Understanding Netflix's Personalization

Understanding how your Netflix home page is tailored to you
"There's no such thing as a 'Netflix show'. Our brand is personalization." –Ted Sarandos, Chief Content Officer

‍

Queen of the Streaming Platforms

With over 207.6 million subscribers and 37 million added in 2020 alone, Netflix is the undisputed leader in the consumer entertainment space. In a market saturated by powerful players, Netflix maintains its position using a unique mix of personalization, original content, and hosting popular shows with critical acclaim. The company claims that 80%+ of the TV shows watched on their platform are discovered through “some form of recommendation”, and the combination of their recommendation engine and personalization has saved them more than $1 billion per year.

In a recent intentional learning session, the Business and Technology Strategy team sought to understand Netflix’s personalization tactics and how they allow the streaming giant to dominate the marketplace.

‍

Strategies for Success

As of Q2 2020, Netflix claimed a greater share of streaming minutes than any other service: 34% of the US total. According to Netflix executives, this success can be attributed to three strategies:

  1. Original content

Netflix gets over 12 hours of viewing per $ of content spent, and in comparison to competitors the platform is 3x as efficient. Popular shows and critical acclaim

  1. Popular shows and critical acclaim

Netflix consistently has some of the most popular television shows, and claims accolades in both TV and film

  1. Personalization

To maximize play time and retention, Netflix is constantly adjusting their features and algorithms.

While each of the three strategies are important, Netflix’s emphasis on personalization remains outstanding.

Their personalization engine combines personalized rating predictions with personalized rankings to create a personalized home page prediction. The home page is populated not only by content but also by thumbnail image based on past behaviors and ratings. In the image below, two top 10 selections are shown, with entirely different cover images based on the viewers’ preferences.

‍

Two Netflix home pages, customized to the viewers' behaviors and preferences

‍

Netflix breaks down user data into explicit (what you tell them) and implicit (behavior-derived). Implicit includes:

  • When you pause, rewind, or fast forward
  • What day you watch content (i.e. week vs. weekend)
  • The date you watch
  • What time you watch content
  • Where you watch (zip code)
  • What device you use to watch (Tablet, TV, computer, mobile, etc.)
  • When you pause and leave content (and if you ever come back)
  • What you search (about 3 million per day)
  • Browsing and scrolling behavior
  • Thumbnail tracking: i.e. when credits start, and allegedly colors, volume, and other scenery content

In addition, Netflix has dozens of professional taggers who go through and tag content to help personalize.

A/B tests are constantly being run, with successes being incorporated into the production system.

Examples of tags that may be attributed to Netflix shows

A Netflix for Everyone

Joris Evers, Director of Global Communications of Netflix has said that: “There are 33 million different versions of Netflix.” The multitude of Netflix personalization options allow for viewers to choose their own adventure, and feel like their experience is bespoke to their varied interests and encourages them to return, time and time again. Next time you login to your Netflix account, you may notice how your own personalized version of the platform speaks to your interests and unique behaviors.

Curious? Send a message to  michael@intentionalfutures.com to learn more.

‍

Back to top