Over the last few months I have been doing commentaries on the subject of Big Data. The enormous amount of digital information has been used by companies to find customers and by government to find terrorists. Hollywood has now found an important use for Big Data.

James Hirsen reports that Google data is now being used by Hollywood executives to predict future box office receipts. By paying attention to search information data, they now claim they have 94% accuracy in box office predictions.

A Google research paper explains that the number of Google and YouTube searches of a movie has the capacity to provide the film industry with a highly accurate predictor mechanism. Will the movie be a blockbuster or a bomb? Find out how many people are searching on the movie title. Of course, it is a bit more complex than that.

By analyzing the number of Google searches for a particular movie, predictions on the box-office gross can be estimated with approximately 70 percent accuracy. If you begin to add in the number of search ad clicks, you can improve the accuracy. Google researchers built a model that included search volume and search ad clicks over a seven day period. This model gave them 92 percent accuracy.

By combining search volume on the movie trailers with other metrics, such as the franchise status of the movie and the season of release, they were able to achieve a model with 94 percent accuracy. It is worth noting that they could make some of those predictions four weeks before the release date.

Obviously, Hollywood executives are interested in Google models. It should help them predict the success or failure of a film. But it will also give them the ability to shape the most effective marketing campaign for future films. I predict they will spend more time working the social media sites in the hopes of driving potential audiences to film trailers and other marketing aimed at getting them to the movies. I’m Kerby Anderson, and that’s my point of view.