Big Data Takes a Star Turn at Legendary Entertainment

by   |   November 12, 2015 3:30 pm   |   0 Comments

CHICAGO – Big data is the latest kind of movie magic, changing how films are made and marketed.

Boston-based Legendary Entertainment is the company behind movies such as the Dark Knight trilogy, the Hangover series, Inception, and Godzilla. It uses deep analytics against multiple data sources to inform everything from the choice of actors and the content of trailers to targeting its digital marketing.

Legendary may spend $60 million to $100 million promoting a new movie, a process it goes through four to eight times a year, and so efficiencies directly improve its ROI.


“Our goal is to reduce our media spend by 15 to 20 percent,” said Matt Marolda, Chief Analytics Officer at Legendary. Using analytics to inform its marketing efforts bring the company’s media spending down by “tens of millions” annually, he said.

Marolda made his remarks Wednesday during a presentation, “Changing Hollywood Paradigms with Analytics,” which he delivered at the Predictive Analytics Innovation Summit. The two-day event attracted more than 250 attendees and more than 40 presenters to Chicago’s Hyatt Regency McCormick Place.

Key to Legendary’s effort is microsegmenting the movie-going audience, using worldwide box office data dating back to 2007.

“This is a new paradigm for Hollywood,” Marolda said, noting that studios traditionally have segmented movie audiences into just four quadrants: male and female, over and under 25.

Take, for example, Legendary’s upcoming June 2016 release of Warcraft, a computer-generated imagery (CGI) and live-action movie based on the popular World of Warcraft video game. The studio considers the devoted fans of the game as very likely to see the movie, but Marolda said his team’s job is to find “the persuadable set” – that is, people who may not care about the video game but might still be drawn into the theaters.

Related Stories

Watson Analytics Breaks Down ‘Minions’ Social Media Buzz.
Read the story »

Redbox finds new kiosk sites by tracking DVD rental behavior data.
Read the story »

Use Data to Measure, Track Marketing Success.
Read the story »

Advertisers Push for Data to Measure Audiences in Multiple Channels.
Read the story »

Marolda said one insight around the release of Warcraft has been to play up the cutting-edge CGI in the movie, which was created by Industrial Light and Magic, Lucasfilm’s visual effects and animation studio. The movie’s impressive CGI is on display prominently in the first trailer for Warcraft, which quickly garnered 100 million views after its release earlier this week.

But testing also uncovered other opportunities, such as the idea that audiences were curious about the relationship between the animated “orcs” and the humans. For this reason, the trailer emphasizes character relationships, not battle sequences.

On a daily and, sometimes, real-time basis, Legendary’s 50-person applied analytics team will analyze how trailers are being consumed. This, in turn, helps the studio to tailor the trailers and customize them for different audience profiles. Legendary will create hundreds of microsegments, pushing different, identified, aspects of its films. It may test thousands of combinations of creative against hundreds of audience segments, Marolda said.

Legendary has access to some 600 million email addresses, representing 200 million households, with personally identifiable information. It also studies social media: 500 million Twitter profiles, 100 million Facebook profiles, billions of tweets, as well as thousands of news sites and blogs.

To conduct sentiment analysis against this diverse data, Legendary uses its own proprietary software, Marolda said.

In other projects, Legendary uses game theory to predict how its films will perform in the context of other movie releases, and studies the content of films using thematic decompositions from machine learning.

In addition to marketing – interestingly, Legendary handles all of its media buying, except for TV, in-house – the company also uses big data to help pick promising movie properties and even actors.

For instance, Marolda said an analysis of an actor’s social media footprint or search rank and can help with casting decisions.

But regarding the use of data to make content or casting choices, Marolda was quick to say that it was just one factor for the producer and director to consider. A data-driven approach may transform the marketing practice, but it only informs the creative decisions in movie making, he said.

Ellis Booker is a freelance journalist based in Evanston, Ill. Email him at Follow him on Twitter: @ellisbooker.

Subscribe to Data Informed
for the latest information and news on big data and analytics for the enterprise.

IBM case study compendium: real-world analytics in the cloud

Tags: , , , , , ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>