AAD - Stories and Analytics Applications

The Dichotomy of Data Visualization: Stories vs. Analytic Applications in Modern Business Intelligence Platforms

Abstract:

Modern business intelligence (BI) platforms offer a spectrum of tools for data visualization and analysis, ranging from user-friendly, self-service "stories" to highly customizable, script-driven "analytic applications." This paper explores the fundamental differences between these two approaches, focusing on their respective strengths, limitations, and the underlying design philosophies that necessitate their coexistence. Drawing upon the SAP Analytics Cloud platform as a representative example, we examine the role of scripting, customization, and user experience in shaping the landscape of contemporary data analytics.

1. Introduction:

The proliferation of data in contemporary business environments has driven the evolution of BI platforms to provide increasingly sophisticated tools for data visualization and analysis. While user-friendly, self-service interfaces have empowered a broader audience to engage with data, the need for highly customized, complex applications persists. This has led to the development of distinct artifact types, such as "stories" and "analytic applications," each designed to cater to specific user needs and technical capabilities.

2. Defining Stories and Analytic Applications:

2.1 Stories:

Stories represent a self-service approach to data visualization. They are designed for business users with limited technical expertise, providing a guided, configuration-driven environment for creating dashboards and reports. The focus is on ease of use, consistency, and preventing users from creating non-functional content.

2.2 Analytic Applications:

Analytic applications, on the other hand, offer a high degree of customization through scripting. They are designed for developers and power users who require complex logic, advanced interactivity, and tailored user experiences. The "Analytics Designer" environment provides the tools to define data models, design layouts, configure widgets, and implement custom logic using JavaScript.

3. Key Differences and Rationales:

3.1 Customization and Flexibility:

The primary distinction lies in the level of customization. Stories are limited to pre-defined configurations, ensuring consistency and preventing errors. Analytic applications, leveraging scripting, enable developers to implement virtually any custom behavior, albeit with a higher risk of introducing errors.

3.2 Scripting and Logic:

Analytic applications are characterized by their ability to incorporate custom logic through scripting. This allows for complex event handling, data manipulation, and user interactions that are beyond the capabilities of stories.

3.3 User Experience and Consumption:

Despite the differences in their creation process, stories and analytic applications aim for a seamless user experience. Ideally, consumers should not be able to discern between the two, as both should present a cohesive and intuitive interface.

3.4 Design Philosophy and User Expectations:

The coexistence of stories and analytic applications reflects the divergent expectations of their respective user bases. Stories cater to business users who prioritize ease of use and guided workflows, while analytic applications serve developers and power users who demand flexibility and control.

3.5 Lifecycle Management:

Analytic applications possess a distinct lifecycle, including startup events and other programmatic control points. Stories, in contrast, operate on a simpler edit-view paradigm.

4. The Analytic Application Development Workflow:

The creation of an analytic application typically follows a data-driven workflow:

• Model Selection: Identifying the underlying data models.
• Widget Placement: Adding tables, charts, and control elements.
• Configuration: Arranging and configuring widgets.
• Scripting: Implementing custom logic through event handlers.

5. Typical Analytic Application Archetypes:

• Table-centric Data Visualization: Applications focused on detailed data exploration through interactive tables.
• Dashboards: Static or interactive overviews of key performance indicators (KPIs).
• Generic Applications: Reusable applications that can be adapted to various data models.

6. Scripting in Analytic Applications:

Analytic applications utilize JavaScript for scripting, enabling developers to implement custom logic in response to widget events. The platform often provides type systems and value help to enhance the development experience and ensure data integrity.

6.1 Type Systems and Validation:

The inclusion of type systems allows for enhanced script validation, ensuring that data types and values align with the underlying data model. This reduces errors and improves the robustness of the application.

7. Conclusion:

The dual paradigm of stories and analytic applications in modern BI platforms addresses the diverse needs of users, from business analysts seeking quick insights to developers demanding advanced customization. While stories provide a user-friendly, self-service environment, analytic applications offer the flexibility and control necessary for complex data visualization and analysis. The effective utilization of both approaches is crucial for maximizing the value of data within an organization. Future research may focus on the development of hybrid approaches that combine the ease of use of stories with the flexibility of analytic applications.