Dashboards – a term that comes up a lot in conversations about Business Intelligence. I hear about them a lot and was asked recently to talk about dashboards at a conference called SAP Insider in Singapore. What I find interesting about dashboards is the diversity of use case that people apply the term towards, and how the term is interpreted differently. So for my session, I tried to decompose the dashboard into the actual use cases. I came up with four, but if you know of any others, leave a comment as I would enjoy hearing about them.
My four are these:
This is typically focused on data that is constantly changing (i.e. not quarterly financials). To me, this use case best follows the source of the term, your car’s dashboard. A monitoring dashboard aims to provide only the key information that you need in a single compressed view. It should be personal and contain just those elements and visuals that are useful to you at a given moment and continuously update. Interestingly a monitoring dashboard will typically contain content from multiple sources and the various items do not need to be related (in fact they probably shouldn’t be related). In a monitoring dashboard the most critical thing is that it all be visible in a single glance, and what is most important is most prominent. Think every network or operations center in the world.
2) Analytical Package
We still see people providing contextually rich collections of information out to others. Today, these are more often called “infographics” or “briefing books”, but are also called dashboards. They contain relatively static collections of data and are delivered to the user but provide that user with dynamic interactions that allow them to press buttons or move dials around – hence the name Dashboard. This is the usual vehicle for quarterly or periodic data that someone has enriched visually to provide a more attractive and exciting experience. The requests for this use case have dropped off, but with the advent of mobility, I suspect these will return as the ultimate presentation shifts from being a slideshow to interactive stories and videos.
3) Analytical Applications
The most common use of the term dashboard today is clearly the custom analytical application. In the Analytical Application, someone has designed a User Experience that has controls to allow the user to navigate around data, drill down to details, and change selection criteria. The ability to design the experience and to provide very pretty visuals seems to be the criteria to differentiate from a report to a dashboard, but one that is very important when gaining adoption and acceptance from user communities. This remains the biggest request from IT organizations looking to make their data highly accessible and fit in their own design and user experience goals.
4) Wall Displays
A growing use case and one that is often most similar to (1) is the wall display. These also show data that is constantly updated, but are more informational. The wall is best exemplified by the flight board at every airport, showing which gate your flight leaves from. The wall display is usually targeting a very broad community with key information they may use. I have customers asking for everything from vehicle tracking, to call center performance, or rolling infographics to be delivered in this form. Today, we are delivering these not just to a wall, but to a mobile device (your personal wall).
There are of course, enormous nuances around all four of these use cases, such as:
· Who should create them
· The sources of information they can connect to
· The amount of information they should and can contain
· The amount of semantic richness that should surround them
· How they should be secured
One thing is clear to me, when someone tells me they want dashboards, I know I have a lot more to understand before I really know what they are looking for. With the most comprehensive BI Suite on the market, the other thing I can say is that I know I have a way to deliver it to them.
My presentation from SAP Insider 2012 Singapore can be found here