What? There's a difference, you say?
Although there are plenty of people out there who would use the words "dashboard" and "visualization" to describe the same thing, that doesn't mean they're necessarily right in doing so. It's a mismatch that I've consistently stumbled upon while reading articles about data analysis, and I'm putting my foot down.
"A dashboard is a collection of resources assembled to create a single unified visual display."
What's the difference?
First off, it's important to distinguish the disparate use cases of the phrase "data visualization." On one hand, the term can refer to the concept of using images to convey intelligence. On the other hand, it can also be used to describe a single representation of multiple sets of refined information, such as a bar graph that deduces how many widgets a factory produced over a given period of time. For the purposes of this post, we'll use the latter definition when discussing the technology.
Now that we've made that clarification, what's a dashboard? Smart Data Collective's Keyan Keihani identified a dashboard as a collection of resources assembled to create a single unified visual display.
Bouncing off the factory example cited above, let's say you also had three other data visualizations, each one providing insight into a different facet of the facility's operations over the course of a year – personnel productivity, equipment malfunctions and energy usage. These four charts would then assemble to create a dashboard. Whether these visualizations are presented as scatter plots, bar graphs, cartograms or circumplex charts depends on three factors:
- The type of data collected
- The way in which the data is manipulated (whether it is reacting to time, tasks, events, etc.)
- The manner in which your team best absorbs information
What makes a good dashboard?
A team of neuroscientists at the Massachusetts Institute of Technology discovered that the human brain can register and process multiple images in 13 milliseconds. For those of you who aren't too familiar with measurements, that's incredibly fast.
"Studies have shown that people tend to naturally direct their eyes toward the intersections or four corners of an image as opposed to the center."
Given the fact that people viewing your dashboards will likely be looking at them for longer than 13 milliseconds, the amount of information they'll be able to derive in a short amount of time is astounding – provided you organize the visualizations appropriately.
The Rule of Thirds
Yes, positioning matters. If any of you have been fortunate enough to receive schooling in the art of photography or film, you're probably familiar with the Rule of Thirds. According to Digital Photography School, the concept refers to breaking down an image into nine equal parts – it kind of looks like a tic-tac-toe board. Near the center are four intersections where the lines across the image overlap, which indicate potential points of interest.
The theory suggests that placing points of interest along these intersections will attract attention from viewers. Studies have shown that people tend to naturally direct their eyes toward the intersections or four corners of an image as opposed to the center.
Filmmakers sometimes employ the Rule of Thirds when trying to contrast power between two characters. For example, if Actor A is placed in the top left of a shot and Actor B is positioned at the bottom right, cinematically, it's proposed that the character whom Actor A is portraying has dominance in a scene.
Conveying information through dashboards
Now, back to dashboards. If there's one particular visualization you would like your audience to weigh with more consideration than the others, placing it at the top right or top left corner of the dashboard will likely draw attention to it, especially if it's sized appropriately.
As I mentioned earlier, the type of visualizations you employ is likely to have a profound impact on how people perceive the intelligence within your dashboard. Delving into this particular topic deserves another blog, so the subject will be discussed in a future post.