Big Data Behind Disney Magic

Disney is using advanced analytics to enhance its business and create personalized, special experiences for customers.

Big Data at Disney: Introduction

Disney is a diversified global entertainment company best known for its high-quality, family-oriented films and theme parks.  While Disney is relatively less known for its commitment to using advanced analytics (likely because the company aims to conceal “the mess behind the magic”), Disney has quietly been investing in big data applications for a decade (1). It is estimated that Disney has about 1,000 full-time employees dedicated to building advanced analytics capabilities (2).  Interestingly, the organization has nurtured an experimental approach to analytics. Teddy Benson, Head of Solution Integration at Disney, explained: “we do proofs of concept almost like startups… we take a small budget and look at the risk, do a few investments with various levels of success and see how far we can get” (2).  Still, there is evidence that Disney is willing to “bet big” on data solutions that can facilitate exciting, frictionless and personalized experiences for its customers.

Below are three known examples of big data usage at Disney, across its business segments.

Optimizing Park Logistics: MagicBands

In 2013, Disney released its “MagicBand” concept at the Walt Disney World Resort in Florida – a program that reportedly cost Disney over $1B to develop and implement (1).  Today, every single guest of the Resort receives a wearable MagicBand bracelet that includes an RFID chip.  The device serves a range of purposes from the customer’s perspective.  It functions as the hotel room key, provides access to the various theme parks, and allows guests to make convenient payments (charged back to their hotel room).   More importantly, the bracelet constantly communicates with sensors spread throughout the parks, generating huge amounts of data around the movements of each individual customer. Disney’s operations team already uses this data to dynamically improve customer experience.  The data might reveal, for example, that a set of customers is waiting in a long line for a ride.  Disney’s team can offer up real-time incentives to redirect those customers to less congested areas of the park (3).  There is huge value created by optimizing park logistics in this way – customer satisfaction grows and resources can be used more efficiently. Further, this wealth of data could allow Disney to develop all kinds of innovative park features.  Analysts speculate that in the near future, Disney will be able to do things like predict a child’s favorite character (Cinderella, Mickey Mouse, etc.) based on how the child behaves in the park, and arrange for that character to locate and surprise the child for an unforgettable meet-and-greet (4).

Optimizing Ticket Pricing: Lion King on Broadway

The unprecedented success of Lion King on Broadway reflects another noteworthy case where Disney harnessed big data effectively.  The show has generated more ticket revenue (about $8B) than any other Broadway show, and yet its ticket prices have never been the highest on average (5, 6). This success can be attributed, in part, to the dynamic pricing model that Disney developed for the show tickets.  Disney used historical Broadway ticket sales data to accurately forecast future demand for Lion King.  The model they developed also predicted the highest possible price for each ticket that customers were likely to accept (7).

Optimizing Film Production: Affective Artificial Intelligence (AI)

When Disney produces films or television shows, an important step in the process is testing market fit. Before big data, Disney Studios would preview early cuts of its content for focus groups, and try to extrapolate from that group’s feedback how a larger audience might respond.  The process was tedious and error-prone.  Now, Disney is creating technology to make these market fit tests faster and more accurate. Specifically, Disney has been an early adopter of Affective AI, an area of computing focused on identifying and interpreting human emotions.  A recent research paper published by Disney and Caltech described placing cameras in a movie theater to monitor each face in the audience. Backend systems could then evaluate the audience’s emotional response to each moment of early-cut content screening. This method generates significantly more granular data than before (about 5K data points per person during an average movie) (8).  Disney scientist Peter Carr explains, “it’s more data than a human is going to look through. That’s where computers come in – to summarize the data without losing important details” (9).  In the future, Disney could use big data and affective AI technology for other purposes.  For instance, to understand how customers are feeling as they move through its theme parks.  Or, even to select in real-time between multiple possible endings to a film, based on predictions about what the audience will most enjoy (8).

