2020 is the year of the dashboard! This is not because dashboards are new, but rather their use has exploded as a result of the COVID-19 pandemic. The COVID-19 Dashboard by the Center for Systems Science and Engineering at Johns Hopkins University remains the most popular dashboard, recording more than 4.5 billion views in a single day.*
Based on the success of the Johns Hopkins University dashboard, COVID-19 dashboards have rapidly multiplied, as seen in the Centers for Disease Control, the Commonwealth of Virginia’s COVID-19 Data Insights, and the local Thomas Jefferson Hospital District dashboard.
Dashboards are not limited to epidemiology, but are common across a broad range of applications, including executive decision-making, sales, marketing, public safety, environmental modeling, election analysis, infrastructure monitoring, and social media assessment. Designing effective dashboards is a challenging task, requiring an understanding of visual theory, the design of charts and graphs, and the combination and coordination of multiple graphics.
This is the first of two Smart Cville posts on books about dashboards. This post will review Information Dashboard Design: Displaying data for at-a-glance monitoring, by Stephen Few. Few is a leading figure in data visualization, having published Show Me the Numbers: Designing Tables and Graphs to Enlighten, Now You See It: Simple Visualization Techniques for Quantitative Analysis, and Signal: Understanding What Matters in a World of Noise.
This book is highly recommended as a first read for dashboard design. Few methodically provides a complete education that covers the basics of visual perception, the construction of graphs, and the integration of multiple graphics in a single display. He builds each topic on a scientific foundation and illustrates them with examples, both good and bad. After finishing the book, the reader will have a solid understanding of dashboard design and an ability to assess the quality of any dashboard.
Few adopts a minimalist approach to design, favoring a few basic graph types and perception-based design recommendations. This approach leads to clear designs, but remains controversial among some modern designers who prefer eye-catching, interactive, and animated graphics.
Few defines the scope of the book early on, saying “A dashboard is a visual display of the most important information needed to achieve one or more objectives that has been consolidated on a single computer screen so it can be monitored at a glance.” The key point is that dashboards are single screen displays, designed for repeated monitoring, giving readers key information at a glance. He goes further, noting that dashboards are not reports, exploratory data analysis tools, portals, or scorecards.
Few briefly reviews the history of dashboards for visualization, noting the early use of car dashboards as the metaphor for early data dashboards. He quickly dismisses designs that mimic car dashboards, especially those that use dials and gauges, primarily due to their inefficient use of space.
Few breaks down the process of dashboard design, starting with the requirements assessment. He then introduces the basics of visual perception. After reviewing the limits of memory and the importance of rapid perception, he describes the visual variables and Gestalt principles of visual perception. He builds on this in an appropriately named chapter titled “Achieving Eloquence Through Simplicity” that sums up his minimalist philosophy. The two key objectives are to 1) remove the non-data pixels and 2) enhance the data pixels.
Having covered visual perception, Few outlines his “ideal library of graphs.” This includes the traditional bar graphs, dot plots, line graphs, box plots, scatter plots, spatial maps, as well as the less common tree maps and heat maps. Two innovative graph types receive special treatment in chapters of their own: Edward Tufte’s sparklines and Few’s own bullet graphs.
Few then addresses the effective integration of multiple graphs in a single display. He covers the most important practices for integrating displays, including organizing information by importance, keeping related information together, maintaining consistency, and preventing excessive alerts.
Few puts combines his guidance when reviewing the strengths of the top results from an education design competition. He addresses the deficiencies of other contributors and submits his own design. He continues with examples of his own design, including dashboards for sales, a chief information officer (CIO), telesales, marketing analysis, and service installation. He finishes by demonstrating “good” examples by others.
Few’s book provides a comprehensive foundation for understanding dashboards. The one limitation is that the most recent edition of the book was published in 2012, so it includes outdated examples of the work of others and does not incorporate the latest innovations in graphic design. Balancing this limitation, Few’s own designs in the book withstand the test of time and he avoids significantly dating the book by remaining software agnostic. That said, a revised and updated volume would be timely and welcome.
Our next post will look at The Big Book of Dashboards: Visualizing Your Data Using Real-World Business Scenarios, by Steve Wexler, Jeffrey Shaffer, and Andy Cotgreave.
References
“Hopkins coronavirus-tracking map is the key source for governments and the public”, Peter Novello And Shivni Patel, May 3, 2020, The Johns Hopkins Newsletter, https://www.jhunewsletter.com/article/2020/05/hopkins-coronavirus-tracking-map-is-the-key-source-for-governments-and-the-public
Resources
Information Dashboard Design: Displaying data for at-a-glance monitoring book.
Stephen Few website.
Perceptual Edge website. Visual Business Intelligence for enlightening analysis and communication