Memo To Apple: A Map To Pandora's Jugular

| About: Apple Inc. (AAPL)


I am disgusted with myself for not seeing what was right in front of me, and I am disgusted with iTunes Radio for the same thing. Since September, I have been short Pandora (NYSE:P), believing that iTunes Radio would be a powerful competitor, powerful enough to knock the luster off P's Tech-bubble 2.0 valuation. Essentially, I was betting that Apple would continue its legacy of creating products that offer a superior user experience. Instead, what they released was a half-baked add-on whose only purpose seems to be demonstrating that Apple (NASDAQ:AAPL) has forgotten its roots. Unlike other Apple products, iTunes Radio seems to be based on the thesis that people will value features over depth of experience. This is a shocking and disturbing outcome for Apple shareholders, since the entire Apple Empire has been built on delivering products that are the antitheses of such geeky reasoning. The good news is that iTunes Radio is highly fixable. Here's how to do it.

Forget the Features! (Know Your Customer)

Apple has put a lot of emphasis on the breadth of its catalog, and the higher sound quality on offer. This is a mistake. While both of these things are worthwhile, they are only marginally additive to the average user experience, which is the only thing that really counts.

First the catalog, let's face it; most people listen to the hits. Yes you will find one in a hundred people who want to hear 1950s French pop music but most of us don't want to hear much more than we would on traditional radio. Ask any musician, the Pareto Principal is in full effect when it comes to the average person's musical tastes. That is to say, 80% of the people only care about 20% of the music, and the real numbers are probably much more like 95% to 5%. This is simply not a differentiating factor for the average streaming radio user, end of story.

Sound quality is the other oft-touted feature that offers only the most marginal improvement to the average user's experience. Think about yourself in the car, how bad does the sound quality have to get before you change the station? For me, if it's a song I like, things have to get pretty awful before I touch the dial. The idea that users are going to flock to iTunes for higher sound quality is ludicrous. Look at how many people use the crappy ear-bud headphones Apple gives out. If quality was really a motivating factor, everyone would be walking around with Sennheiser headphones - they're not.

Fix the Algorithm! (Quality of Selections)

This is a difficult claim to make. But I'm gonna do it anyway. The iTunes Radio song selection algorithm is broken, and needs to be fixed. Reasonable minds will inquire; do you have any data to support that conclusion? And the answer is no. I have only my own experience. Nevertheless, I feel confident in asserting that the iTunes algorithm is at present, far inferior to the one Pandora employs. I've reached this conclusion after trying ten of my favorite artists on both stations and comparing the musical selections each platform played. The result; I was blown away at how bad iTunes Radio performed at selecting relevant songs. That said, I encourage readers to run a similar experiment for themselves, and come to their own conclusions. Don't take my word for it - go try it out.

The bright side of this, (if you can call it that) is that the effect of the inferior algorithm on the user experience is so obvious, I honestly cannot imagine that it's not being worked on as we speak. In addition to the tweaking of the broader algorithm, we can reasonably expect improvements in iTunes' performance as data from user input is incorporated into the system. Pandora for instance benefits from years of people plugging away at their thumbs up / down icons, helping to fine-tune what gets presented to users and the algorithm itself. As Apple accumulates more feedback from its users it should be able to improve its algorithm to the point where it is competitive. But it's vital that this happens ASAP. A critical point is that the algorithm doesn't have to be as good as Pandora's but it has to be better than it is or things will never work out.

A second order result of poor song selection is that on iTunes Radio you quickly run up against your station song skip limit. This is detrimental to new user conversion, hitting the skip limit on multiple stations makes people much more likely to abandon the platform out of sheer frustration.

With regard to the skip limit, it's worth pointing out that Pandora (pictured above) at least gives you a nice, short explanation of why you can't keep skipping, and how to listen to more music. iTunes Radio on the other hand provides only the somewhat terse message: "You can skip six songs per hour on each station." Being nice and helpful still counts for something in this world, not to mention it's cheap. Apple, take note.

