Marketing TechnologyData & AnalyticsHow Netflix uses big data to create content and enhance user experience

How Netflix uses big data to create content and enhance user experience

An estimated 80% of content streamed on Netflix is influenced by its recommendation system. Stats/examples how shows like House of Cards keep users engaged.

With a 51 percent market share of the American streaming industry and over 148 million streaming subscribers worldwide as of Q4 2018, Netflix is certainly a force to be reckoned with.

More interestingly, Netflix is on track to be profitable. The chart below, courtesy of Statista, shows Netflix’s annual revenue from 2002 to 2018, and one thing is clear: Netflix is growing consistently and exponentially.

Stats on Netflix’s annual revenue from 2002 to 2018

Unlike most other brands, Netflix’s growth is attributable more to content and user experience than to marketing, and this content is largely influenced by big data.

Big data is helping Netflix thrive despite counter-intuitive decisions

While many organizations have yet to effectively leverage data available to them, Netflix is a noteworthy exception.

Netflix is easily one of the most counter-intuitive companies out there. A huge example of Netflix’s counter-intuitive nature is shown through its decision to flat out block VPNs in 2016.

This is despite the fact that at the time, more than 30 million Netflix users lived in countries where Netflix’s service is unavailable without using a VPN or other location-masking services (and where Netflix is now recording most of its subscription gains).

The same year, Netflix hiked its prices and refused to back down despite protests from users and loss of hundreds of thousands of users.

Yet, Netflix has only grown since.

The following chart shows Netflix’s subscriber growth since it made its controversial decision to ban VPNs and hike its prices in 2016.

Stats on Netflix’s subscriber growth after 2016 prie hike

So how is Netflix able to continue rapid growth despite alienating a significant portion of its base? By leveraging big data to find out exactly what users want and giving it to them.

Netflix is betting big on content and user experience, the larger chunk of Netflix’s budget is spent on content. In 2019, Netflix is committing a $15 billion budget to content. For comparison, they are committing a meager $2.9 billion for marketing.

While it’s easy to focus on Netflix’s huge content budget, it would be a better idea to focus on the process used to come up with ideas for this content and how much of a role big data plays.

Netflix’s big data infrastructure

Netflix uses data processing software and traditional business intelligence tools such as Hadoop and Teradata, as well as its own open-source solutions such as Lipstick and Genie, to gather, store, and process massive amounts of information. These platforms influence its decisions on what content to create and promote to viewers.

Netflix doesn’t use a traditional data center-based Hadoop data warehouse. In order to allow it to store and process a rapidly increasing data set, it uses Amazon’s S3 to warehouse its data, allowing it to spin up multiple Hadoop clusters for different workloads accessing the same data. In the Hadoop ecosystem, it uses Hive for ad hoc queries and analytics and Pig for ETL (extract, transform, load), and algorithms.

It then created its own Genie project to help handle increasingly massive data volumes as it scales. All this points to one thing: Netflix is very particular about having a lot of data and being able to process this data to ensure it understands exactly what its users want.

The result has been nothing short of amazing. Netflix has been able to ensure a high engagement rate with its original content, such that 90 percent of Netflix users have engaged with its original content.

Netflix’s big data approach to content is so successful that, compared to the TV industry, where just 35 percent of shows are renewed past their first season, Netflix renews 93 percent of its original series.

House of Cards: A Netflix case study in big data

One of the most oft-cited examples of Netflix’s use of big data to conceive successful content is the House of Cards TV series. For good reasons.

Some quick facts:

  • When Netflix wanted to introduce the House of Cards show in 2013, unlike was the standard practice in the TV industry, Netflix didn’t launch a pilot. Instead, it commissioned two seasons of the show (for an estimated $100+ million), even before the first episode aired. A very big gamble for a show with no guarantee of succeeding, or so it was thought.
  • The House of Cards show was an instant hit, and six years later, despite the turmoil surrounding its star, Kevin Spacey, the program still boasts an 8.8 out of 10 rating from over 420,000 reviews on IMDB, putting it in the league of blockbusters like Avatar and The Sopranos.
  • According to Netflix, House of Cards was such a success that it was the most streamed piece of content in the United States and 40 additional countries at the height of its success.

While Netflix’s commitment to two seasons of House of Cards was a gamble to outsiders, insiders already knew that the show would succeed

In fact, Netflix’s confidence in the success of House of Cards was such that an executive told GIGAOM in an interview that they didn’t need to spend millions to get people to tune into the program. They just knew people would watch it.

Due to the direct relationship Netflix has with its subscribers, as well as an abundance of data on how audience members interact with their content, the company could easily determine what kind of content people wanted.

In the case of House of Cards, by analyzing its data, Netflix realized that a significant percentage of its 33 million subscribers at the time had streamed director David Fincher’s work, The Social Network, from beginning to end on its platform, and that films featuring Kevin Spacey were always successful with its audience.

