Part 1: Understanding the Potential of Immersive Analytics (IA) in BI Tools
There will come a time when you can simply ask an analytics application, “What was my company’s revenue for the last three quarters?” And, instead of complex charts and tables, the application will learn the context of your query and simply retrieve the exact figures for your review.
While we are yet to reach this state, ensuring good user experience (UX) is critical in today’s data-driven, digitized economy. In reality, however, almost one-third of US analytics users find it difficult to use business intelligence (BI) tools, citing complicated interfaces.
As advances in the field of Immersive Analytics (IA) continues, these challenges will soon become a thing of the past. Just like Amazon’s Echo has simplified the shopping experience with the power of voice commands, a Natural Language Processing (NLP) engine can form and infer semantic relations from verbal inputs—allowing end users to interact with the BI applications in the mode that is most natural for humans. In fact, the demand for platform-agnostic BI tools that use innovative IA technologies is high—47% of executives are interested in accessing analyticsvia voice-activated virtual assistants.
Making the Case for Immersive Analytics (IA)
While IA is set to firmly establish itself as a default interface, it’s important to understand this from the perspective of end users, business operations, and the overall technology landscape. The current omni-channel mode of interaction demands IA be a de-facto offering at minimal cost while ensuring zero-compromise on data security and confidentiality. Users can interact with their applications anytime, anywhere, in their preferred mode—voice, text, gestures, or interactions — enabled by Virtual and Mixed Realities (VR/MR). In turn, this will improve enterprise-wide productivity through self-service and remove the need for extensive training.
Such a solution will also exist as an enabler and should be implemented after considering its utility to the end user. Enterprises must identify applications that can seamlessly integrate with an IA interface in their as-is built state, without the need for over engineering. At the same time, it is important to pinpoint and assess applications that need enhancements to prepare for such an interface. When deploying IA, it’s critical to test and validate if the system can accurately capture and translate interface actions to a language understood by the application—for continually delivering (near) real-time responses. And, these interactions should not just be unimodal, but allow end users to seamlessly toggle between multiple modes. For instance, while using a text-based chatbot, one can interact using voice and switch to gestures as required with the visualized output.
Preparing for the Road Ahead
To build a cohesive IA implementation roadmap, businesses will need to decide if they want to develop these capabilities as part of their value proposition or invest per market demands.
The more pertinent question that should follow is – ‘Will my enterprise’s current IT infrastructure be able to support this technology?’ This also means correctly assessing if the necessary investments have been made, and skill sets are available for sustaining such a program. Before introducing new IA for analytics tools, businesses must also consider the potential ROI for doing so. Finally, the system needs to be continuously improved so that IA is institutionalized across the organization.
But, how will enterprises achieve this while ensuring business as usual? Please share your views in the comments section below, and watch out for the next episode in the series, as we breakdown the solution schematics.