Netflix: The power of data
In the convergence era, we’re theoretically able to have great control over how and when we consume content. But with the big data trail we leave behind thanks to new media, are we becoming more of a herd than ever before? As Siegel (2013) noted, prediction is power, and this power offers a competitive advantage to online platforms that have strongly defined audiences. Netflix is an excellent example of a company harnessing the power of data mining: for creating new shows, acquiring and pricing shows aired by other providers and suiting these shows to user tastes, here's a short insight into their work.
Leonard (2013), argues that data mining turns viewers into mindless puppets, since Netflix uses it’s database of 29 million subscribers to find out who is watching, when they pause, when they stop and what kind of device they watch on. Instead of relying on pilot episodes like most television studios, Netflix made a $100 million gamble on House of Cards through data mining, reducing the risk through predictive analysis (Siegel 2013), which saw Netflix analyse user data on Kevin Spacey movies, political dramas, the original British series of House of Cards as well as producer David Fincher (Leonard 2013). An official competition run by Netflix also awarded $1 million to a team of scientists who improved their recommendation system (Siegel 2013. It’s an interesting comparison between these investments considering how incredibly important this overhaul of the recommendation system was to Netflix, and can further arguments about exploitations through crowdsourcing. This is important to Netflix’s business model before they run on thin margins at only $8 per month, so to continue developing, the subscriber base needs to grow. For this subscriber base to grow, people have to get value out of the package. To get this value, they need recommendations that suit them, so they continue their subscription. This is a complex form of association learning (Furnas 2012).
In a way, despite the fact that broadcasters now offer viewers more ways to catch-up on TV, the user input very much limited due to the strength of recommendations available from data mining. But it also raises questions for the future of television. Will it become a series of cheap thrills, where data makes viewing habits so open and available for exploitation that television finds a way to become even more formulaic?
The Netflix example also offers interesting perspectives on the engagement between television and social media. In an era where live-engagement with ‘event’ television, can be seen as a prerequisite for a full viewing experience (Harrington 2013), Netflix focuses on wider buzz through its delivery strategy.
Set your clocks and put the coffee on. House of Cards goes live at 12.01 AM PST tonight. pic.twitter.com/7LmeuoSBwE
— House of Cards (@HouseofCards)
February 14, 2014
Season 2 of House of Cards went live at 12.01am American Time on Valentines Day, and despite this, still gained incredible buzz around the world.
This creates a unique scenario where customers are happy at getting valid recommendations due to their tastes, while Netflix benefits from the data consumers (somewhat unknowingly) share.
Reference List
Furnas, Alexander. 2012. "Everything You Wanted to Know About Data Mining but Were Afraid to Ask." Accessed May 9, 2013. http://www.theatlantic.com/technology/archive/2012/04/everything-you-wanted-to-know-about-data-mining-but-were-afraid-to-ask/255388/.
Harrington, Stephen. (2013). “Chapter 18 Tweeting about the Telly: Live TV, Audiences, and Social Media.” In Twitter and Society edited by Katrin Weller, Axel Bruns, Jean Burgess, Merja Mahrt & Cornelius Puschmann, 237-248. New York, NY: Peter Lang.
Leonard, Andrew. 2013. "How Netflix is Turning Viewers Into Puppets." Accessed May 9, 2013. http://www.salon.com/2013/02/01/how_netflix_is_turning_viewers_into_puppets/.
Siegel, Eric. 2013. "Introduction - The Prediction Effect." Predictive Analytics, 1-16. Hoboken, NJ: Wiley.















