Every moment counts. Look around you. We live in a fiercely competitive, interconnected, big data world where speed to insight and action have become more critical than ever before. If you are a decision maker, you might not have the luxury of relying on dated, stale, static dashboards that can only get refreshed periodically.

Historically when a project sponsor asked me for real-time reporting, I would ask them to clarify what real-time meant to them, what was the true information need and level set expectations with regards to what existing technology could reasonably deliver. Often we compromised on a near real-time or a right-time dashboard that directly connected to a relational database or used a ROLAP (relational online analytical processing) solution. Today real-time is easy to deliver if your dashboard platform is capable of rendering live visualizations. Note most dashboard solutions in the market actually can’t handle it. They require adding JavaScript refresh snippets or other programmatic tricks to update the dashboard visual display.

Two weeks ago I delivered a real-time dashboards session at INFORMS Analytics 2015. INFORMS is the Institute for Operations Research and the Management Sciences. It is the largest society in the world for professionals in the field of operations research (O.R.), management science, and analytics. If you are not familiar with INFORMS, you should review their web site and plethora of fantastic resources.

520Sample dashboard built by Microsoft engineering

In that presentation, we walked through real-world use cases and shared who is likely deploying real-time projects. Typically we see real-time needed in the areas of operations, healthcare, crime, support, sales, IT, marketing and finance where timely decisions are crucial.

I also shared evolving modernized data platforms, technical solution architecture design patterns and common technologies being used in real-time analytics today. Topics included:

  • Changes to data warehousing patterns using both Hadoop and massively parallel database systems
  • Modern ingest and load versus extract, transform and load techniques with event hubs, Kakfa, message queues and Apache Storm
  • Best practices, tips and patterns of sending data to multiple destinations
  • Streaming data sources with Azure Stream Analytics and using temporal queries
  • Creating a real-time Power BI dashboard with Power BI REST APIs and SignalR

Feel free to peruse my SlideShare to dig much deeper into this super hot topic.


How To Build a Real-Time Dashboard

To get started building a real-time dashboard yourself, check out this excellent article by Jeff Stokes that was posted earlier this week during the Microsoft //build/ conference. The recently added Power BI output destination for Azure Stream Analytics makes development a breeze these days.


Contemplating the future

Seeing so many solutions get automated or become trivial to build like this real-time dashboard with live streaming data sources that used to be a technical feat for an entire team of BI and IT pros, I am actively exploring what else is out there to prepare for the future. I simply don’t feel complacent in the business intelligence and analytics space today…not even for one millisecond. The great news is that there are many new, amazing areas of technology being introduced into our data ecosystem and the world around us. The future possibilities are endless once you start looking beyond your current subject matter expertise.