Delivering the ShinyConf 24 keynote for the Shiny in Enterprise track, Eric Kostello, Executive Director of Data Science at Warner Brothers Discovery, discussed the instrumental role Shiny and its evolution has played in projects he has been involved in over the last decade. 

Big things are happening at ShinyConf 2025! Stay ahead of the curve, check out what’s in store and save your spot.

Eric’s experience using Shiny has evolved with the technology since its early days in 2012. At that time, the team collected data through phone interviews, and they started to use Shiny because they needed an interactive userface to explore data pathologies and to understand and make judgments on new and existing data. Shiny's role in enabling interactivity and accessibility started with technical teams, and has now moved to empower both technical and non-technical stakeholders in effectively understanding and leveraging data insights.

Shiny’s Role in Data Science

Before Shiny, it was a significant challenge to integrate statistical computing and analysis with a graphical user interface. It tended to fall on data scientists to create their own handwritten, code-heavy solutions. Shiny was a game changer, with its framework that enables interactive applications without the need for extensive programming experience. Building on the legacy of the S language that facilitated statistical computing, Shiny has made data science accessible to a broader audience.

Building and maintaining R Shiny apps isn’t always easy. Learn how to get the right support for your team in our new blog.

Watch the Video

Evolution of Shiny

The early days of Shiny had initial limitations with debugging, UI options, and only basic functionality.

The advent of event-driven programming and reactive graph evaluation was a turning point for Shiny, particularly in companies like Warner Brothers Discovery where it allowed for greater efficiency in data processing. For developers, the shift meant that it was easier to implement dynamic features. Managers and non-technical stakeholders also benefited from more actionable insights with less technical intervention.

Today, Shiny integrates with Python and enterprise frameworks such as Rhino and provides scalable and maintainable solutions that cater to diverse business needs.

Want to create powerful web apps in Python? Our ultimate guide to Shiny for Python has everything you need to get started.

Case Studies: Shiny in Action

1. Updating Survey Methodologies 2012

Shiny played a crucial role in transitioning from traditional phone surveys to online platforms. For Warner Brothers Discovery, leveraging Shiny in this way meant the organization efficiently oversaw data adjustments that resulted in $5 million in annual savings. The Shiny app achieved this by efficiently sorting the different combinations of responses, where hundreds of pathological answers needed oversight or were poorly behaved for one reason or another. Rather than the existing practice of having to write one-time-use response codes, Shiny was able to efficiently handle each of these responses. The interactive review and adjustment of survey data proved invaluable in ensuring a smooth transition and more reliable results that required less manual handling.

2. Analytical Applications for TV Surveys 2014-2019

Citing the annual TV ratings data analysis, Eric discussed how he used Shiny to facilitate the development of the survey administration and analytical applications. There were challenges with corporate adoption, however, Shiny's flexibility allowed the organization to create highly customized tools that provided long-term value. While the initial reception was mixed, the adaptability of Shiny ultimately contributed to the successful launch of a product that continues to be used over a decade later.

3. Multi-Scale Applications 2019-2024

Shiny is now demonstrating its versatility in both proof-of-concept and enterprise-scale solutions. The teams at Warner Brothers Discovery are now able to quickly iterate on survey methodologies with real-time oversight and interactive adjustments that significantly improve efficiency and decision-making. This collaborative approach allows the creation of tailored applications that meet diverse business requirements and empowers end users, both technical and non-technical, to interact with data in meaningful ways.

Warner Brothers Discovery successfully integrated Shiny into their tech stack, however Eric attributes its success largely to the early involvement of subject matter experts in the development process.

Behind the scenes of a digital transformation. Discover how we helped a leading movie studio streamline operations with data-driven solutions.

Key Insights: Lessons from a decade of Shiny

In the keynote, Eric Kostello highlighted several key insights regarding Shiny's application in enterprise settings:

• Democratizing data science by empowering users through collaboration: Shiny bridges the gap between technical and non-technical users by allowing non-programmers to interact with data in meaningful ways. By involving end users early in the development process, you can create more effective solutions that align with business goals and operational needs.
• Minimum viable interactivity: Before investing too many resources, test your ideas using the simplest level of interactivity needed to validate the effectiveness of your application. Shiny's flexibility allows for the rapidly development of proof-of-concept applications with just enough interactive features to test ideas, gather feedback, and make data-driven decisions. This approach helps teams to iterate and refine their solutions in a way that ensures meeting user needs without overcomplicating the development process.
• Enterprise-class solutions: Tools like Rhino provide scalability and maintainability and make Shiny suitable for long-term enterprise deployments. Rhino ensures that Shiny applications are robust, secure, and easily integrated into existing IT ecosystems. Developers benefit from structured workflows and enhanced debugging tools, while management is given confidence in the reliability and scalability of data-driven solutions.
• Matching requirements: Choosing the right tool for the job is essential for your project's success. Shiny excels in scenarios that require interactive data exploration, rapid prototyping, and collaboration between technical and non-technical users. It is ideal for when organizations need to move beyond static reporting to more dynamic, data-driven decision-making processes. However, it can be worthwhile to use tools like Tableau or Power BI when only simple reporting is required. Recognizing the distinctions will help to maximize your investment and resources assigned to Shiny while ensuring your analytics strategy aligns with operational goals.

Challenges and Adoption

Despite its advantages, there are still barriers to Shiny's widespread adoption, such as organizational resistance and reluctance to change from committed users of established tools like Tableau and Power BI. Organizations can overcome these challenges by having power users embedded in the organization who will showcase Shiny's unique and powerful capabilities such as its ability to provide real-time data interactivity, integration with R's extensive statistical libraries, and its flexibility to build highly customized analytical applications. 

Choosing the right tech stack for decision support systems? Our blog breaks down when to use React, Python, and R for scaling effectively.

Conclusion

The insights from the keynote at ShinyConf24 clearly demonstrate how Shiny has become a critical tool in accessible, enterprise data science by providing powerful analytics capabilities and how it is evolving and adapting to the needs and demands of its users. Through both rapid prototyping and enterprise-scale deployments, its flexibility allows organizations to build interactive applications that drive real-time insights and informed decision-making, and its capabilities and accessibility are growing daily with the addition of frameworks like Rhino. As companies navigate and utilize increasingly complex data landscapes, and more teams and stakeholders are becoming involved and reliant on insights from data analysis, Shiny is a practical and effective solution for addressing organizations' analytical needs.

For more insights into Shiny's evolving role in enterprise settings, join us at ShinyConf25.