"Whenever I go into a restaurant, I order both a chicken and an egg to see which comes first"

Saturday, February 2, 2013

House Of Cards–Netflix, Big Data, And Creativity

Netflix premiered its own movie February 1st – House of Cards, a political drama series    directed by David Fincher, starring Kevin Spacey and based on a successful 1990s BBC mini-series. There would be nothing particularly new in this, except that the decisions to make the film were based on the mining of big data.  In other words, based on the billions of bits of information collected from its millions of consumers, Netflix knew exactly how to produce a blockbuster hit.  In an article for Salon (2.1.13) Andrew Leonard writes:
For at least a year, Netflix has been explicit about its plans to exploit its Big Data capabilities to influence its programming choices. “House of Cards” is one of the first major test cases of this Big Data-driven creative strategy. For almost a year, Netflix executives have told us that their detailed knowledge of Netflix subscriber viewing preferences clinched their decision to license a remake of the popular and critically well regarded 1990 BBC miniseries. Netflix’s data indicated that the same subscribers who loved the original BBC production also gobbled down movies starring Kevin Spacey or directed by David Fincher. Therefore, concluded Netflix executives, a remake of the BBC drama with Spacey and Fincher attached was a no-brainer, to the point that the company committed $100 million for two 13-episode seasons.

Image result for images kevin spacey house of cards

Netflix’ understanding of consumer habits and preferences is becoming increasingly sophisticated.  Since most movies are now downloaded, the company can monitor exactly how we watch them.  For example, every time we pause the film, Netflix knows at what point in the movie we have gotten up to pee, stir the soup, or let the dog out and for how long.  They don’t care what we did when we paused the movie.  They are only interested in when. If a significant percentage of viewers paused during a particular scene – a love scene, one with children, or at an office – Netflix can infer that something is wrong with the subject, the pacing, the format, the cinematography.  They can neither tell why we paused or what we did when we did, but they realize they need to do more investigative data mining to learn more and to better tailor-make their product. 

Based on subscriber information Netflix knows the likely gender of the audience for a particular film.  Although a man may watch a film with his date, Netflix knows at least the number of men who are watching.  Since Netflix’ algorithms are tracking all movie-watching habits, the company can know if you always paused films during the same type of segments.
In 2012, for the first time ever, Americans watched more movies legally delivered via the Internet than on physical formats like Blu-Ray discs or DVDs. The shift signified more than a simple switch in formats; it also marked a major difference in how much information the providers of online programming can gather about our viewing habits.
This switch to digital viewing has been a boon to Netflix, for they can determine with great precision exactly what you are watching, for how long, and with what interruptions:
The scope of the data collected by Netflix from its 29 million streaming video subscribers is staggering. Every search you make, every positive or negative rating you give to what you just watched, is piped in along with ratings data from third-party providers like Nielsen. Location data, device data, social media references, bookmarks. Every time a viewer logs on he or she needs to be authenticated. Every movie or TV show also has its own associated licensing data. The logistics involved with handling every bit of information generated by Netflix viewers — and making sense of it — are pure geek wizardry.
Netflix’ Senior Data Analyst Mohammed Sabah recently stated that Netflix was developing new ways to analyze what’s going on in the movies themselves:
Sabah said it already captures JPEGs and notes the exact time that credits start rolling, and it’s looking to take into account other characteristics. It could make a lot of sense to consider things such as volume, colors and scenery that might give valuable signals about what viewers like (Gigacom, Derrick Harris 6.14.12)
In other words, user data derived from Pause, Rewind, Volume, color adjustments can be filtered, checked against the actions of millions of viewers to give the company insights in what consumers like and how new productions can be tailored to suit these tastes.

Netflix data also includes the time of day and location viewers watch movies.  Advertising for those movies showing at preferred times can appear prominently when a viewer presses his On Demand button.

This all means increased revenues and profits for Netflix.  No longer do the scenes in Robert Altman’s movie The Player resonate.  In the film screenwriters pitch stories to a busy Hollywood executive who, based on his intuition and personal experience, decides which movies get made.  Today intuition, personal experience, and individual judgment are out the window, and data mining is in.
The movie Moneyball has many more relevant scenes.  There is one, for example, where Billy is telling his staff how recruiting will be based on numbers, especially on-base percentage.  His old-guy staff grumbles and complains.  Numbers don’t tell the whole story, they say.  We are the ones with thirty of years of experience.  We have been in the stands watching thousands of young players, studying their habits, their abilities, and their personalities.  No numbers can do that. 

“I don’t care about that”, replies Billy.  “I only care about on-base percentage because when a player gets on first base, more often than not he scores”.  End of discussion.  History has proved Billy Beane right.

The groaners and naysayers are many.  Billyball will sap the game of its spirit, spontaneity, and unpredictability, they say; and Netflix will turn us all into zombies, consuming exactly what we want based on what we wanted.  Hollywood will no longer gives us new, innovative, creative films that challenge our assumptions, turn new corners, touch parts of us we never knew we had.

What we forget is that movie-making, like baseball, is a business; and that Hollywood is out to make a profit.  If that profit can be derived from producing and reproducing movies with familiar themes and actors, there is no reason why they should break from a successful business model.  Moreover, given their new big data acuity, Netflix will be the first to know if old retreads are being tossed in favor of something new.  Big data lets consumers dictate what they want to see, when, and by whom.

This is not new.  Why is it that we want to watch the aging Bruce Willis and Sylvester Stallone yet again in an action movie?  Because people like to watch them blow things up.  For the moment Stallone,pushing 70 is still a cash cow.  Car manufacturers rarely completely redesign their product, preferring to make small, incremental changes.  Detroit will never again make the same mistake it made with the Edsel.  Moreover, it is no accident that cars from different manufacturers all seem to look alike.  Companies are risk-averse, and find that the promise of regular smaller profits is worth far more than taking a big risk for possible higher rewards.

The issue of creativity in films raised by Leonard is a non-starter because there have always been Hollywood blockbusters and small indie films and before them ‘art films’ and intellectual ‘foreign films’.  There will always be a small group of people who prefer the unusual, the new and different, and the challenging.  For now, there numbers are sufficient to justify small films.  When these numbers decrease and when and if these selective viewers turn to the mainstream, then indies will disappear; but it will not be because Netflix has turned them into zombies.  It will be because their preferences have changed.
The companies that figure out how to generate intelligence from that data will know more about us than we know ourselves, and will be able to craft techniques that push us toward where they want us to go, rather than where we would go by ourselves if left to our own devices. I’m guessing this will be good for Netflix’s bottom line, but at what point do we go from being happy subscribers, to mindless puppets?
I do not agree.  No one will push us to where they want us to go, because we are the ones providing user data to Netflix and Amazon.  They are only transforming those data into product.  Netflix has stated that approximately 75 percent of its subscribers are influenced by what the company has suggested to them.  This is not because Netflix is this mad, corporate, evil mastermind; but because we have taken the time and effort to rate films, thus codifying and categorizing our preferences. The fact that so many of us watch films suggested ‘by Netflix’ is because their algorithms are working very nicely to gauge what we like.


  1. Excellent discussion from both a free market economics and a marketing research perspective.

  2. I like this post! Truly nice idea!
    I need to keep this tips in mind!


Note: Only a member of this blog may post a comment.