Potential Challenges

 With its customer-centric big data solutions, Disney will need to be very mindful of customer privacy.  Already, with MagicBands, Disney collects the consent of its customers before it tracks their movements within the park.  Some have speculated, though, that Disney customers will be more tolerant of the “convenience-surveillance” tradeoff given their longstanding trust in the brand.  Disney World, for example, is “wrapped in an idealized vision of life that’s as safely self-contained as a snow globe.  Disney is thus granted permission to explore services that might seem invasive anywhere else” (1).

Do you agree?  Will Disney receive an unusual amount of leeway to collect and analyze customer data?  Or will customers eventually push back, finding these data-driven solutions unsettling?

 

  1. https://www.wired.com/2015/03/disney-magicband/
  2. https://www.computerweekly.com/news/252439803/Doing-analytics-the-Disney-way
  3. https://www.nytimes.com/2013/01/07/business/media/at-disney-parks-a-bracelet-meant-to-build-loyalty-and-sales.html?utm_source=datafloq&utm_medium=ref&utm_campaign=datafloq&pagewanted=2&_r=0
  4. https://www.forbes.com/sites/bernardmarr/2017/08/24/disney-uses-big-data-iot-and-machine-learning-to-boost-customer-experience/#50bdbd443387
  5. https://www.forbes.com/sites/leeseymour/2017/12/18/the-lion-king-is-making-more-money-for-disney-than-star-wars/#1824a2e11ff0
  6. https://www8.gsb.columbia.edu/bizanalytics/content/why-“-lion-king”-roars-so-loud-business-analytics-disney
  7. https://www.computerweekly.com/news/252439803/Doing-analytics-the-Disney-way
  8. https://www.fastcompany.com/90134144/disneys-next-movie-it-could-be-watching-you
  9. https://www.caltech.edu/about/news/neural-networks-model-audience-reactions-movies-79098

Next:

23andMe, who does it profit?

Student comments on Big Data Behind Disney Magic

  1. Disney is one of my favorite American brands and I’m happy to see that they are making improvements through the use of magic bands and AI. However, as the question of privacy starts to show up in the public space, how can Disney assure parents that their children’s data will be protected? Do you see a future where some parents may want to exempt their children from facial recognition technologies?

  2. Thank you for this post! I haven’t been to a Disney Park since the launch of the wearable MagicBands, but I have friends that loved their experience with it. It demonstrates the power of developing something that captures value in data assets while simultaneously creating value for the customer by enhancing their experience through improved conveniences rather than taking away from it. In this particular case, I believe that the benefits gained from the improved experience outweigh concerns about “surveillance.”

    I’m a bit more skeptical regarding the use of facial recognition for affective computing in movie theaters. Somehow the idea of being “seen” and watched while watching a movie in a more intimate, albeit public, setting feels more unnerving. Even if identity is anonymized, it feels as though it would take away from the experience knowing that I’m being watched in a way that is different from being observed in a Disney Park/customer service setting.

  3. Interesting article. Many attraction parks are currently exploring options to mine the amount of data gathered on-site to improve customers’ experiences or even lower operating costs. All these opportunities often come at the expense of the privacy of the customers and I am very curious to see how this dilemma will evolve in the future.

  4. Great read, thank you! It’s fascinating to read about how Disney has built a capability in digesting and analyzing data. It seems to be embedded in the culture of the company now, which in turn has led to more customer satisfaction and revenue. The examples you gave all provide different approaches and metrics that the company has used to optimize the customer experience. It really shows that building a data capability requires flexibility and the appropriate skilled workers (i.e. data scientists) to carefully evaluate and leverage the information.

  5. Thanks for sharing about Disney’s latest big data initiatives! Out of the three programs discussed here, I personally like the MagicBands idea the best because waiting in long lines – sometimes for hours – is a shared painful memories amongst millions of Disney fans worldwide, and MagicBands can definitely help alleviate this problem by redirecting visitors to less congested areas. The wait time has been hurting customer experiences, and may lead to safety concerns as seniors and little children simply can’t stand and wait for too long especially during summer and winter times. Echoing the previous comments on privacy, I would also feel uncomfortable watching movies in a cinema when a camera is constantly monitoring my facial reactions. That’s just not the way movies shall be enjoyed. And this trade-off makes me wonder to what extent do we want state-of-the-art technologies to enter into our daily activities? Does more tech mean better life? Not necessarily.

