Analytics is becoming increasingly pervasive. Managers are now relying on data to drive decisions more than they ever have in the past. These changes have increased the complexity and importance of adhering to and understanding data visualization best practices.
One critical item to keep in mind when displaying data is Tufte’s Data-Ink Ratio. The data-ink ratio is the percentage of ink that is necessary to convey the meaning of the data. We want a high percentage of data-ink and a low percentage of non-data-ink.
Below is an example of a graphic with a low data-ink ratio. Notice the extra lines included in the graphic.
The second graphic has a much higher data-ink ratio. It is much easier to read and the reader is not distracted by non-data portions of the graphic.
Below is another example illustrating the differences between high and low data-ink ratio graphics.
Tufte refers to the following five principles related to data-ink:
As an example, the following 1 minute you-tube video gives a fantastic illustration of these principles. The video can be found here. The video is a very quick illustration of how to turn a diagram with a low data-ink ratio into one that displays the data very well.
Before moving on, consider what this graphic says.
Suppose the point of the graphic is to convey which stadium is the loudest. When you glance at the graphic, which stadium is the loudest and by how much (a lot or a little)? The takeaway to most casual observers is that Arrowhead Stadium in Kansas City is the loudest of the five by a fairly wide margin (look at the stadium pictures - it is a wide margin, clearly).
Below is the same graphic, transformed with the principles of Tufte, which seems to show a much different story.
Arrowhead Stadium is still the loudest - but by less than 10% over TurkTelekom Stadium in Istanbul. The extranious bells and whistles in the first graphic distort the message which should be that there is not much difference between the five loudest stadiums in the world.
Obviously, there are many opinions on what constitutes best practices. Use what follows as a guide.
When creating visualizations, consider the audience.
Do not create a visualization that meets everyone’s needs; but rather, consider the needs of a smaller group of stakeholders.
When selecting your graphic, consider why you are doing so.
Do not simply include a visualization without a goal in mind. Stakeholders need to be able to use your visualization to make decisions and in an appropriate time-frame to align with decision types. Once a visualization is clear, insights can be drawn and action can be taken.
Consider how your visualization will drive action. If there are thresholds that need to be achieved or comparisons to be made, how will you include those in your visualizations?
Accountants like tables - everyone else likes visuals. However, selecting the correct chart type is not trivial.
Data visualizations should tell the story clearly, avoiding distortions. Avoid the use of visual representations that do not accurately represent the data set, like pie charts in 3D.
Content marketers have long known that content needs to be snackable, and this applies to data too! If everyone is going to create their own data visualizations at your company, you’ll want them designed for maximum impact. When these type of design power practices are used, your improved visualizations will help various audiences to:
A good graphic does not need explanation. The graphic IS the explanation. If you need to add words to your slide in addition to your graphic, it is not a good graphic. Keeping the user engaged in the graphic will ultimately keep your client on topic and engaged with you.
Camm, Jeffrey D. Business Analytics. Third edition. Boston, MA, USA: Cengage, 2019.
Toptal Design Blog. “Data Visualization – Best Practices and Foundations.” Accessed April 29, 2021. Available here.
Tufte, Edward R. Envisioning Information. Fourteenth printing. Cheshire, Connecticut: Graphics Press, 2013.
Tufte, Edward R. Visual and Statistical Thinking: Displays of Evidence for Making Decision. 4. print. Cheshire, Conn: Graphics Press, 2008.
Tufte, Edward R. The Visual Display of Quantitative Data, Graphics Press, 1983.
Tufte, Edward Rolf. Visual Explanations: Images and Quantities, Evidence and Narrative. Cheshire: Graphics Press, 2019.
Mullet, Kevin and Sano, Darrel. Designing Visual Interfaces, Sun Microsystems, 1995.