How the Revenant used Big Data as a Movie Audience Feedback Loop

How the Revenant used Big Data as a Movie Audience Feedback Loop

I’m constantly watching out for ways that Hollywood is using big data analytics (or as I like to call it, film analytics). Here’s a recent example of The Revenant using big data to gather feedback from audiences. The idea is pretty exciting (or creepy depending on which side of the Matrix your on)! Let me break the idea down with a personal application:

My first film was for a 5th grade project. It was sort of an attempt to be SNL while teaching about our research on the rainforest (which was basically that it existed). I don’t remember a stitch about what I learned on the rainforest, but I remember every bit that we did for the 5 minute film. It was hilarious and we were 11 year old geniuses. We had a spoofs on classroom disruption, educational space documentaries, Little Red Riddinghood forcing a disinterested Wolf to chase her… I’m telling you, it was a goldmine. We laughed so hard making this thing we were sure that we could hardly wait to have it screened for a live audience of likeminded peers.

Sure, my friends mom didn’t think it was funny, but she didn’t much approve of me anyhow. And yeah, we expected the teacher not to like it. After all, we were lampooning her classroom and the teaching environment in our assignment. But we were sure to get an A for all the effort and creativity anyway. And we were going to kill that classroom. We were sure to be kings from then on and I was positive that Theresa and Sarah were going to be fighting over me by the end of the day (yeah, I thought that way at 11).


Sourced: Flikr creative commons


We were giddy with anticipation by the time our turn came. We had to go last due to the inconvenience of getting a big boxy TV and a VCR player on top of a wheeled podium from the office. My friend and I were cracking up before the first sketch even got rolling. As the film progressed I started to realize that we were the only ones laughing. I assumed they were just really attentive, but as I looked around I noticed that many of them were in what looked like real physical pain. Not the emotion I was trying to evoke.

Certainly, we could have tested and screened our ideas earlier on and perhaps saved ourselves the embarrassment from what turned out to not be enthusiastic peers, but rather a set of typical overly-judgmental pre-teens. Just visually seeing the reaction might have been a strong indicator on whether we were on the right track. But today’s movies can be infinitely more subtle. The combination of shots, lighting, coloring, sound (or lack of sound), and on and on all lead to very nuanced emotions in the audience. With that kind of depth (competing with the ever increasing expectation of the audience) it can be hard to gauge if you are on the right track by just watching facial expressions. And asking a test audience what they thought about scene x 23 minutes into the movie may be like asking what they remember about the womb (soothing, warm, but a little cramped in my opinion).

For the Revenant, Fox Studios contracted with a bio-analytics company to help track the audiences’ reaction to the movie. The contracted company, Lightwave, modified fitness trackers to see the audiences reactions in real-time. The were able to see how many times they got the audiences’ hearts racing (15 times) and how much time was spent motionless and spellbound at the unfolding story (4,716 seconds). In all, they were quite convinced at the strength of the film.

Revenant stats

From Yahoo!s blog post

There is very little doubt that Alejandro González Iñárritu did a stellar job of directing or that he and Mark L. Smith adapted Michael Punke’s novel in an exciting visceral way. Obviously Leonardo DiCaprio (who finally got his Oscar) and Tom Hardy performed wonderfully as did the whole ensemble, but these things are typical in a great film. They are staple predictors of film success and evoking the intended emotional response from an audience. But what about someone who’s still learning?

With a more sophisticated audience, a filmmaker could (and I might argue, should) use bio analytic feedback along with studying facial reactions to determine if they are communicating to the audience properly because that’s what the art of film is all about, communication. I would even push further that the sooner the feedback is given in the creation process the faster an ensemble of artist can pivot to communicate effectively to an audience. That’s what film analytics is all about. That’s what I’m going to help you figure out how to do so that you can have an edge on your next film.

What is your opinion on using biofeedback to track film impact? Do you do audience feedback and how does that impact your creative process?

Leave a Reply

Your email address will not be published. Required fields are marked *