  6. Great article. The key question that I think about as it regards this how data-derived decision making may affect the ability of Disney to attract new customers whose preferences may not be reflected in its existing dataset. Particularly as it regards certain groups (racially, Socio-economically, etc.)

  7. Great post! I agree that “Disney customers will be more tolerant of the ‘convenience-surveillance’ tradeoff given their longstanding trust in the brand”, and that the Disney park is “wrapped in an idealized vision of life that’s as safely self-contained as a snow globe. Disney is thus granted permission to explore services that might seem invasive anywhere else”. However, I am not sure this is a good thing. The cameras in the movie theaters definitely has a creepy sci-fi vibe to it. The eerie aspect of surveillance seems all the more sinister when people are trusting without question. That said, I grew up on Disney films and going to the park as a child was a dream come true, a defining experience for a kid. I hope that right balance is found.

  8. The example that I found most interesting was the MagicBands. I can see an added convenience for the consumer, by being able to use one device to gain access and pay at all of Disney’s venues. What caught most of my attention, however, is that Disney does not yet have a use case for the data. Was one of the objectives behind the MagicBands creation to generate data? I see the value in starting to create database, even if the company does not yet have a use case for the data. However, I have the impression that a lot of companies don’t yet know what to do with the data they are collecting. I would think that you would first want to find a problem and then see what type of data you should capture to solve that problem.

  9. Thanks for writing on this! We’ve seen in this class with Viacom that the entertainment industry must adapt to this new paradigm, and incorporating real-time data into content decisions is a ‘must-have’ to be able to survive. With the addition of their new streaming service Disney +, it’ll add another input into their systems.

    Disney definitely takes it a step further with its parks though. The enhanced customer experience would be incredible though, particularly if it translates to less wait times and increased satisfaction. There may also be an opportunity to generate higher inventory turnover in slow-moving products (or rides) by pushing ads. Looking forward to seeing what happens with this!

  10. Thank you for the valuable information on how Disney is taking a nearly century old company into the 21st century age of data analytics. I love the idea of magic bands and using AI to recognize people’s emotions…. from Disney’s stand point. However, I worry about the privacy issues in the future. I could imagine that Disney will connect all the data it has about current customers to the same customers Disney Plus (their new video streaming platform) profiles. With the rise of watching videos on tablets and phones – both devices which have a customer facing camera, I could imagine Disney falling into temptation of taking advantage of the camera on those devices and observing people watching its content to gather data about emotional reactions from facial expressions. This has a potential to be a slippery slope.

  11. Thank you for writing such an interesting post, I am particularly amazed by the real time movie ending option, it is such a creative way to use affective AI! You are raising the question of consent which I find being the most recuring issue when looking at big data in general, but as you mentioned, in this case, it seems that when it occures within clear geographic and time boundaries – a movie theater or a theme park, it somehow makes it more acceptable, especially if it is considered a form of payement, at least in the context of the theme park magicband users.

  12. This was a really fun read! It was really interesting to read about how Disney is working on using data in so many different ways. Like many of the previous comments above, data privacy concerns of especially children are on my mind.

    It will be interesting to see how they continue to use big data for the Disney streaming service they are going to launch soon.

  13. Many comments address the privacy issues potentially created by Disney’s big data initiatives. This is certainly a valid point. However, I am wondering if the company even risks loosing something fundamental to its business model – creativity and magic. The article does not explicitly discuss it, but Disney also uses data to decide which products or content to produce. I am sure, data helps in this regard, but I am always skeptical when creativity is replaced by mechanics. Think of the early times when GPS systems for cars (without dynamic route optimization) were introduced. I remember that suddenly everybody was using the same routes, and traffic jam was created on roads which used to be empty. If content companies such as Disney focus too much on data analysis, could they loose their competitive edge?

  14. Thanks for the post Kate! It exhibits a very interesting juxtaposition between the value of data to buyers and value that can be created for consumers. It really causes a problem for privacy advocates when the data collected is in such a personal setting, but does create delightful experiences.

Leave a comment