The research, published in the IoP and German Physical Society’s New Journal of Physics, comes up with a way of forecasting how many people will go to see a film, based on monitoring advertising data and word-of-mouth in the form of online blogs and social networking sites. The scientists’ model was originally designed to predict how word-of-mouth communication spread over social networks. They applied it to conversations about movies in particular, and found that when they overlapped their predictions with the actual revenue of the films they focused on, the numbers were very similar.
Interestingly, but perhaps not surprisingly, the findings seem to demonstrate that the web has the power, and that the size of a studio’s marketing spend doesn’t much correlate to a movie’s level of hit-ness. The timing of advertising campaigns turns out to be far more important than their size.
It’s all weirdly like Issac Asimov’s psychohistory system of predicting future human behaviour, but limited to whether people are going to watch Pirates of the Caribbean
. Maybe forecasts of intergalactic warfare and migration come later. For those who can tell their WOM from their SNS and their exogenous factors from their coefficients, the research is here
(with a handy video abstract). Executives behind future John Carter
s and Green Lantern
s are already reaching for their calculators.