When it comes right down to it, you can amass all the data in the world, create oodles of visualizations that reveal scads of valuable information, and still not have a clear understanding of what it all means. Why? Because data visualizations, no matter how beautiful or powerful, don’t really have meaning unless they tell a story. But how do you tell that story?
Author of Storytelling with Data and speaker Cole Nussbaumer Knaflic gives insight into how to build a narrative around data during her Skype presentation at the Eastside Tableau User Group (ETUG). In what was the most interesting discussion of charts and graphs I’ve ever experienced, she broke down the way that people understand information and walked the group through examples of data transformation that adds real value.
Principles of Visual Perception
To begin, Cole covered the basic principles of how human brains interpret visual information:
- Proximity – elements which are closer together will be seen as belonging together.
- Similarity – elements which share visual characteristics such as shape, size, color, texture, value or orientation will be seen as belonging together.
- Enclosure – elements which share an enclosed space are seen as belonging together.
- Closure – when a figure is incomplete or a space is not completely enclosed, people perceive the whole by filling in the missing information.
- Continuity – elements arranged on a line or curve are seen as more related than elements not on the line or curve.
- Connection – elements x that are visually connected are perceived as more related than elements with no connection.
Knowing how the mind perceives and processes visual information will help you build more intuitive and useful visualizations. After briefly illustrating each principle, Cole dove into how best to apply them.
Crafting a Data Narrative
The easiest way to coax a story from your data is to eliminate the noise. In short, you remove anything that doesn’t add information but does add confusion. This is also thought of as improving the data-ink ratio. Cole approached this challenge in five ways:
- Leverage How People See – In short, take the principles of visual perception and apply them thoughtfully to your viz. Consider what story you want to tell and present only the essential information in a way that makes sense.
- Employ Visual Order – Avoid unnecessary design elements, such as large blocks of text, multiple font sizes or weights, too little white space, or overuse of multiple colors. Be mindful of the aesthetic design of your viz.
- Create Clear Contrast – As the saying goes, “it’s easy to spot a hawk in a sky full of pigeons.” If you want to call attention to something, make it stand out from other elements. For example, in a graph showing the value of multiple companies over time, show your company’s data in blue, and the competitors’ in grey.
- Don’t Overcomplicate – Keep things simple: make sure all the text on your viz is legible, keep the space tidy and uncluttered, use straightforward language, and remove anything that adds unneeded complexity. Minimalists, rejoice!
- Strip Down & Build Up – Sometimes the best thing you can do to improve your visualization is to start over. Give yourself a blank canvas and be thoughtful as you add information back to the page, following all of the above guidance and holding your narrative at the center of everything.
Take it to Tableau
Cole spoke about data visualizations in general, using examples of graphs created in Excel, but the basic principles apply to anything you create in Tableau, as well. To illustrate this, blogger and Tableau whiz Ben Jones re-created Cole’s “strip down and build up” example using Tableau instead of Excel. You can see what he did on his blog, DataRemixed.
If you need help getting started with Tableau and creating your own powerful visual stories, consider attending contacting our team to help you get settled, in training or to answer questions.
Written by: Sarah Reebs