The ZingChart team recently read Win with Advanced Business Analytics: Creating Business Value from Your Data, by Jean-Paul Isson and Jesse Harriott. The book was useful for understanding use-cases for analytics in a variety of job roles and industries.
But we especially liked the CONVINCE acronym they used to ensure their data visualizations adhered to principles of usability and had maximum effectiveness. We will outline some very effective principles for effective data visualization for you!
CONVINCE: Principles for Effective Data Visualization
According to the authors, the CONVINCE dataviz principles are:
C onvey Meaning
V isual Honesty
I magine the Audience
E ncourage Interaction
At ZingChart, we constantly scour the web for dataviz inspiration and data visualizations that could be improved with our library. But it is surprising how many charts, graphs, and dashboards lack an obvious purpose.
Lesson: Use titles. Good ones. And, remember to label things.
A simple example shows the power of titles and axis labels to make it clear to the user why they are viewing this piano chart.
Some of our favorite authors on data visualization are Stephen Few and Alberto Cairo. Sometimes they even go online and call out the makers charts and graphs who have an obvious agenda with their visualization, rather than objectively displaying the data as it exists.
Lesson: Chart selection and proper scaling are key
By adjusting the scale and time period, one can manipulate how data appears.
Isson and Harriott say, “Don’t boil the ocean.” Are you including more information than users really need? Dona M. Wong from the Wall Street Journal is a big proponent of providing only what is needed in a chart.
Lesson: Avoid clutter.
By removing background items and unnecessary lines and ticks, the data being communicated is quickly comes to the fore.
Similar to objectivity, this item seems to be included by Isson and Harriott to ensure unconscious/unintentional visual missteps do not make it to the final project.
Lesson: Proper use of axes, scales, & chart types
Maybe you’ve seen this snapshot making the rounds online lately?
The problem with this design may stem from a misunderstanding of how Venn diagrams work. Venn diagrams are meant to show the relationships between a set of items.
As such, this diagram is saying that Thomson Reuters values do not overlap very much with the ones in the orange bubble. Oops!
Another common culprit for unintentional visual dishonesty is 3D. 3D charts can be confusing to users if grid lines and markers are not carefully planned.
Imagine the audience
According to Isson and Harriott,
“A sizable percentage of nonanalytical business partners don’t have the ability, the time, or the interest in expending cognitive effort trying to figure out what your graphs mean.”
Lesson: Get to the point!
These static and interactive chart features can improve usability of your data visualization key audience groups.
Similar to necessity, but with chart data. The recommendation here is to keep your audience interested by focusing on one topic at a time. But what if your data has more to say? You may have to save it for another time. Or create an interactive visualization that can accommodate those users who want to explore further.
Lesson: Less is more.