AI Comes to the Magic Kingdom
Artificial Intelligence has started to play a prominent role in the Media & Entertainment industry, enabling companies to use data to create customized content and user experiences. While streaming companies such as Netflix have led this movement to date, no company is better positioned to benefit long-term from machine learning than Disney.
The media and entertainment landscape today has been galvanized by two driving forces: 1) an explosion of demand for original content and 2) a change in consumer preferences, particularly among millennials, valuing experiences over goods[1]. Armed with the technology and capability to cater to customers’ media consumption according to their preferences, companies have invested heavily in Artificial Intelligence, turning large swaths of data into personalized content and carefully curated user experiences. While Netflix and Hulu have gained widespread publicity for incorporating machine learning into their core strategy early on[2], no company is better positioned than Disney to use AI to take advantage of these secular trends.
The Walt Disney company’s portfolio includes media networks (Disney Channel, ABC, ESPN), parks and resorts (Disney World, Disney Cruises), studio entertainment (Walt Disney Studios, Marvel) and, within the next year, streaming services (Disney+)[3]. While Netflix and Hulu have successfully used AI to learn and customize content, their data is only limited to in-home TV/movies and their customer base is skewed toward millennials and Gen Xers[4]. Disney’s competitive advantage lies in the fact that it can more accurately predict the demands of a larger customer base. Disney uses its large portfolio identified above to see both the in-home and out-of-home preferences of a customer. The Company’s theme parks, movie franchises, TV shows, and hotels have allowed the company to acquire data on entire families and understand multiple aspects of their entertainment preferences, beyond just their content viewing habits. This allows Disney to create a more wholistic personalized experience for its customers.
Management has already started to invest heavily to use AI to create these personalized experiences. Disney’s short-term strategy revolves around Disney+, the Company’s direct-to-consumer streaming service that will aim to take market share from Netflix/Hulu[5]. The company plans to use AI to predict consumers’ demand for its existing franchises and help determine its next series of franchises. While the company’s short-term strategy aims to catch-up with Netflix/Hulu, their long-term innovative use of AI will help separate them from the competition. For example, Disney now gives customers wristbands that act as identification, hotel keys, movie tickets, FastAccess (for theme parks) and even as a credit card[6]. This enables Disney to know where their guests are, what they are doing, and what they need, enabling them to help deliver a personalized future experience. Not only does this use of big data help with product development, but it also facilitates process improvements. The Company now better understands where the bottlenecks are in their theme parks and hotels and can add staff in congested areas to improve overall throughput time and, consequently, customer experience.
Disney research has also started using AI to improve movie experiences, most recently collaborating with several Universities to analyze the facial reactions of movie-goers. The research team then attempts to find correlations between reactions (e.g. laughing, gasping) and movie scenes. The application of this will help the Company when conducting target audience analysis and group behavior[7], allowing Disney to more effectively produce content and market them to consumers. Within the actual product of these films, the Company is also using machine learning to accelerate and enhance animated renderings. As seen in Figure 1, the company used machining learning capabilities in the movie Finding Dory to help transform the initial graphic design (left) into a higher-quality image (right) in both a time and cost-effective manner.
Figure 1: Disney uses AI to improve visuals in Finding Dory movie
While AI has the potential to help Disney use its data to become a more effective player in media and entertainment industry, management must address a few additional issues. Firstly, Disney will need to maintain its strong culture as it shifts towards hiring more engineers and data scientists rather than candidates with traditional media/business/creative backgrounds. Furthermore, the Company will need to balance its ability to collect substantial amounts of data on its consumers with privacy concerns[8]. Disney’s ability to create an improved, customized in-home and out-of-home comes could potentially cross the line into personal intrusion. Educating consumers as the benefits of this data collection could be a useful first step in addressing this problem.
Looking at the potential AI could have for Disney, there are a few criticial questions that come to mind. Firstly, will the benefits of using AI outweigh its substantial costs? In other words, does Disney really need to do this given its dominant market share and already strong grasp on its consumer demand? Furthermore, what are the implications for consumers in terms of privacy? Is the incremental benefit of customized experiences worth giving companies an all-access look into your personal life?
Word Count: 778
[1] Uptin Saiidi, “Millennials are prioritizing ‘experiences’ over stuff”, CNBC, (May 2016): accessed November 2018
[2] Bill Taylor, “To See the Future of Competition, Look at Netflix”, Harvard Business Review, (July 2018): accessed November 2018
[3] S&P Capital IQ (2018). Disney Inc.: Public Company Profile, accessed November 2018
[4] Netflix Amended Annual Report (10K), https://www.netflixinvestor.com/financials/sec-filings/default.aspx, (February 2018): accessed November 2018
[5] Erich Schwartel, “Disney Lays Out Its Plan to Fight Back Against Streaming Giants”, The Wall Street Journal, (August 2018): accessed November 2018
[6] Edgar Alan Rayo, “Artificial Intelligence at Disney, Viacom, and Other Entertainment Giants” TechEmergence, (February 2018): accessed November 2019
[7] David Nield, “Disney is Developing an AI That Can Judge What Makes For a Truly Great Story”, ScienceAlert, (February 2017): accessed November 2018
[8] Timothy Morey, Theodore Forbath, and Allison Schoop, “Customer Data: Designing for Transparency and Trust”, Harvard Business Review (May 2015): accessed November 2018
Very interesting essay — It’s fascinating to learn how Disney, the king of creativity, has embraced the use of AI in its development process! You mentioned the necessary shift to hire more engineers & data scientists, and I’m intrigued to learn more about how specifically Disney incorporates data into the creative process, a topic that I imagine the company holds close to the chest. In the 2008 Harvard Business Review piece titled, “How Pixar Fosters Collective Creativity,” the Disney Pixar president Ed Catmull refers to data, but primarily including “the rates at which things happen, how often something has to be reworked, whether a piece of work was completely finished or not when it was sent to another department, and so on” [see source reference below]. It’s likely that artificial intelligence was not a large component of Disney/Pixar’s process back in 2008, but the specific process of collaborating between data scientists & wildly creative idea generators seems like it could be a managerial challenge.
