratings. The world would become a better place, and two companies— one with an amazing tool and some great technology, the other with market share and a direct line to 75% of the world’s headphones— could share the love.
PS.
Although this is not a conventional love story, it’s interesting to note that the two friends I’ve mentioned became engaged— to different people— on almost exactly the same day. Congratulations to both of you!
