With the growth of Business Intelligence (BI) and Analytics, dashboards are now pervasive and considered commonplace in the modern enterprise. We see them visualizing information for every type of organization, across every industry. They are embedded in websites, and they even (on occasion) make cameos in movies. Despite its ubiquity, the day of the dashboard as a static, stand-alone technology is quickly coming to an end. On the continuum between minds and machines, they are overly reliant on human inputs and are constrained by such elements. Yesterday’s dashboards are being upgraded with Artificial Intelligence and Machine Learning algorithms that transform them into dynamic tools—they are now able to use AI-powered recommendations to ensure that the intelligence being delivered to users across the entire enterprise is both relevant and actionable.
Recommendations Are Already Pervasive
Recommendations are already pervasive in our everyday lives—impacting everything from the way we select a movie to what gadgets we buy online. Netflix uses a complex Machine Learning algorithm to recommend new shows and movies based on user demographics and previously watched content. Travel apps like Hopper and Travelocity recommend the best time to purchase flights for the lowest fare. And the global leader in retail, Amazon, has integrated recommendations into nearly every aspect of the purchasing process. These types of recommendations work to personalize the consumer experience—improving engagement levels and boosting loyalty.
As this type of recommendation becomes more and more common, businesses are adopting AI-driven recommendations internally, too. For example, NBC uses algorithms to recommend not only which ads to show customers, but where to place them on a webpage. Another example of this type of enterprise recommendation is IBM’s Watson Marketing, which can use consumer demographic and social behavior data to recommend customer outreach programs for marketing departments.
Recommendations in BI?
Aside from Machine Learning applications, what can recommendations do for the BI and Analytics industry? For starters, it can speed up the process of finding insights and make creating dashboards and charts more accessible to every business user in an organization—even those without particular data skillsets. For example, recommendations implemented in a dashboard could immediately analyze the data and suggest suitable charts for visualizing important KPIs, all with zero clicks.
Natural language processing (NLP) can also be used to create recommendations. With Natural Language querying, users can ask questions about their data in dossiers and receive answers in the form of new visualizations. Analysts can also quickly gain insights through dossier recommendations, which suggest dossiers authored by other users within an organization who have used similar metrics and attributes in their dashboards.
Building on the momentum achieved so far, AI-powered improvements will only continue to revolutionize dashboarding and BI.