Designing an effective dashboard requires focus on certain aspects which, if overlooked, could render the tool useless. Understanding that cosmetics are just as important as content in some cases, Faithful+Gould’s Consult Group examined how human elements come into play when designing dashboards. During the study, we discovered several gaps:
- The lack of focus on overall layout
- Showcasing the wrong or irrelevant data
- Improper use of effective graphics or visualizations
- Representing too much data instead of information, reports acting as dashboards
- Not considering the element of human perception
Understanding the purpose of a dashboard versus a report is a good place to start closing the gaps. With rows and columns of data, reports dive into a low level of detail, are typically static and may include some graphics. They usually require consumers to apply their own judgment and/or calculations to get the information they want. They are a snapshot in time and are typically not “live.” Dashboards, on the other hand, are meant to provide the consumer of the data instant views of what’s going well, what to keep an eye on, or where to focus and take immediate action. If the consumer sees areas of concern, the data can then be interrogated at a lower level, either in sub-dashboards or from the source data itself.
A great solution to the wrong problem will always fail!
Before drawing a single pixel for your dashboard, you must understand and define the problem you’re trying to solve and for whom. A common approach to this is developing user stories. A user story is an actionable statement comprised of three components—the user, the need and the goal. It summarizes who the user is, what the needs are, and identify why the needs are important to that user. User stories not only help identify the requirements and set the stage for the design process, but they also provide criteria to measure the effectiveness of the final product. It’s tempting to jump right in and start building a dashboard; however, this can lead to wasted energy and effort if the design, elements and data are incorrect and need constant revision. After we understand the purpose and scope of a typical dashboard, applying that to the design is the next step, with several more factors to consider.
How will you present the data?
How to present the data may be the most challenging part of dashboard design. Choosing the wrong type of graphic or visualization could lead to confusion and misinterpretation of data. Pie charts are popular, but they’re not always the best choice to deliver the intended message, like effectively communicating quantitative comparisons, for example. A bar chart would be the better choice.
To see which one works best for the intended purpose, run sample data through various types of charting methods – column vs. pie, bubble vs. scatter, list vs. tree map. The choices and combinations are numerous, so it’s important to fully vet how the information will be presented and stick with that format for similar designs.
What goes on the dashboard? And where?
Define the layout and prioritization of the dashboard elements. Most Western cultures read from left to right, top to bottom. Also, our eyes are naturally drawn to white space. Understanding how people consume data is an important step in designing your dashboards. Just as important as the layout is the positioning of the data. Try to position the most important information top left and cascade the data across and down according to importance or corporate interests. Apply the use of conditional formatting to your metrics/information, or use automated icons driven by carefully considered thresholds to draw the eye to areas that require attention. However, be careful to avoid the temptation to place too many elements on a single dashboard. Doing so will lead to difficulty presenting the most important information and will quickly diminish its effectiveness. Brainstorming this with the consumers of the data will go a long way in helping to determine the positioning of the dashboard elements.
What is the overall look?
Keep it simple and keep your audience in mind. We know that a dashboard is meant to provide the user with real, relevant and up-to-date information. Avoid forcing the user to drill-down into the data as much as possible. This generally applies more to operational, executive and tactical dashboards, but is a good rule-of-thumb in all designs. Try to use graphics over tables of data. Choose a color scheme that is easy on the eyes and not too overwhelming. Be aware of the importance of white space.
NOW you can start building!
Go through the proper steps to develop an effective dashboard, taking the time to identify the requirements, understand the context, select the proper elements and data, and use effective visualizations. Without these steps, you could potentially deliver useless information and resources that provide little benefit. With the steps, you can create a valuable resource and equip users and clients with the right tools to quickly assess information from real-time data so they can make timely and accurate business decisions, which ultimately improves project performance. When we help our clients succeed, we establish ourselves as a trusted partner.
Tim Haar is a senior project manager in Faithful+Gould’s Consult Services Group which provides industry-leading dashboard design and implementation as part of our comprehensive PMO Implementation service line. Tim focuses primarily on project controls for civil construction projects in excess of $750 million. He is a certified Project Management Professional and is passionate about data and the use of technology to enhance the delivery of work through developing more useful and accurate information, streamlining and automating processes and team collaboration.