As my girlfriend was kind enough to point out, one good thing about Apple's message is that it tells you how many how many songs you are allowed to skip per station. Pandora's by contrast is ambiguous and can make users nervous about using the skip button.

While we're on the subject of niceties, another nice thing about Pandora is that they provide suggestions of similar artists you might like to make a station for (pictured above). Something akin to this would be especially helpful for iTunes Radio because it would minimize user loss due to the frustration that occurs when you run up against the skip limit. Put another way, I'm already mad when I run out of skips, I don't want to have to think about what else I might want to hear. Anyone familiar with the paradox of choice will immediately recognize the value of this or any feature that minimizes the user's cognitive workload.

Fix The Feedback Interface! (Ease of Feedback)

"What I propose is a road map to modify our man-made systems to let the simple - and natural - take their course. But simplicity is not so simple to attain. Steve Jobs figured out that 'you have to work hard to get your thinking clean to make it simple.' The Arabs have an expression for trenchant prose: no skill to understand it, mastery to write it." - Nassim Taleb, Antifragile

User feedback is not only an important part of tuning the song selection algorithm, it's an integral part of the user experience as well. And so the fact that iTunes Radio serves up an inferior interface puts it at a disadvantage on both fronts. I call it inferior because it's simply more difficult to use. It takes two clicks to perform any preference action aside from skipping the song. First you have click this little star button (pictured above), and then choose one of three menu options (pictured below). Pandora by contrast offers every single preference option just one click away. It's hard to express how this little difference can be such a big deal. Steve Jobs would get it. The only other way I can think to put it is that people are really, really lazy. I want to feel like I'm in my favorite armchair listening to the radio, not organizing my iTunes library. Easy - get it?

The other problem is that the feedback options aren't sufficiently granular. The positive feedback options are okay. You can choose between, "Play More Like This" and "Add to iTunes Wish List." On the negative side however, your only choices are, "Never Play This Song," and to skip. This dichotomy comes to feel like a nuisance, especially when you run out of skips. Additionally, the "Never Play This Song" button is ambiguous. Are we never playing this song again on just this station, or on all stations? I suspect it's the former but the ambiguity will definitely make many users hesitant to use the button at all. The fix for this is that every preference action be only one click away, and non-committal / non-threatening.

This failure, yes the failure of the iTunes Radio user interface is surprising for two reasons. One, this is what Apple is supposed to be good at. Two, they have Pandora to copy! And therein lies the solution. I don't know whether the thumbs up / down interface is patented, but it shouldn't take hiring Dan Brown to come up with some alternative symbols. This should have been a lay-up and they blew it.

Make it an App! (Accessibility)

iTunes Radio isn't easy to find, and thus far maybe that's a good thing. In fact, I would wager that half of Apple users have never opened the platform, simply because they don't know where it is. More importantly, even after you figure out where it is, it's annoying to have to go into iTunes to get it. This could be easily remedied if Apple would make a separate App for the radio. Even if that app were just a shortcut to bring you into the radio part of the Music App, it would go a long way to making the product user friendly. Make it pre-installed, make it easy to see. You'll be amazed how many new users you get.


Tim Cook should have been draping the bloody carcass of Pandora across his shoulders at the last Apple event, as proof that Apple is still primus inter pares. Instead, iTunes Radio is languishing in the depths of OS 7 - a blemish on an otherwise unparalleled legacy of delivering products that recognize the primacy of user experience above everything else.

Before I wrap up, let me be clear about something; Pandora is overvalued. But not because it doesn't make any money, it's overvalued because intelligent investors wouldn't pay $5 billion plus for a startup with a replicable product that's surrounded by competitors looking to take a bite out of it. Not the least of which (Apple) also happens to be the distribution platform for 1/3rd of its users. Pandora has been granted a stay of execution thanks to Apple's shoddy implementation of what has the potential to be a far superior product. However, as I have demonstrated, the road to Pandora's ruin is not complicated. Apple could implement most of the changes in a weekend. The only question is whether the giant in Cupertino has enough of its former mojo to strike the killing blow sooner than later.

Disclosure: I am short P. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: No position in AAPL.