Conceptualizing an effective business dashboard
Conceptualizing a business dashboard is a complex process. It requires in-depth knowledge of the business processes involved and deep technical knowledge of how to go about implementing it too. But another aspect that is often overlooked is an understanding of how the end user will interact with the dashboard. A poor understanding of that leads to poorly designed dashboards, which hinders decision-making, and kills the very purpose of the dashboard. With this article, we aim to equip you with a basic understanding of how end users interact with a dashboard and more generally, how you can conceptualize your business dashboards better.
What is a business dashboard?
Definition: A business dashboard is a multi-layered application, built on a business intelligence and data integration infrastructure that allows organizations to monitor, analyze and manage business process more effectively.
This definition suggests that a dashboard is an application that is wired to the organization's database/ERP system from where it draws data and presents it in a meaningful way. The fact that a dashboard is multi-layered implies that it presents different hierarchical views of data. Using a dashboard, executives can monitor, analyze and manage a business process.
How do people interact with a dashboard?
Dashboards are all about visualizing data. According to Wayne Eckerson, the director of research and services for TDWI (The Data Warehousing Institute) and a leading authority on business dashboarding, business dashboards should allow visualization of data at three levels:
- Summarized view
- Multi-dimensional view
- Detailed view
Prior to explaining the three layers of data presentation, we’ll give you an insight into how requirement of data varies depending on the circumstance. In fact, the need for layered presentation of data is born out of the fact that people don’t want all the data at one time. What they do want is the right piece of information – just when it is needed.
As suggested in the definition above, a business dashboard should let users monitor, analyze and manage business processes. The summarized graphical view of the dashboard allows a user to monitor a business process. In case there is a problem in the process flow, the user drills down to the next level which is the multi-dimensional layer. The multi-dimensional layer provides a much more detailed view of data, thereby allowing the user to determine the cause of the problem. Sometimes the cause of the problem cannot be detected even in the multi-dimensional layer – so the user has to drill-down further to the detailed view layer. The detailed view layer provides details of individual transactions from where the user can easily identify the fundamental cause of the problem.
Architecture of a Dashboard
While building a dashboard application, planners and developers should establish well-defined levels of abstraction with respect to data. This not only makes the application user-friendly, but also makes it deployable at all levels of organizational hierarchy.
As discussed previously in conjunction with end user interaction with dashboards, a dashboard should present data across three levels, namely summarized view, multi-dimensional view and detailed view.
The design specifications of each data presentation layer of the dashboard are:
- Summarized view: This is the topmost layer of a business dashboard and it should display essential KPI's graphically (i.e. using graphs or gauges). The layer must also have an in-built control mechanism that triggers an alert when a KPI exceeds or drops below the normal value range.
- Multi-dimensional view: This layer should supplement the metrics displayed in the top layer with additional data. Analytical tools must be built into this layer in order to allow users to perform computational analysis on data.
- Detailed view: This layer should facilitate users to view reports pertaining to individual transactions (e.g. invoices, shipments etc).