Each year, influencers post their top reading lists. While I’m not Bill Gates or Oprah, my favorite read of 2017 prompted several “ah-ha” moments as I poured through the book each evening. Good Charts: The Harvard Business Review Guide to Making Smarter, More Persuasive Data Visualizations marries geeky data with business and psychology principles to relay the secret behind storytelling with charts.
I am not a data analyst, I’m not even all that good with numbers, but I love a good chart. And I’m not alone. Data is the new language of business. But the tenants of storytelling are what make data meaningful. Even when working with data, it’s ultimately about your message. Experts at Harvard say not to start with data, but start with the idea you want to convey, and let these tips guide you.
Substance before style
Spend more time thinking about your message than the style of chart. This might seem a bit contradictory when you are trying to find the best chart style to present your data. However, if you haven’t taken the time to really understand your dataset and the message you’d like to convey, even the loveliest chart can present a flimsy analysis.
Collaboration and diverse opinions will yield richer results
Talk out or whiteboard the problem with colleagues and pay close attention to the words you use. The conversation will influence how you frame the data, which parts are elevated, and the chart type. For example, the phrases “huge dip,” “crazy maze of a process,” or “flowed from,” evoke immediate images and suggest certain types of charts.
Focus on the essentials
Remember, less is more. Many chart standards such as gridlines, values and even data labels may not be necessary to get your point across. Unlike reading text from left to right, people’s eyes dart all around a chart and they focus on the items that stand out. Emphasize your key points with bold colors, shapes, large fonts, or strategic placement. Conversely, if something is not relevant to your message, consider excluding it.
Use established terminology to understand each other
Chart structures are common for a reason; for instance, the Y axis typically shows time. Flouting convention will just make your charts harder to understand.
I also appreciated the thorough discussion of data integrity. It is easy to manipulate data. We can exaggerate points, or omit points altogether; we can show two different views of the same data and come up with opposite conclusions. The book gives practical advice to help us determine if charts are deceiving.
When I put these principles into practice, I was surprised to find that not everyone is ready to ditch the bar charts and legends in favor of a simpler chart. Folks who work in the trenches with data, often like to see every data point on a chart. However, I’ve noticed leadership teams can be more interested in the analysis and implications, rather than the full dataset. With a chart that tells a story through data, you can quickly and easily communicate key points.
If your job involves data—which, these days, most jobs do—you can up your data game this year, and in the process, improve your business discussions too. Good Charts can make you a better storyteller with data.