Business Intelligence Blog from arcplan

Building Your BI Dashboard Foundation: A Developer’s Guide


Over the last couple of years, the growing hype around big data analytics has rekindled interest in data visualization and a long-standing BI tool – the dashboard. BI dashboards are an avenue for presenting pertinent chunks of information at a glance, and in 2013, they can present more complex, insightful representations of data than ever before. But the key to making your dashboard usable and long-lasting is to keep it simple, organized, and intuitive.

In other words, the way to address the growing need for better analytics and visualization is not to build prettier, more animated charts. A solid foundation has to precede these bells and whistles. Here are 2 foundational dashboard elements that developers should master before moving on to the fun stuff:

1) The Path of Analysis
There are many ways to dissect and analyze data, but the first step is determining how to make the path of analysis most intuitive for users. Your options include:

  • Functional Breakdown: This is a very common way for arcplan customers to organize dashboard data. In this structure, sales reports, operations reports, or financial reports – along with their corresponding KPIs – are grouped together, providing a dedicated area for members of each department. If your organization is building a comprehensive dashboard application, a functional breakdown structure could be hidden behind a login; users would login separately and be able to access the information relevant to their position. Or, if your organization would rather open access to all reports for C-level users for example, then each department’s reports would be grouped together and accessible via the navigation. All areas of the dashboard app should have a similar design (4-6 charts/screen, corporate colors, etc.), making it easier for developers to maintain the overall application.
  • arcplan_summary-to-detail_sm

    Click to enlarge

    Classic Drilldown or Summary-to-Detail: See image to the right. Information is presented at the highest overview level, such as the United States region, with the ability to drill down into lower levels of granularity to see data by state, county, zip code, or individual store. Drilldown is often used with the functional breakdown structure – they are not mutually exclusive.
  • Reverse Hierarchy: It is technically possible to invert the classic drilldown structure, allowing users to view information from the lowest level of granularity and roll up to higher overview levels. This is an unusual and uncommon way to structure your path of analysis, but we have seen some users prefer to work this way. Though most of your users may take your recommendation for the path of analysis that works best, some users will find your way antithetical to their work style.

2) The Navigation Construct
Once you have chosen how to guide users through their data analysis, you must determine how they will navigate through the application.

  • Tab Navigation: One very popular option is a dynamic tab-based navigation, which presents report tabs corresponding to the users' level of access. For instance, if a user does not have access to finance reports, then he or she will not see the finance tab. Though this method is straightforward, effective and meets many of our design rules, one significant limitation is that you can run out of tab space, especially if your user has access to many KPIs.
  • arcplan_horizontal-nav_sm

    Click to enlage

    Horizontal Menus: When tab navigation does not suffice, horizontal menus with parent-child relationships provide supplemental navigation. Developers should not create more than 2 levels of horizontal menu breakdowns. Efficiency will go down the drain since few users will have the patience to find whatever is on sub-menus 3 through 7.
  • Vertical Menus: A vertical menu is a clever way to contain a lot of data using hierarchical trees. Keep in mind, however, that many BI practitioners are metric and KPI hoarders, reluctant to delete a KPI or metric even if it has not been used in ages; so after a while the vertical hierarchy set up can break down as well. The vertical orientation of the menu system takes up valuable real estate on the screen, but developers can compensate (to an extent) by using expanding and collapsing frames to make things work.
  • Search Capability: Users have come to expect a search function in every application, and BI is no exception. For an organization with thousands of reports, there’s no way a new employee without a personalized dashboard can figure out which 10 of them are important to their department. Search and bookmarking capabilities are critical new features of modern dashboard applications and their use may soon override the navigation structure you spent so much time developing.

At this stage of dashboard development, application designers would do well to remember that substance – not style – guides decision-making. You can make your dashboards pretty after you’ve mastered the basics.

Dwight deVera

About Dwight deVera

I'm Senior VP responsible for Solutions Delivery at arcplan in North America. I also present on a lot of arcplan webinars, so you can sign up to hear me - the events listing on our website is located here: You can also follow me on Twitter: @dwightdevera.
Comments (0) Trackbacks (0)

No comments yet.

Leave a comment

No trackbacks yet.