Furthermore, I appreciated your point on the need to balance customization with personal intrusion. I’m curious if the Disney park ‘Fast Access’ wristbands will be eventually linked to a consumer’s Disney+ account, or if that connection between in-home and out-of-home data would be viewed as too intrusive.
Source for HBR article: Ed Catmull, “How Pixar Fosters Collective Creativity,” Harvard Business Review, September 2008. https://hbr.org/2008/09/how-pixar-fosters-collective-creativity, accessed November 2018.
I agree that AI has and can significantly change the media industry’s approach to content creation and marketing. I also liked how you connected that in Disney’s case, this extends to a wider audience base, most notably kids, and a wider space- homes, theaters and parks. I do think that there are significant benefits of using AI for Disney- even though it already has a very strong grasp on consumer demands. In this age Netflix and Prime are gaining importance and market share, and already has a large amount of data on its consumers, especially Amazon that I would argue has the highest amount of knowledge about its consumers. They are probably ahead in the curve and better positioned to address its consumer’s demands than Disney is. To answer your question, I do believe Disney needs to do this given how the market is shifting, if it wants to remain a prominent player in the market. Given its time and cost for development of movies is much higher than that of Prime and Netflix (which so far have not made animated movies requiring significant technical costs), it can greatly benefit from using AI/ its facial expression recognition to predict consumer demand, especially that of kids, a market which is untapped and can give them significant competitive advantage over others.
The author raises an interesting question regarding the impact of machine learning and AI on privacy. To what degree is it acceptable for a company to gather data on their customers? The answer to this question depends on many factors, including the types of data a company is gathering, the country in which the company is operating (privacy laws vary significantly between countries), and the loyalty of its customers.
Recent events (i.e. Facebook and Cambridge Analytica) highlight the reputational risk that companies take when collecting customer data and the safeguards they must put in place around the types of data they are collecting, how they are collecting it, and with whom they are sharing it. These considerations are especially important for a company such as Disney, whose brand it build on consumer trust and a “family friendly” image. When integrating AI and machine learning into its strategy, Disney should carefully evaluate potential customer risk as well as political and regulatory risk.
I think this is a really cool concept. Disney is a company that has branded itself around “magic” and creativity, so the use of an algorithm to predict their customer demand is interesting. I agree with one of the earlier comments that pondered how well the data scientists that will oversee the AI and the more artistic employees will interact. How well can the AI mimic the intuition of an individual who has a knack for seeing where the next big trends will go? I think this would create a really interesting dynamic within Disney, but I think that this application does open up “A Whole New World” of possibility!
Enjoyed reading your analysis on the broad AI efforts Disney is launching in each of its different product segments. You touch on this briefly, but in this age of machine learning, I believe having proprietary data to develop your algorithms is a key competitive advantage. This is actually Disney’s biggest strength in that it owns its own theme parks and are on the way to owning their own streaming platform (which will be attractive because they have their own content).
You brought up some open-ended questions that I think are valid concerns. In the short term, I believe consumers will find the incremental benefit they get from theme park convenience, enhanced movie graphics, and better targetted content to be worth the privacy sacrifices they make, given that these seem like less personal traits that Disney is collecting on its consumers.
This was very interesting to read. You made great points and brought up fantastic questions. I think the tradeoff between customer experience and invading customer’s privacy will be at the heart of the issue Disney will face when they utilize machine learning to model customer behavior. The difference in how they use the data, on an individual level per person or by overall trends for a type of visitor, will be crucial to how I view your question. I imagine they will lean towards generalizing trends by user type, in which case I see less of an issue on the invasion of privacy front. This is definitely something Disney will have to keep an eye on as they move forward with machine learning technologies.
This was a really great article, Kabir! Really enjoyed reading about Disney (of course, lifelong fan of all things Disney and Pixar!) and about their future.
While tremendous amount of data and AI enable Disney to create curated content that is extremely personalized for a larger audience to create worlds that can mean something to so many people, I think you bring up several good questions regarding the trade offs. Disney is in a unique position to obtain data on several touch points – whether it’s film, TV, theme parks, etc. But what is the line between leveraging data to create better innovation and invading the privacy of your customers? Being viewed as a primarily children-driven brand, Disney could face massive ramifications were a breach to happen. Most importantly, on an actual content level, will AI really still be able to look at the past and generate content that organically connects? As you mentioned, if the culture shifts more towards data scientists and engineers, do they move away from the creatives who might create a concept that no historical data supports but could strike a cord with the audience? The past is not always the best indicator of the future and our data may not always reflect what will connect – we can’t crave what we haven’t seen. Super interesting article!
Disney has always been one of the best content creator in the world! With AI and ML, it would further improve the experience that to be enjoyed by consumers, which is excited to see. However, there is always a line between the ultimate customer experience and customer privacy. When reading that they will be able to capture audiences’ facial expression, which is a bit scary to me. Is that permitted by the audience. I think most people would not be happy to have their facial expression been monitored. If from audience side, the information could not be captured, is there other ways to improve Disney’s understanding about audience?