Statistics are The Magic behind Disney Resorts


  • Author: Carlos Grajales
  • Date: 15 Jan 2019
  • Copyright: Image: Ian Dagnall/Alamy Stock Photo

It is hard to know when Walt Disney, entrepreneur and animation icon, first thought of creating a park. One of the earliest signs of the idea was described by his wife, who recalled that the family used to visit the Griffith Park merry-go-round and Walt used to complain: “There’s nothing for the parents to do… You’ve got to have a place where the whole family can have fun.” [1]

It wasn’t until the late 1940s that Walt Disney, a model train enthusiast, started playing with the idea of creating an entire miniature village, populated with his personal collection of layout miniatures. As the project grew in scope he even hired layout artist Ken Anderson to paint scenes, that Walt would bring to life with his miniatures, as well as a sketch artist to help him. By then, the idea was to exhibit it around the country as a series of “jukeboxes”, where kids could use coins to “play” the different scenes. Always a fan of technology, Disney devised a series of mechanical artefacts that would “move” the miniature scenes to simulate the effect of a real, living town. By then, the project was called Disneylandia [1].

As a separate project, in the spring of 1948, Walt discussed the idea of constructing a scale-model passenger train to circumnavigate Disney’s animation studio, part of a guided tour he had imagined [1]. One employee pointed to Walt a profitable railroad concession at an amusement park outside San Francisco. Walt loved the idea and immediately related it with his miniature village project. Disneylandia would then become Disneyland, no longer a miniature village but a full-sized town with its Main Street, a train station and several areas for guests to visit and marvel at.

thumbnail image: Statistics are The Magic behind Disney Resorts

Three thousand workers and a million square feet of asphalt later, Disneyland finally opened to the public in July 1955 [1]. According to urban planner James W. Rouse, by the time it opened, Disneyland was amongst the greatest pieces of urban design ever [1]. Its design was far from conventional at the time, following Hollywood based principles that forever changed the amusement park industry. Disneyland’s proportions were manipulated to achieve cinematic effects: lower floors on Main Street were nine-tenths scale, second floors were eight-tenths and the third floors were built with a seven-tenths scale. Some other sections and objects of the park were scaled according to their functions and positions, to achieve a realistic effect that subconsciously allowed guests to see the town as a “toy”. Attractions were devised as “movie scenes”, going from one attraction to another involved a harmonious transition, with different scenarios and even a different texture of pavement, simulating the smooth transitions seen between sections of a film [1].

Since its inception, Disneyland was designed to become a one-of-a-kind experience. More than 60 years after the first Disney park opened, the company is still looking to achieve this same landmark, but this time, instead of relying on Hollywood for the next generation of park services, Disney Resorts are investing heavily on statistics to improve customer experience: from high-tech wearables to statistical modeling.

Back in 2013, Disney introduced the MagicBand, a water proof wristband wearable by every guest. The band is equipped with short range Radio Frequency Identification (RFID) technology and a 2.4 GHz transmitter that tracks your location while inside the park. With an estimated 1 billion USD dollars of investment, the bands serve guests as hotel keys, credit cards and tickets [2] [3] [4]. They are also used to constantly transmit information on the whereabouts of the guests and where they are spending time and money. This huge dataset is managed with a combination of servers using Hadoop, Cassandra and MongoDB and feed several dashboards and data-mining operations that drive the daily operations of the parks [5].

Data from these wearables is used to optimize resources within Walt Disney resorts [3]. The location of costumed characters is modified in real-time when traffic surges are detected; intelligent stock reallocation in restaurants and stores is driven by data as well, to avoid shortages in the sections of the parks with higher agglomeration. The same dataset allows the park to identify seasonal peaks and popular attractions, which help the personnel prepare in advance with additional resources as required. Even more, statistics are a powerful weapon to combat one of the most notorious issues within amusement parks: waiting lines. The bands allow visitors to schedule their rides and attractions even before they reach the park [2]. Any guest can program in advance what sections of the park he/she is interested, and the system will automatically suggest an optimized route, which considers not only mathematically optimized paths but also the statistics of ride use and saturation. The park will thus recommend the best time to visit popular attractions and might even offer incentives to avoid certain attractions that are overflowed with visitors [5]. This can even happen in real time: if you happen to be in line for a very demanded attraction, the band might suggest a tailored alternative to avoid the waiting line [2], all based on your recorded preferences. This data-driven optimization allowed Disney to accommodate 3,000 additional daily visitors due to optimizing resources [4].

