Enter a search term such as “mobile analytics” or browse our content using the filters above.
Check your spelling or try broadening your search.
Sorry about this, there is a problem with our search at the moment.
Please try again later.
Epagogix is a company that analyses scripts in an attempt to forecast US box office performance or TV audience size.
On the back of this script analysis, the company recommends box-office-increasing enhancements.
Nick Meaney of Epagogix was speaking to the audience at PUNCH, the Festival of Marketing’s creative-focused event in the East end of London.
He started with a revealing anecdote about a brilliant script that was nevertheless splitting its audience. Sat in a meeting with some studio execs, they were trying to decide if a particular script was a zombie movie or an undead movie. There is, of course, a difference and the genres are distinct.
The data analysis suggested that this film should decide upon either ‘zombie’ or ‘undead’, or risk diminishment of its box office.
These kind of script assessments are performed using Epagogix’s neural network. Neural networks work on accumulated intelligence. Over time, as more data is fed into the network, the weight attached to variables will change according to the relative success of films analysed.
Predominantly, this kind of analysis is a risk management tool, and indeed Nick comes from an insurance background in the city. Scripts are analysed early on before too much money is spent on them. Some may consider it heretical when it comes to the creative process, quantitively scoring it in such a way, but the system is described as ‘shockingly’ accurate.
Nick used a phrase taken from Malcolm Gladwell, an advocate of their methodology, who said it’s important you ‘don’t defer to the superior expertise of insiders’.
Looking at some of the numbers involved in Hollywood, it’s easy to see why risk management is important. An average movie costs around $60 million, and scripts themselves often require as much as $14m before they are ‘pregnant’ and ready for greenlight (apparently the industry lingo).
Nick said that the network is probably more effective at detecting flops than smashes, simply because any novel event is harder to predict from past data.
He gave the example of Avatar. The system would have predicted that Avatar would make top end box office, but as no film had done $500m before, it was impossible to forecast such a number.
Applying this kind of system to advertising is something we may see more and more of. Although it’s difficult to predict what campaigns and movies will win awards (indeed, Nick said an agency had explicitly asked if the tech could help win awards!), it’s capable of fairly accurately predicting audience engagement with a story.
Nick also suggested that a takeaway for brands outside of Hollywood could be to be more forthcoming with their data. If using quantitative methods to assay your creative is effective, sharing data with your agency about your audience and products is required.