Netflix Continues to Tweak Their Recommendation Algorithm
How has Netflix scored so well with their original content like House of Cards and Orange Is The New Black?
The shows are well made and entertaining. But data - and more importantly, canny data analysis - has also had a role in the way that these shows have found an eager and appreciative audience.
In the video above (posted Aug. 16th, 2013 to YouTube), Demitrios Kalogeropoulos talks about the role that improved algorithms have played in the recent visibility that Netflix has earned.
And, in a Dec. 12th, 2013 post to Mashable, Seth Fiergerman goes into even more detail about how Netflix has evolved their predictive software such that "75% to 80% of the videos that Netflix users end up watching on the service come directly from the company's recommendations about what to watch next."
Why does Netflix care so much about their ability to match users with content?
As Netflix improves their ability to predict content that will please unique customers, they reduce the amount of users who sample but then cancel their subscriptions after a month or two: "The better the suggestions Netflix can make, the more videos users will stream, and the more customers will want to continue paying for the service."
As Seth Fiergerman explains, to achieve optimal recommendations, Todd Yellin, VP of Netflix product innovation, and his team have moved away from the five star user-rating system that Netflix started with. Instead, Netflix is focusing on user behaviors as they traverse the site, tracking data from user interactions including "things like the "velocity" of how fast a user makes it through a video and whether or not they stalled out five or ten minutes into it. They [also] track whether the user is more likely to view an edgier sitcom, like It's Always Sunny In Philadelphia, late at night or watch comedies on a particular day of the week to better dole out recommendations."
Mr. Yellin's team even tracks how users interact with the list of suggested films: "Netflix has started to track how users scroll down the page and where they click to see which suggestions they ignore. "It's one thing to know what people play. It's another to know what they didn't play," Yellin says. "If we know what you saw in front of you, we can know how many times you saw that title." All the other data may suggest that the user should want to watch Skyfall, but if they repeatedly ignore it, Netflix will eventually stop suggesting it."
Randy Finch's Film Blog:
Thoughts from a film producer about making and distributing films.