The New York Times has an article about how Netflix execs supposed knew before they launched House of Cards that it would be a hit. I think it's easy in hindsight, but just because people stream lots of David Fincher movies is no guarantee they'll like a David Fincher-helmed series, or even the next David Fincher movie.
But what is interesting about Netflix's model for HOUSE OF CARDS is that they have metrics. In the video game industry, we have rather elaborate metrics. On a given level of a multiplayer game, we can build a "heat map" of where players are spending most of their time -- every player, not just playtesters. That's because every player of a multiplayer game is on the game company's server, so the game company is getting a firehose of information which a smart designer with the right data miners can use to improve the gamer's experience.
Film doesn't have that yet. It's a lot of work to gauge audience demographics, let along which part of the screen they're looking at. You have to bring in focus groups, and some movies that focus test well flop, and some that focus test great soar, probably because people behave differently in focus groups than they do in real life.
Netflix knows a lot about its audience. It knows how long it takes for someone to stream the next movie, and what sort of movie that next movie is. Do people watch a slew of comedies, or do they watch a comedy and then an action movie? When do they stop watching certain movies or tv series?
Smart execs will be able to mine those data and discover truths about the audience experience. It will influence filmmakers and tv creators. There is a tendency among artists to dismiss whatever the execs have to say ("bean counters!"), but just because the interpretation of the data have in the past been wrong does not mean that new methods of mining data will be just as flakey. Technology gets better.
It's hard to make predictions, especially about the future, but there ought to be an interesting synergy between how easy it is becoming to make a cheap movie, and how much better systems are becoming available to match the audience with just the product they want.
But what is interesting about Netflix's model for HOUSE OF CARDS is that they have metrics. In the video game industry, we have rather elaborate metrics. On a given level of a multiplayer game, we can build a "heat map" of where players are spending most of their time -- every player, not just playtesters. That's because every player of a multiplayer game is on the game company's server, so the game company is getting a firehose of information which a smart designer with the right data miners can use to improve the gamer's experience.
Film doesn't have that yet. It's a lot of work to gauge audience demographics, let along which part of the screen they're looking at. You have to bring in focus groups, and some movies that focus test well flop, and some that focus test great soar, probably because people behave differently in focus groups than they do in real life.
Netflix knows a lot about its audience. It knows how long it takes for someone to stream the next movie, and what sort of movie that next movie is. Do people watch a slew of comedies, or do they watch a comedy and then an action movie? When do they stop watching certain movies or tv series?
Smart execs will be able to mine those data and discover truths about the audience experience. It will influence filmmakers and tv creators. There is a tendency among artists to dismiss whatever the execs have to say ("bean counters!"), but just because the interpretation of the data have in the past been wrong does not mean that new methods of mining data will be just as flakey. Technology gets better.
It's hard to make predictions, especially about the future, but there ought to be an interesting synergy between how easy it is becoming to make a cheap movie, and how much better systems are becoming available to match the audience with just the product they want.
0 comments