Best Practice for Dashboard Design
When you think about dashboarding, you think about a sleek, colourful webpage that visualizes the current condition of your organization [WC1] with up-to-date data. A common mistake when designing a company dashboard is wanting to put too much information on one dashboard. The result is a mix ‘n match of data containing different levels of granularity. This is hard to navigate and does not give you the desired insights in an efficient way. That is why we work according to the DAR principle.
The DAR principle is a methodology for creating an effective dashboarding application, whilst still giving you room to be creative. DAR stands for Dashboarding, Analysis and Reporting. The idea of DAR is that every page contains a different level of granularity. The dashboard page contains high-level information, whilst the analysis pages visualize more complex and detailed information per business topic. Finally, the reporting pages provide a deep level of granularity to explore every detail. This way, you can easily navigate you company data and take grounded actions when needed.
The dashboard page functions as a table of contents, as it provides an overview of the most important information about the organization. This page allows the user to scan for status updates in one or two minutes. Based on what you see on this page, you head to the analysis and reporting pages to get more detailed insights.
To attain this functionality, the information on the dashboard should be general and high level. As a result, there is only a limited amount of KPI’s and filtering options. Furthermore, there should be a hierarchy to your information to enhance scanning. Important information is visualized in a larger font and in a more prominent place. Ideally, this page should be entirely visible on a desktop monitor.
Figure 1: Example of Dashboard page. (Source: http://princetechsolutions.com/dar-methodology-of-qlik-sense/)
The analysis pages allow the user to explore the data and to find answers to questions that have arisen on the dashboard page. The analysis can consist of multiple pages that scroll vertically, each focussing on a particular business topic. This allows the user to conduct a structured and targeted search.
To facilitate this search, these pages are more interactive and provide more filtering options. There is also a strong focus on guided analytics, as more complex interactive visual interfaces are introduced. Usually, a user spends more time here.
Figure 2: Example of Analysis page. (Source: http://princetechsolutions.com/dar-methodology-of-qlik-sense/)
The reporting pages are the third component of the DAR principle. It is this part of the dashboard that should ultimately lead to action. Here, we can find tabular data consisting of the most granular information. There are a lot of sorting and filtering options to get as much details as possible. These details will be the one that will help the user taking different actions at an operational level.
It is often said that you can’t manage what you can’t measure. The reporting dashboard will help the user to stay in control of his/her business at all times.
Figure 3: Example of Reporting page. (Source: http://princetechsolutions.com/dar-methodology-of-qlik-sense/)
Lately, there has been an increasing focus for data analytics applications to provide ‘actionable insights’. We could rethink the DAR acronym and use the R for Recommendation.
This recommendation will help the user acknowledge a problem and analyse the cause of it. Once this analysis is done, it becomes possible to recommend a concrete action that could solve the problem by dealing with his identified causes. In the end, the user would get some more useful information.
Now that we identified the different components of a dashboard, it is important to note that DAR isn’t a linear process. The user can navigate through the pages and come back to the previous ones at any time. Moreover, once some actions are taken, it is important to review the information on the dashboard, see what changed and start the loop again.
QlikView. (2013, November). Dashboard, Analysis, Reporting (DAR)