What’s more, Netflix’s data revealed that the British version of House of Cards on its platform was a hit. And that those who had watched the British version of House of Cards had also watched other films acted by Kevin Spacey or directed by David Fincher.

Relying on this data, Netflix concluded that an already successful show in Britain, starring much-liked actor Kevin Spacey and director David Fincher, for an American audience, will be a big hit.

Netflix was right

Within three months of introducing House of Cards, Netflix added 2 million subscribers in the US and 1 million additional subscribers internationally.

This meant that an estimated $72 million was added to the company’s bottom line, nearly paying off its initial investment in the House of Cards show in mere months.

With a 93 percent renewal rate for its shows after the first season, the success of House of Cards isn’t an isolated incident. Other series like Orange Is The New Black, Arrested Development, and The Crown were introduced to acclaim using a similar process that relies on big data.

How Netflix uses data to enhance the user experience

When it comes to gathering data, Netflix’s huge user base of over 148 million subscribers gives it a massive advantage. It then focuses on the following metrics:

  • Date content was watched
  • The device on which the content was watched
  • How the nature of the content watched varied based on the device
  • Searches on its platform
  • Portions of content that got re-watched
  • Whether content was paused
  • User location data
  • Time of the day and week in which content was watched and how it influences the kind of content watched
  • Metadata from third parties like Nielsen
  • Social media data from Facebook and Twitter

Once data has been gathered, Netflix uses this data in a lot of ways. One of the most important uses is formulating and validating original programming ideas, as discussed in the above House of Cards example.

Arguably more significant is the manner in which Netflix has mastered effective use of data to get people to engage with its content.

Netflix is so good at targeted content promotion that an estimated 80 percent of content streamed on its platform is influenced by its recommendation system.

This recommendation system is designed in such a way that:

  • Netflix focuses on giving each user just what the user wants through a personalized content ranker that organizes each Netflix user’s collection based on personal information collected about the user. Like Netflix, you can use big data to ensure that content delivered to each user is influenced by the user’s personal activity and interaction with your brand, ensuring the content experience is unique for every user.
  • Netflix ranks top and trending content not only based on how popular the content is but also based on personal information available about the user. The content is promoted on the basis of the user’s Netflix activity. The key lesson here is that while people are interested in what is popular, they still want it to be influenced by their interests. When promoting “top content” to users, it is important to make sure it is relevant to their personal interest.
  • Recently viewed content is sorted based on an analysis of whether users are expected to continue watching or rewatching, or whether users stopped watching due to not finding the content interesting. This is key in ensuring that Netflix doesn’t bore its users; it can be tempting to want to keep promoting the same content since you’ve invested in it. If user activity indicates a lack of interest, it is better to relegate the content and offer something more interesting.
  • A content affinity algorithm recommends content similar to content a user just watched. It is important to note that people are more likely to want to consume content similar to the one they just consumed.

In conclusion

Without getting bored with the technicality, Netflix is clearly a great example of the power of big data. While you might not have the resources to create your own project for more big data efficiency like Netflix did by creating its Genie project, the big data industry is rapidly evolving and a lot of open source tools exist to help you collect and process the essential data to understand exactly what your users want.

By following Netflix’s example, it is possible to effectively leverage big data to enhance your content and user experience and ensure the growth of your business.

Gabrielle Sadeh is a Digital Marketing Consultant. She can be found on Twitter @GabrielleSadeh.

Whitepapers

US Mobile Streaming Behavior

Whitepaper | Mobile US Mobile Streaming Behavior

4m
Winning the Data Game: Digital Analytics Tactics for Media Groups

Whitepaper | Actionable Analysis Winning the Data Game: Digital Analytics Tactics for Media Groups

4m
Giving a Voice to Your Brand

Whitepaper | AI Giving a Voice to Your Brand

4m
Mobile Messaging Masters

Whitepaper | Mobile Mobile Messaging Masters

4m

Related Articles

Research: How the human body can predict what makes content go viral

Content Marketing Research: How the human body can predict what makes content go viral

2m Domenica D'Ottavio
Digital analytics for media: Four use cases for quality data

Data & Analytics Digital analytics for media: Four use cases for quality data

4m Emily Alford
How Best Buy, Nordstrom and Nike bring personalization to physical stores

Data & Analytics How Best Buy, Nordstrom and Nike bring personalization to physical stores

7m Mike O'Brien
Common mistakes to avoid when choosing an attribution model

Data & Analytics Common mistakes to avoid when choosing an attribution model

11m Tereza Litsa
Does your content generate ROI?

Data & Analytics Does your content generate ROI?

12m matthewmurray
Bad data breeds bad leads: Cleaning up data for the sake of the B2B sales cycle

Data & Analytics Bad data breeds bad leads: Cleaning up data for the sake of the B2B sales cycle

1y Mike O'Brien
Introduction to Customer Data Platforms

Data & Analytics Introduction to Customer Data Platforms

1y David Raab, CDPI
GDPR: How US businesses are preparing

Data & Analytics GDPR: How US businesses are preparing

2y Clark Boyd