Why The Defenders is Netflix's Biggest Behavioral Data Play Yet

Interana Insights

If you love Dave Chapelle’s comedy special on Netflix, you will likely enjoy Iron Fist, a Marvel comic book story about a rich white guy who is raised by monks to become their supernatural protector.

The recommendation makes complete sense, maybe not to an experienced TV executive, but to Netflix—who base their content decisions on hard evidence, generated from their viewer’s actual behavior and run through their algorithms.

House of Cards, the first flagship original program Netflix created, was purchased primarily on the strength of affinities between fans of:

  • David Fincher
  • The U.K. House of Cards
  • Kevin Spacey

Netflix Venn Diagram Source: UpX Academy

After watching thousands of politics junkies binge on movies like American Beauty and Fight Club, Netflix had enough data to suggest that a show bringing together all of these different elements would be a success. Now Netflix is going meta, actively experimenting with their own original content with The Defenders.

While The Defenders originated in a Marvel comic book series, the move to produce the show is a risk because the four original shows — Jessica Jones, Daredevil, Luke Cage and Iron Fist — have vastly different audiences.

Netflix is confident because they have the data to support their decision to put all four characters in one series, but the experiment goes much deeper than whether or not the show will succeed. Here’s why The Defenders is a unique data experiment that will drive content innovation even if the show fails.

Tweet: Why The Defenders Is Netflix's Biggest Behavioral Data Play Yet

The Defenders is an Experiment to Test Human Behavior

The Defenders is a show set in New York City based on the comic book series from Marvel. It features four famous superheroes who each like working alone. They soon realize that they can’t win fighting solo, which leads them to form The Defenders. Before releasing the show, Netflix featured each superhero in their own program:

  • Daredevil is Matt Murdock, a lawyer who was blinded in an accident that exposed him to a radioactive substance. As a result, his other senses became heightened giving him special abilities that he harnesses to fight crime.
  • Jessica Jones was also exposed to radioactive chemicals as a child when a military convoy hit her car, killing her entire family. She has super strength and other abilities, but she suffers from post-traumatic stress syndrome and is reluctant help others with her powers.
  • Luke Cage developed super strength and became bulletproof after a series of experiments he volunteered for while in prison. He lives in Harlem and uses his abilities to keep his neighborhood safe. Cage was featured in the Netflix program, Jessica Jones, having a failed romantic relationship with the female superhero.
  • Iron Fist is Danny Rand, who survived a plane crash at 10 years old and was rescued by monks and trained to be a warrior. When his chi is focused, his abilities are enhanced, and he possesses a glowing fist with super strength. Rand, an heir to a multi-billion dollar corporation, returns to New York to avenge his family’s death.

Netflix Defenders

Although Netflix doesn't release viewership numbers, they confirmed that only a small percentage of people watch all four of these Marvel superhero shows. However, many watch one or two. When you put all the characters together in one show and measure viewer response, it becomes an audience experiment. What happens when you bring four diverse audiences together?

Netflix will be able to see who watches the show and also track the journey each viewer takes to find the show. This valuable behavioral data will help them determine if there is a market for other shows that bring together popular characters for an ensemble series.

While they want The Defenders to be a success, the data they collect regardless of how many people view the show is just as important for future programming and recommendations. For example, if people who watched The Defenders originally discovered Daredevil because they first watched House of Cards, then Netflix can start connecting the dots to optimize other content.

Dexter vs Walter White Source: AgoraVox

More importantly, if the ensemble cast attracts a large viewership from the audiences of each show, it can lead to a new genre of programming that teams up popular characters into a new type of show. Let's be honest: it would have been interesting to see Dexter and Walter White taking down evil drug kingpins together.

How Behavioral Data Connects the Dots for Netflix

Behavioral analytics is at the center of Netflix's decision-making from recommending programs to deciding which shows to produce to actual creative direction like talent casting.

Netflix relies on this type of data because behaviors never lie. In the past, companies relied on surveys to get a feel for what customers wanted, but declarative data collection is often misleading as people have natural biases that unintentionally skew their answers. It is hard to answer a survey about creative content, let alone trust the results to run your business.

