The Future of Business Intelligence: Invisible Analytics and Embedded Insights

The Future of Business Intelligence: Invisible Analytics and Embedded Insights

sisense logo The best analytics are invisible. Now, there’s a whole industry devoted to creating compelling visualizations meant to take complex, dynamically unfolding events and making them instantly comprehensible and insightful. These visualizations are beautiful but, alas, they represent a creative workaround for technology’s inability to deliver what we really want – an answer, right when we need it.

Analytical dashboards are designed to get you the answers to your pressing questions as fast as possible. No matter how well they’re designed, they have a number of important drawbacks. First, they take time to interpret, meaning they only appeal to a subset of users and personas that have the mandate to uncover insights and make strategic decisions.

Second, and more importantly, they require the individual to switch away from whatever they’re working on to use them, unless the analytics application itself is the core workflow.

Now, analytics are augmenting the dashboards experience with analytics which are embedded into workflows, allowing people to get answers at the moment that they’re needed, and to act on them directly. These embedded analytics are changing the way that people work, and are the first important step towards what I call “Invisible Analytics.”

What Embedded Looks like Now

Embedded analytics are usually created to be a seamless part of the workflow of the client, or the customer of the client. This is a contrast to traditional standalone dashboards which are “swivel-chair” tools, that is, information presented outside of the usual workflow, that require the user to consult a different screen or app, before turning back to their email, phone, or workspace to act on the insights that they receive.

Those embedded insights aren’t just for office workers. They need to go where the user is, no matter what kind of user it is. Employees on the road will get them via email or text, nurses will receive them on screens at bedside, Salespeople will get them through their CRM, supply chain managers will get them through their ERPs… you get the idea.

The insights that are delivered differ by company and role. However, one common element between a lot of the embedded use-cases that we see are feedback mechanisms for users on how the aggregate of users are acting on the same tools, (e.g. “Other users viewing this information are achieving an average cost of $.23 per unit.” for someone in an operations role).

Recommended: What is Deloitte’s New Public Data Visualization Tool All About?

Amazon-Like Embedding and AI

Much the way that Amazon.com can tell you what other consumers like you are interested in – different versions of the same product you’re looking at or complementary products – embedded analytics can give you an idea of better ways to complete your task, other tasks that you can jump on soon after, or how your performance in a task stacks up against similar users in the firm.

This Amazon-like benchmarking has been shown to raise individual contributor performance significantly. Going beyond the benchmarks – with a little bit of Artificial Intelligence to analyze the aggregate data on the fly – they can be used to reveal best practices as well, (e.g. “Did you know that 80 percent of the people who are achieving higher Sales than you are making their calls before 10:30 am each morning.”). This combination of analytics and AI have led to some interesting applications.

In Healthcare, Machine Learning is already being used in places like imaging to determine things like skin cancer with greater accuracy than can be achieved with a human professional – even one with decades of diagnostic experience.

This type of Artificial Intelligence can be augmented with analytics for cases which are more difficult, even for AI to diagnose. For example, infant brain tumors look very similar to naturally occurring structures in the brain, to human eyes or even Machine Learning. However, when this AI is positioned with a mashup of other data, the quality of the predictions go way up.

Answers at the Point of Decision

Invisible Analytics are now science fact, not science fiction. To be sure, while the applications for this technology are being explored rigorously in the healthcare space, other industries have been slow to realize its potential. The builders of analytical apps are now just starting to leverage embedded analytics to deliver key insights. Many of the business users that they’re working with are getting these insights and acting on them, without ever seeing a dashboard or a chart.

The bottom line is that very few workers want to depart from their daily workflows to look at a dashboard. They want insights when they need them, how they want to receive them, embedded into the workflows they are already using so they can act on them immediately. Analysis that is never viewed, and never acted upon, doesn’t add any value.

Still, most firms that adopt a data-centric approach are seeing the benefits of better, faster decisions by their staff, and increased productivity. For these firms, invisible analytics will present an even bigger leap forward, ensuring that important insights are never ignored and never overlooked, by the users who can benefit from them.

Read more: The 3 A.I. Scenarios: Artificial Intelligence, Augmented Intelligence, and Intelligent Automation

Previous ArticleNext Article

Leave a Reply

Your email address will not be published. Required fields are marked *