Netflix Announces Pivotal Change To Search, Recommendation Algorithms To Make Your Movie Choices Much Better

Following its expansion into 130 new countries in January 2016, Netflix has announced a pivotal change to the way its search and recommendation algorithms will now serve an increasingly global audience.

The shift, detailed in an extensive blog post, seeks to “to connect members worldwide with stories they’ll love,” according to Netflix — despite the fact that content licenses are region- and country-specific. Accordingly, Netflix’s previous recommendation methods were also region-specific in order to avoid the hiccups caused when a movie is available in one country and not another.

Going forward, however, the algorithm takes these regional disparities into account, but is also able to compute tastes and preferences across regions. If French users are fans of horror films, for instance, they will now see recommendations based on internationally-derived data rather than just local preferences.

The new algorithm was roughly one year in the making and was perfected by 70 engineers, according to The Verge, which noted that it has been live since Netflix announced its global expansion last month at CES.

While Netflix’s previous algorithm also took cultural differences into account — “we expect that Bollywood movies would have a different popularity in India than in Argentina,” the company explains — the updated approach now encompasses a more global view. Netflix describes it thusly: “If a member likes Sci-Fi movies, someone on the other side of the world who also likes Sci-Fi would be a better source for recommendations than their next-door neighbor who likes food documentaries.”

Netflix, which says it supports 21 total languages, is also working on how language barriers can impact recommendations and search. Because the company does not collect data on the languages its users speak, it runs the risk of suggesting films in foreign tongues, it says.

“Our global journey is just beginning,” said the company, which recently increased watch time by 20 percent simply by testing new title images throughout the platform. “We look forward to making our service dramatically better over time.”