Movie has always been an integral and integrated part of our lives, from the classics like ‘Mother India’ and ‘Sholay’ to underrated ‘Aks’ and recent blockbusters such as ‘Chennai Express’. Movies have always fascinated people and influenced us in some way or other. Now if we leave aside the debate about the dearth in novelty of ideas in the industry these days. What still excites us, is the number that these flicks garners , Whether it’s Talaash (1969), which was touted as the first movie to have budget of 1 crore to Ra1 which had the budget over 150 crores. The volume of money and efforts getting involved is very high. With more than 600 Television channels, 100 million paying-for- set-top-box households and over 1000 films produced annually, we are looking at an industry which will reach to Rs. 1661 billion by 2017 (FICCI-KPMG report).
A bitter truth, but now it’s not about the content of the movie which counts, it’s all about how much money it’s raking and in how many ways. Earlier it was only the revenue from the tickets and up to an extent the music sales but now the game has been change drastically with media houses finding alternate sources of revenue such as pre-selling satellite rights and home video rights. According to a study tickets sales now only accounts for 60% of the revenue generated by a film.
Now what Analytics can do for an already flourishing business?
The answer is it can help to make it efficient and even more profitable. Within this entire hullabaloo, what arises is the ‘Risk involved, room for improvement and prediction’. Now focus groups and audience polls are already there in the industry, but these are approaches are outdated and ineffective and hence results in disasters like Joker and Tees Maar khan.
With the help data mining techniques, statistical analysis combined with an understanding of the film industry to give production houses and filmmakers important insights into the potential for financial success of their film projects.
The film industry, which we see now a days, consist of a wide range of components that can be quantified and reduced to any form of statistical based analysis. Such as you can actually predict that a Don 3, with Shahrukh Khan and Priyanka Chopra, directed by Farhan Akhtar, releasing on Deepawali next year would be a hit or a flop or what amount of revenue it would be generating for the makers, all this before the film is not even being started. Now with these kinds of insights, the makers of the film can actually decided the budgets and over-under spend accordingly, to make more profit, with less risk involved.
The Data: Does lots of it means a lot?
Recently Google announced that it has developed an algorithm that can guess movie box office hauls with 94 percent accuracy. It will use the timing and category of Google searches and paid clicks and now this is something which is relevant and can be captured. With the Pages of the movies being made with the announcement of the movie, the twitter fights, find the buzz that is being generated by the movie, and measure it quantitatively to derive a conclusion.
Besides the already being used source of data such as Facebook and Twitter, Wikipedia is emerging as a dark horse in the run of data paradise. In a recent study the researchers compiled the 312 movies released in America in 2010 that had Wikipedia pages. They measured four variables: number of views, number of human editors, number of human edits, and collaborative rigor (a function of edits per editor; a higher rigor value means the edits are spread out among more humans).
The correlation was almost perfect; getting an R-squared value of 0.94 (an R-squared value of 1 would indicate perfect correlation). People are reading about, what they would be watching in the coming future, editing it, adding to it. So the most basic and rich source of data would be the audience data and its efficiency would depend upon how deeper we can dig into it.
Implementation of the Idea: To do or not to do and if do how much to do and when to do!
If we talk about implementation, People may argue that, the use of Data Analytics can only predict whether the movie will bombed or not, and it will affect the creative process of film making. My argument would be First, it is not only about that movie will be a hit or flop, with the right prediction we can actually, cut costs to lessen the risk involved or change is the cast of the film or in an extreme case do not make that movie at a first place, Possibilities are endless.
Now coming to second point, the kind of data analytics services which we are talking about would be efficient and effective for movies which have budget on the higher level and not the Independent art movies. However the inputs from the data analytics firm and expert input from a industry will hold equal or proportionate weightage (Depending on the content of the film, of course) and hence not be polluting the outcome. A human point-of-view is very important to any form of data analysis. It is not about giving importance to one over the other. Analytics insights and Creative department of the film will be working together as a system of creative checks and empirical balances.
In the end:
It may sound like an overenthusiastic idea but looking at the facts, it can be safely said that we are sitting on a gold mine. It all depends on, who will move his butt and see it first. Till date there is no firm in India, which provides any kind of Data Analytics services for Film and Entertainment Industry in India and when we talk about the whole Industry, it will bring in the Television, Video Games, and Music etc. So though I have said it earlier, I will say it again.
Possibilities are endless.
Try deep learning using MATLAB