Big Data Analytics Goes Hollywood
Big Data analytics are ready for a close-up at companies involved in the movie business.
A movie theater used to receive a film reel from a distributor and display the title for weeks until the next reel rolled into town. A hit movie's name might remain on the marquee for a couple of months. Since the intellectual property was tied up in the celluloid itself, rights, royalties and fees were fairly easy to control.
Things are much more complicated now that films are transmitted digitally. A film may have little more than a one-week window to succeed before the next movie hits theaters and, within a week or two, might be available on premium channels and/or available for rent.
"In the old days, we could manage the film fairly well," said Theodore Garcia, executive vice president of Strategy and Development at FilmTrack, a company that offers a software as-a-service (SaaS) suite that streamlines the complex IP management lifecycle from project inception to distribution. "Digital technology has exploded the number of channels available globally."
The company's goal is to achieve end-to-end visibility and real-time access to contracts, rights and royalties data to help concerned parties within the industry accelerate decision-making, drive productivity and impact the bottom line by deriving insight from the vast amount of available data.
The barriers to achieving this are not insignificant. Disparate studio systems make it hard to track ownership rights to know which rights were exploited, which ones required collection of revenue and which ones were about to expire in different locations. The available channels have mushroomed into a labyrinth-like structure that includes network TV, cable, DVD, Netflix, on-demand, Internet streaming and more.
Imagine trying to sort out all those channels for the thousands of movies and shows produced each year. When you factor in various geographies, channels and rights details, it creates a quagmire of potential conflict, lost revenue and failed opportunities. Garcia offered an example of foreign rights. A company might have German speaking rights for a movie, for example, that applied to Austria, Switzerland and Bavaria but not the rest of Germany. Another contract might apply only to Austria or only to Germany, or even to certain regions within Germany.
To manage this complex environment, "We realized that all the channels had to be managed with advanced analytics," Garcia said.
FilmTrack's SaaS system operates on Microsoft's .NET platform and contains a wealth of metadata on the various properties it manages on behalf of movie industry players. The system also integrates with studio systems, which are often built around Oracle or SAP.
On top of this, the company is adding analytics from Information Builders (IBI) to analyze and model opportunities that may be exploitable. This enables the company to create reports based upon existing contracts within FilmTrack and visualize various parts of the world using a red/yellow/green setup to display which contracts are running out.
"It is estimated that about 44 percent of overall international potential is unrealized," said Garcia. "Now you can check how well you were exploiting your assets. Before, you just didn't know."
Eventually, FilmTrack aims to augment this with more data from financial, point-of-sale, social media and other data sources to move toward more predictive analytics. Garcia gave the example of being able to tell clients which genres and titles would do best in which regions in advance of a property's release.
Data Quality and Data Visualization
Many movies these days come out in 3D. If you went to see "Avatar," for example, you used 3D glasses to create the sense of large pieces of exploding debris landing on your lap. Most 3D movies are shown on a system by RealD, which leases its equipment to exhibitors and earns revenue based on 3D film viewership at locations running its platforms.
However, RealD was having difficulty capturing attendance information in a useable form. Clients sent the data via PDFs, spreadsheets, accounting system outputs and other reports generated by a wide range of enterprise applications that housed the data.
"We needed to be able to capture, consolidate and improve the quality of our data and then present it accurately via customer invoices," said John Dohm, senior vice president of Operations at RealD. "We lacked a consistent way of knowing if we were collecting the right revenue before."
A team in accounts had to compile the data into reports, then estimate the billable amount. Lack of precision led to customer calls and long discussions to more precisely work out the proper compensation.
To solve the problem, RealD hired a business process outsourcing firm to clean up the data and put it into a single data warehouse. It uses IBI's Data Quality Manager software to profile the data, flag exceptions and have them verified.
The result: A database now has one integrated entry for a film such as "Smurfs 2" as opposed to several different entries under Smurfs 2, Smurfs #2, Smurfs Two and so on. Such typos and variations in reporting were commonplace, Dohm noted.
"We have aggregated the data, improved its quality and can visualize it on dashboards as well as reports," said Dohm. "IBI WebFocus analysis tools and dashboards deliver details about the films that exhibitors show to help us comprehend issues and take action."
The next phase is automatic invoicing. As confidence in the cleanliness of the data grows, invoices will be automatically generated. However, they will go via accounts for a side check before being sent out. RealD also uses the system to profile exhibitors, predict expected revenue and maximize yield from its systems. To facilitate this, it provides a visual representation of theaters in a region using a geographic information system from ESRI.
"We have enough granular information to get a global view of all locations and how they are performing competitively and on a time scale," Dohm said. "We can then show theaters how they are doing relative to competitors in areas with similar demographics and socio-economic conditions."
Drew Robb is a freelance writer specializing in technology and engineering. Currently living in Florida, he is originally from Scotland, where he received a degree in geology and geography from the University of Strathclyde. He is the author of Server Disk Management in a Windows Environment (CRC Press).