Apple held a somewhat underwhelming press event on Tuesday, September 9th, but while the deafening buzz Apple’s unveilings typically generate made this one seem a little dull by comparison, I found it quite interesting. The beautiful (and very colorful) new iPod nano wasn’t what made me sit up and take notice, though. What caught my attention was a feature in the new iTunes 8 called Genius.
Genius is, in a nutshell, a music-recommendation feature that works with the songs in your own library. It does basically two things for the user: it can suggest songs similar to the one being played that the listener might like to buy from the iTunes Store, and it can instantly sift through the listener’s library to generate a playlist of songs that are musically similar to the one currently playing. The former functionality is a transparently good idea to inspire more purchases (tailoring suggestions to what the listener is demonstrably in the mood for at any given moment makes a lot of sense), while the latter has already come in handy for me as it has shown me songs from my cavernous music library that I was in the mood to hear but had forgotten about.
So why is this interesting? Genius takes advantage of the wisdom of large numbers of people to recommend music in a way that makes Apple’s job easier and makes the service more accurate. As Steve Jobs (vaguely) explained in his speech, Genius will initially recommend music based on a proprietary, Apple-designed algorithm, but as more and more users turn on Genius it will (anonymously) gather data about users’ listening and playlist-management habits in order to “get smarter” (i.e., refine recommendations and more accurately determine which songs share qualities).
Pandora, an Internet music-streaming service that plays songs that share qualities with songs or artists you like, bases its recommendations on the mammoth Music Genome Project, which requires very smart people to spend up to half an hour per song creating a database of musical “genes” or shared qualities. But why spend all that time and effort (and money) when the preferences of the people you actually care about – end users – can easily be aggregated to produce recommendations that may even be more accurate?
Finding ways to take advantage of the information waiting to be gathered from large numbers of people is advantageous in many areas. Amazon.com, which disrupted brick-and-mortar retailers through an online offering with a limited ability to interact with customers, doesn’t need to develop a sophisticated recommendation system for determining which of its products go well together; it can simply track purchasing habits and tell you what other people combined with the purchase you just made.
The Dash Express, a potentially disruptive GPS navigation device (see here), doesn’t use the hard-to-gather and often inaccurate traffic information provided by the complex variety of traffic monitoring services; instead, it simply aggregates the positions and speeds of its users to come to more accurate, real-time conclusions about traffic conditions. Amazon and Dash are particularly interesting in that they have utilized this kind of information to strengthen their highly disruptive offerings by making them much better than competitors’ products along the dimensions that matter most to their customers (i.e., quality of product recommendations and quality and quantity of real-time traffic data).
Genius thus joins a long list of systems that leverage the “wisdom of crowds” to create improved products and services. The system may not make iTunes a more disruptive product (it adds features without any trade-offs), but it has the potential to be a powerful sustaining move. More broadly, seeking out and using crowds’ wisdom is easier than it has ever been, and many more new ways of taking advantage of it are undoubtedly yet to be discovered.