netflix

Date that article was published: Feb 27, 2019

Article Summary

There are concrete reasons and ways that video-streaming giant, Netflix, needs data-driven processes (i.e. data science, machine learning, AI) to function and continue to grow its loyal consumer base. The bulk of their business problems are reliant on data in order to be solved, with some examples below.

Use Cases Business Problem Solved
1. Personalization of Movie Recommendations Uses watch history of users with similar tastes to keep users intrigued by new recommended shows/movies
2. Auto-Generation and Personalization of Thumbnails / Artwork Calculates likeliness of users to click on a movie based on thumbnail based on whether similar users have clicked or not
3. Location Scouting for Movie Production (Pre-Production) Using data to help decide on where and when best to shoot a movie set
4. Movie Editing (Post-Production Using historical data of when quality control checks have failed in the past to predict when a manual check is most beneficial
5. Streaming Quality Using past viewing data to predict bandwidth usage to help Netflix decide when to cache regional servers for faster load times during peak (expected) demand

What Data Does Netflix Use?

…and how are you targeted?

We can consider the second use case of “Auto-Generation and Personalization of Thumbnails” as an example. Besides, you may be wondering how AI works its magic to produce click-baity thumbnails. In order to truly understand this, first consider these questions:

  • What data does Netflix use to create these personalized thumbnails / artwork?
  • What data does Netflix use target these custom-created thumbnails to the appropriate individual?

For the first question, consider these facts:

  • A 1 hour episode of Stranger Things has >86,000 static video frames
  • These video frames can each individually be assigned certain attributes that are later used to filter down to the best thumbnail candidates through a set of tools and algorithms called Aesthetic Visual Analysis (AVA). - This is designed to find the best custom thumbnail image out of every static frame of the video
  • Netflix Annotation — Netflix creates meta data for each frame including brightness (.67), # of faces (3) , skin tones (.2), probability of nudity (.03), level of motion blur (4), symmetry (.4)
  • Netflix Image Ranking — Netflix uses the meta data from above to pick out specific images that are highest quality (good lighting, no motion blur, probably contains some face shot of major characters from a decent angle, don’t contain unauthorized branded content, etc) and most clickable

Consider these facts as well for the second question:

  • Number of movies watched, number of minutes of each show watched
  • % of completion for every video/series
  • Number of upvotes, which movies were favorited, etc
  • % of overall watch content that is attributable any specific show (and therefore level of affinity that user has to a specific show or related cast members)
  • Any seasonal or weekly trends related to a user’s level of engagement, etc.

Most Useful Takeaways

This article heavily emphasizes how part of Netflix’s success in this new world of AI and business is that they are able to identify their business problems first before trying to come up with a cool AI solution to grow the business. Incorporating AI is extremely useful and probably vital to entertainment platforms across all industries, but it is not just about adding more AI elements, it’s about identifying important aspects of the company that can grow and figure out where AI fits into that. Give this article a read if you want to get a more in-depth read on how Netflix specifically uses their data and relies on AI to do a lot of the hard work that human error would definitely mess up.

Why Is This Important?

As the leading video-streaming platform, Netflix sets a prime example for all others. It is clear that the company with the best ML and AI practices who can effectively connect use cases to business problems will generate the most money. This is just one example of how powerful knowing how to collect and manage data, while also knowing how to apply that newfound knowledge, is becoming increasingly relevant in our society today. There will also be questions of ethics that come to light with data usage, especially with Netflix. We need to acknowledge where the world is headed and be on top of knowing how these big companies are using their data.