“You probably recall all the excitement that went around when a group finally won the big Netflix $1 million prize in 2009, improving Netflix’s recommendation algorithm by 10%. But what you might not know, is that Netflix never implemented that solution itself.”
The long reason why they never implemented the winning solution is pretty interesting. For those that don’t want to read long explanations, the long and short of it is that once Netflix shifted to a primarily streaming service, the recommendation mechanism also shifted.
It turns out that the recommendation engine for streaming videos is different than for rental viewing because the algorithm needed is a matter of “watch now” rather than “watch in a few days.” For streaming customers Netflix essentially needs a 1:1 recommendation:success ratio. That’s no easy feat.