Posted on Tuesday, April 17th, 2012 by Germain Lussier
Several years ago, Netflix ran a contest that would pay someone $1 million for improving their already solid recommendation engine. The algorithm, called Cinematch, took how you rated movies on Netflix’s five star system and gave you personalized recommendations based on those ratings. The contest winner only needed to improve the system by a small percentage and a team called BellKor Pragmatic Chaos took the prize. It was a win win for the company. They got positive press and new technology for a very small amount of money, relative to what it usually costs.
Years later, Netflix tried to do a similar contest that was shut down over privacy issues. Since then, the Netflix Prize more or less left the public consciousness. Did Netflix actually use the winning algorithm? The answer is “No” and the reason might surprise you.
Netflix ran a detailed blog on the situation which was picked up on several tech blogs such as Kottke and others. Head there for the long, detailed explanation but the gist is this: The movie rental game changed.
One of the reasons our focus in the recommendation algorithms has changed is because Netflix as a whole has changed dramatically in the last few years. Netflix launched an instant streaming service in 2007, one year after the Netflix Prize began. Streaming has not only changed the way our members interact with the service, but also the type of data available to use in our algorithms. For DVDs our goal is to help people fill their queue with titles to receive in the mail over the coming days and weeks; selection is distant in time from viewing, people select carefully because exchanging a DVD for another takes more than a day, and we get no feedback during viewing. For streaming members are looking for something great to watch right now; they can sample a few videos before settling on one, they can consume several in one session, and we can observe viewing statistics such as whether a video was watched fully or only partially.
Yup, just like with everything else, the fact that the Netflix business model became more Instant Watch focused made it less important to improve how they recommend DVDs. The extended blog also talks about the fact that Netflix moved primarily off internet to mobile devices and more, making the star system less important.
If you needed another pieces of evidence that their DVD service is eventually going away, now you have it.