It is not only schedules that are optimized with data. Some other experiences are now integrated seamlessly thanks to this analytics program. For instance, reservations for a restaurant can be linked to the wristband, so the personnel know in advantage when you are coming, based on your location. If you happened to order in advance what you wanted to eat, the team starts preparing the order as soon as you come near the place. So by the time you are seated at the table for only a minute or so, your hot meal is served [2]. Another current application of the MagicBand relates to Photo services offered by Disney Resorts. Professional photographers are around the park, taking photos of the moments you and your loved ones are enjoying. The information on the bands allow proper identification of the guests, so they can receive their photographs in their hotel room, after reviewing them online and choosing the ones they prefer. These real time IDs are also used for some of the most amazing uses of real time analytics ever recorded: costumed characters in the park greet you and your family by your name [2]. Welcome to the future.

Another way they use data is through geo-targeting, which basically means sending guests real-time information relevant to their current location. In practice, this is mostly ads, like the current discounts found in the store you are passing by. But merged with the detailed information gathered from your consumptions within the parks, these ads are tailored to be more relevant. Even after you leave the resort, statistical analyses of your behaviour during your last visit is used to send micro-targeted e-mails offering incentives for you to return, such as great discounts on some type of souvenir you bought or maybe a reservation for a restaurant you encountered closed in your visit [5]. These micro-targeted incentives are specifically designed to drive you to come back.

Some of the most state-of-the-art approaches to data analysis are also being integrated into Disney resorts, even statistical models that were born within other parts of the company. Among the oldest tactics that movie companies have used to test their features is through pre-screenings with test audiences. A group of people usually preview early cuts of films and give their feedback, which is then used to make changes based on their suggestions. Well, after years of doing it the traditional way, Disney is currently deploying a statistical driven approach to do this. Instead of directly asking the public what they think of the movie, they use a series of advanced video cameras to record their reactions and then use statistical models to translate those reactions into sentiments, similarly to what a social-media-based sentiment analysis would do to classify social media posts [2] [4]. This model yields way more information on people’s reaction to a film, allowing for deeper and richer analysis. Now, it is expected that this technology will be implemented within the Disney parks as well, with cameras recording and analyzing the emotional reactions of the visitors on each attraction, ride or restaurant [4].

And just as the late Walt Disney jumped from idea to idea to finally create Disneyland, 21st century Disney is playing with several different applications of data intelligence across their full enterprise. Disney Research, an innovation center within the Disney corporation, has a full department devoted to research in machine learning and optimization algorithms [6]. Their topics of research are not as straightforward as you might expect. Sure, they have some work on using deep neural networks to optimize computer animations, but their work on data-driven soccer analysis is also remarkable. For those interested in analyzing the defensive movements of the Swansea City football team, feel free to review their paper about it [7].

It was over 60 years ago that Walt Disney disrupted the amusement park business with innovative architectural and design principles. Now, his company is trying again to disrupt the same ecosystem, but this time with the use of statistics.


[1] Gabler, Neal. Walt Disney. The Triumph of the American Imagination. Vintage Books Edition , 2007
[2] Marr, Bernanrd. Disney Uses Big Data, IoT And Machine Learning To Boost Customer Experience. Forbes Website (Aug, 2017)
[3] Madhavan, Archana How Behavioral Analytics Changed Disney World Forever. Big Data Zone Website (Apr, 2018).
[4] Disney World Meets Big Data. Medium Website (Aug, 2017)
[5] Van Rijneman, Mark. Walt Disney’s Magical Approach to Big Data. SlideShare Presentation (Dec, 2015)
[6] Disney Research Website.
[7] M. Le, Hogan et al. Data-Driven Ghosting using Deep Imitation Learning. MIT Sloan Sports Analytics Conference 2017 (March, 2017)

Related Topics

Related Publications

Related Content

Site Footer


This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.