What? Data visualization is a language to be learned, and most people need to go back to the basics.
So what? An infographic should be a clear and concise carrier of information, rather than a pretty image to take up space on a slide deck or in a report.
Now what? Before creating any data visualization models, ask yourself what you want to say, why you’re creating a chart to tell the story, and how you can best visualize the data to create actionable insights.
visualization is a language to be learned, and most people need to go
back to the basics.
Dona Wong tells stories. As the vice president of digital strategy and communications at the Federal Reserve Bank of New York, her job is to tell stories using data, not words. ”Data visualization is about telling good stories. We’re wired to remember good stories,” she said during her keynote at the 2016 AMA Analytics With Purpose Conference. “Visualization stories help audiences imagine, experience and connect. When a story is well told, it helps people understand what’s behind that story with actionable insights.”
Wong categorizes most graphics being created today into three categories: trash (which is 99.99% of the graphics out there), pretty art, and highly functional visualizations of data. Her aim is to teach people how to create the latter. “Your data should be a carrier of information,” she says. “You should always think about your audience, not your boss.”
In her presentation, Wong laid out the “seven deadly sins” of data visualization: mistakes most marketers make when creating infographics either for internal meetings or external communications.
People don’t know why they make charts. People just want charts for the sake of having them. Strip down extra colors, extra info and numbers. Pare your graphs down to create only actionable insights so you can create promotions or some other kind of action around the information.
They put information too far away from the visualization. Put your graph and the legend as close as possible. Think about how your audience will read your info, and make it easy for their eye to take it all in at once.
They choose any chart form they want. Chart form follows data. There are four basic measures of data: 1) Slope, which shows growth; 2) vertical bars, which show qualities over time; 3) horizontal bars, which show ranking; and 4) pie charts, which show share of a whole.
They create cluttered pie charts. Pie charts should have no more than five slices. Use horizontal bars for ranking more than five.
They begin bar charts anywhere besides zero. All vertical and horizontal bar charts must start at zero.
They use 3D. Never use 3D. Ever.
They use color just for the sake of making something look colorful. Don’t just use color to make your chart interesting. Think about what it does to your chart. Limited colors, usually just one, will help you show a pattern and tell a story. Also, avoid red and green since 10% of men have a hard time seeing those colors.
Before you visualize your data think about what you’re creating, why you’re creating it and how you will customize your data visualization program to fit your needs. “When done well, infographics can be more universal than written language,” Wong says.
Watch AMA's CEO Russ Klein interview Dona Wong at the 2016 Analytics with Purpose Conference below:
For data visualization inspiration to get you started, check out Wong’s work at the Federal Reserve Bank of New York.