Recommendation system at Netflix

Netflix was initially an American company in the business of DVD’s rental and was launched in 1997 and headquartered in California. Its initial core business was renting and mailing DVD’s to their clients. They launched their own website in the same year and developed several new rental models such as the online rental model and the monthly subscription model. At that moment, they had already developed video recommendation systems based on the experience of their customers that they cautiously recorded. For instance, they implemented systems for the clients to express concern about movies and to create movies’ ranking.

However, the market significantly changed over the last ten years with the development of the digital movies at the expense of the physical DVD’s. As a result the market was progressively moving away from the physical DVD’s to a more digital business based market. Netflix has subsequently responded to this trend and has changed dramatically over the last ten years. They introduced an online video on demand service in 2007, which has become the instance streaming service that we know nowadays. Now Netflix is an online provider of streaming media, especially focusing on movies and series. This change and redirection toward streaming service online has allowed Netflix to develop and streamline a very strong passive recommendation system. Indeed, with the launch of online services, it has become much easier for them to have access and to record more information about the consumers. They could then use more advanced algorithm with the integration of more information in order to provide more efficient and personalized recommendation and finally get closer to the consumers.

For instance, the welcome page of Netflix presents different set of movies presented in the forms of different row of movies, and is personalized for each visitor, based on the information owned on that visitor. Netflix attaches also great importance to the consumers’ awareness. This means that most of the time, Netflix provides information about why such recommendation is made to the visitor. This allows them to promote trust among their clients, but also to request some feedback about the relevance of the recommendation, which enable Netflix to constantly improve its recommendation system. Netflix recently introduced a Facebook connect feature, which allows them to make recommendation based on friends’ preferences as well. Finally, Netflix also focuses on similarities in the large sense, such as between movies or between customers. They also distinguish multiple dimensions in customers’ similarities, such as the movies type preferences, the time of the day of the viewing, the customers’ rating, etc. All these similarities constitute additional data to be used in their algorithm to improve their recommendation system.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s