How Netflix’s Recommendations System Works

Our business is a subscription service model that offers personalized recommendations, to help you find shows, movies, and games we think you might enjoy. To do this we have created an innovative recommendations system. This article provides a high level description of our recommendations system in plain language.

The basics

Whenever you access the Netflix service, our recommendations system strives to help you find a show, movie, or game to enjoy with ease. We estimate the likelihood that you will enjoy a particular title in our catalog based on a number of factors including:

• your interactions with our service (such as your viewing history and how you rated other titles),

• other members with similar tastes and preferences on our service, and

• information about the titles, such as their genre, categories, actors, release year, etc.

In addition to knowing what you have watched on Netflix, to best personalize the recommendations we also consider factors including:

• the time of day you're enjoying Netflix,

• the languages you prefer

• the devices you are enjoying Netflix on, and

• how long you enjoyed a Netflix title.

All of these signals are used as some of the inputs that we process in our algorithms. (An algorithm is a process or set of rules followed in a problem solving operation.) The recommendations system does not include demographic information (such as age or gender) as part of the decision making process.

If you’re not seeing something you want to watch, you can always search the entire catalog available to you. We try to make searching as easy and quick as possible. When you enter a search query, the top results we return are based on a number of factors, including the actions of other members who have entered the same or similar queries, our prediction of what you’d like to enjoy on Netflix, and others.

Below is a description of how the system works over time, and how these pieces of information influence what we present to you.

“Jump starting” the recommendations system

When you create your Netflix account, or add a new profile in your account, we ask you to choose a few titles that you like. We use these titles to “jump start” your recommendations. Choosing a few titles you like is optional. If you choose to forego this step then we will start you off with a diverse and popular set of titles to get you going.

Once you start engaging with titles on the service, this will “supersede” any initial preferences you provided us. As you continue to enjoy Netflix over time, the titles you engage with more recently will outweigh titles you engaged with in the past in terms of driving our recommendations system.

Rows, rankings and title representation

When you look at your Netflix homepage, our systems have ranked titles and positioned them in a way that is designed to present the best possible ordering of titles that you may enjoy.

On each page, there are multiple layers of personalization. For example, in a row, we may personalize the following:

• the choice of row (such as Continue Watching)

• which titles appear in the row, and

• the order of those titles.

The most recommended titles go to the top. The most strongly recommended titles start on the left of each row and go right -- unless you have selected Arabic or Hebrew as your language in our systems, in which case these will go right to left.

How we improve our recommendations system

We take feedback from every visit to the Netflix service (for example, what titles you start watching, if you finish those titles, and how you rate those titles, such as thumbs up) and continually update our algorithms with those signals to improve the accuracy of their prediction of what you’re most likely to watch. Our data, algorithms, and computation systems continue to feed into each other to produce fresh recommendations so that your Netflix experience remains relevant and helpful.