Tweet: Behavioral #analytics is at the center of @netflix decision-making

Behavioral data is different because it records user's actions, not what they think. As a result, Netflix knows exactly how each Marvel show connects to others because of their viewers' behaviors. They know that Jessica Jones and Daredevil have an audience overlap, but viewers of Luke Cage who never saw Jessica Jones or Daredevil tend to watch Iron Fist next.

Netflix did not expect these types of correlations. Luke Cage is also a character in Jessica Jones so one would assume the viewer would watch both programs, yet this was not the case.

Supershow For Everyone Source: CNET

Beyond connecting the Marvel shows, Netflix can track the other kinds of programming that fans of particular shows also like, which gives them deeper insight into recommending new content that is likely to stick.

  • If you're into the relatively light and optimistic Iron Fist, you'll probably enjoy coming-of-age tales like Shameless and Love.
  • If you're into the acerbic Jessica Jones, you'll likely enjoy the sharp humor in Orange is the New Black and Master of None.
  • If you're into the darker Daredevil, you'll probably enjoy morally ambiguous shows like Dexter and Breaking Bad.
  • If you're into the antihero Luke Cage, you'll likely enjoy the dangerous worlds of The Walking Dead and Black Mirror.

While some of these programming connections make sense, such as matching Dexter with Daredevil, both dark vigilantes, other show correlations can come as a complete shock.

For example, a show that led people to Iron Fist was Grace and Frankie, a comedy-drama series starring Jane Fonda and Lily Tomlin about two friends brought together after their husbands announce that they are in love and plan to get married.

In a million years, no one would have thought that Iron Fist would share a similar audience with a program starring two septuagenarians whose husbands fall in love. Yet, these are the types of unexpected insights that can only come from behavioral data and help you steer your business in a whole new direction based on your customers' actions.

Intuition is powerful and important. But often, data will trump it.

What Netflix is Trying to Learn with The Defenders

The central hypothesis Netflix is looking to explore with The Defenders is this:

  • If you take four shows with distinct viewerships, and combine them into one mega show, can you bring those disparate audiences together? If not all, then which viewers will engage, and why? And can you use that information to create better programming in the future?

Netflix is going beyond recommendations to pursue insights on how people's behaviors and affinities overlap across audiences. If most of the people who watched Luke Cage and Iron Fist are the ones watching The Defenders, then that tells you something about the potential crossover between the antihero-loving audience and the coming-of-age audience.

When you think about future programming, an audience crossover of Luke Cage (dangerous worlds) and Jessica Jones (sharp humor) would indicate something about the viability of more shows like Black Mirror that straddle that median.

Netflix wants to go beyond knowing which content people like, to why they like it. The decision to produce House of Cards relied on data to validate the decision, but Netflix is taking it a step further with The Defenders by deconstructing their shows based on audience behaviors to create something entirely new.

Netflix Data Model Source: Truly Deeply

It's not a simple problem to solve. Netflix is attempting to create a successful formula through hypothesis, trial-and-error, and experiment, the same way any business should. In the meantime, we are happy to binge-watch their experiments.

Tweet: @netflix wants to know not just what #content people like, but why they like it

But is Anyone Watching It?

Netflix's VP of Product Todd Yellin told Wired “what we're going to be learning as a company from both the algorithm side and the content side, what happens when you pair up shows that aren’t exactly the same thing,” and just “what happens when you put the mélange together.”

The initial results collected by the third party firm Jumpshot are not encouraging when it comes to wide adoption—currently, it appears The Defenders is Netflix's lowest-watched Marvel premiere (at least within its first thirty days.)

That might not matter. What Netflix needs to do is provide their subscribers, and fans of the Luke Cage, Jessica Jones, Iron Fist and Daredevil franchises, with content that creates the most happiness per dollar spent. The calculus is complicated here—it's not a flat measurement of how many people watched The Defenders. It's based on how engaged its viewers were, whether they rewatched it or renewed their subscription because of it.

John Ciancutti Answer

Whether or not The Defenders ends up being a success, Netflix will learn something valuable from its experiment on how to work with user preferences and negotiate the complex journeys viewers take from one piece of content to another.

Previous article Blog Summary Next article