Every Saturday we present you a company and its way of using data. Today, a company known to all: Netflix.

‍

Presentation of Netflix

‍

Netflix presents itself as a subscription streaming service that allows its users to watch movies and series without advertising with a device connected to the Internet.

This company was created in 1997 and started as a DVD rental company. It was ten years later, in 2007, that it began offering its streaming service. Since 2020, Netflix even owns movie theaters. House of Cards is the first series produced by Netflix

Today in France, the platform broadcasts more than 5000 different titles between movies, series, documentaries. It is even starting to produce its own games, available on its mobile application only for the moment. Netflix now has more than 200 million subscribers worldwide.

‍

Data usage at Netflix

‍

Its recommendation engine

Recommending the right movies and series to the right people is part of the mission that the Netflix company has given itself. The Netflix Prize in 2009 is an illustration of this. It was a competition asking participants to propose the best collaborative filtering algorithm - a recommendation algorithm. The winner then won $1,000,000 and allowed Netflix to improve the accuracy of its predictions by about 10%. 

To do this, Netflix collects a lot of data on users: number of views per movie and series, pause and resume times - a user who doesn't pause during his movie seems to be much more interested than a user who stops several times, whatever the reason. The company will also seek to collect the passages of movies and series watched multiple times.

Today, Netflix goes as far as optimizing the poster it presents to you among those it has. Compare with your friends, you will see that you don't all have the same posters. If you have the same tastes as your friends, chances are they are very similar because you are in the same "taste group".

Data and user experience

Netflix uses data to improve its users' experience. The platform perfectly remembers where you left off, which episode and where within the episode. This seems like a small detail, but not all of its competitors are able to do this. Notably, if you are in the end credits of an episode and you stop there, then when you restart to resume your series, Netflix will not leave you on the remaining 2 minutes of credits but on the next episode. 

Another detail: if you watch your episodes on the channel, Netflix will skip the summary part of the previous episodes but displays it if you are at the start of the application then you will have the summary of the previous episodes (if it exists in this series). Also, if you want to skip the credits, Netflix will cut perfectly at the right place and you can then resume the series in the following. 

These many details made possible by data allow Netflix to be fluid to be used and makes the platform enjoyable. 

Creation of series 

The data collected by Netflix allows it to understand what kind of content its consumers want. Netflix will therefore be able to calibrate its series to appeal to the greatest number of people. 

The result is that in the United States, 25% of the series produced will have a second season. For Netflix, this statistic is 80%. So Netflix knows how to control its market better than anyone else, thanks to its data. 

The data will be used to plan the shooting: best locations, time and day of shooting according to the availability of each of the stakeholders needed to produce the film or series. The data will allow planning and therefore reduce production costs.