We see them everywhere in the news and on social media: charts, graphs, and other data visualizations calling out bold facts and figures. Across industries, organizations turn to data viz to reinforce messages—but some data visualizations are more effective than others. What makes a successful data visualization, and how should audiences read them in a digital environment rife with misinformation? Mary Reddy, LEFF’s senior adviser on data visualization, shares her expert perspective on the current state of data viz.
What should good data visualizations do?
Mary Reddy: In communications and marketing, data visualization uses facts to ground any assertions an organization may be making—it lends gravitas. In research and analysis, data becomes fodder for organizations to tell certain stories.
Data visualizations should be used to add meaning, not just for the prettification of a page. When I see a whole row of icons and factoids across a page, my eyes just kind of glaze over: with that many, they lose meaning, they aren’t novel anymore, and they don’t draw the eye.
Where can data visualizations go awry?
Mary Reddy: As I see it, there are three ways that things can go badly. One is failing to set context. For example, if you’re reporting on ESG [environmental, social, and governance] and you say you’ve increased the role of women in leadership by 50 percent since last year, but you fail to note that that meant you added one more woman. So you now have two women out of 60 or 70 leadership roles.
Another failing is trying to say too much in a single display. If you have an exhibit that contains too many charts, data points, or words, you’re going to lose the audience. They won’t spend time trying to figure out which things are the most important.
Finally, the surrounding text has to work with charts and visuals. People consume information in different ways, and you want to hit all the notes so that your audience gets it without a struggle.
What should organizations consider when developing digital data visualizations over ones in print?
Mary Reddy: The digital environment has produced an audience that wants something delivered instantly—and if it doesn’t come fast enough, they’re going to move on. When creating in a digital space, you’ve got to think of how to make it more novel.
I like reinventions that retain the integrity of a chart—you’re not breaking any rules, but you do something to make it a little more appealing and distinct from everything else somebody will have seen on their feed. Use headlines that make people think, “I’m curious. What does this mean?” It’s a way to delight the audience, get their attention, and increase their desire to come back and find more.
Why might data visualization be better than video or other formats in terms of communicating quick takeaways?
Mary Reddy: In data visualization, there’s a clear, simple message. It requires less of a time investment from your viewer; if you just put the chart and the takeaway in front of them visually, they can get it right away.
A more nuanced data display should also have a simple high-level takeaway, but there may be details and nuances that the reader is invited to explore at their own pace. In a video, unless somebody’s scanning ahead, you’re locked into the time the producer has taken to convey the message.
What trends are you seeing in the development of data viz?
Technology is allowing us to visualize data in exciting new ways. For example, by using well-crafted AI prompts, a person with no knowledge of coding can build a model to visualize predictive scenarios for environmental change using real-time data. Translating massive spreadsheets of data into maps or charts distills the findings into a broadly accessible format. Such clear illustrations of trends are invaluable for taking action to mitigate climate risks.
What organizations are making strides in elevating data visualization? Who does this particularly well?
McKinsey has a long history of intellectual rigor in their charting, coupled with really good design. Their charts consistently hit the right notes.
Bain & Company has elevated their design and data visualizations, including creating visuals for LinkedIn. They keep refreshing to align with visual trends on social media while standing out from the crowd.
The Economist’s Graphic Detail is also good, solid storytelling without too many bells or whistles. It tells an entire story in a little snippet of data around countries, revenue, climate change—you name it.
What can you say about the role of data in ESG reporting? Where can it make the greatest impact, and when should people be cautious about how they use it?
Mary Reddy: With new regulations, companies are required to submit a lot of ESG data. Data tables can do some of the work, but the Corporate Sustainability Reporting Directive (CSRD) also mandates some narrative disclosures to explain the relationships between sustainability and information about finance and risk, which is a new requirement. Good data visualizations are important here. They can really bridge the gap between raw numbers and the story companies want to tell, and they can help companies to make their sustainability messaging stand out in a world awash in data.
Everybody already knows about greenwashing, but greenhushing has also become a noticeable trend. If an organization is reporting its progress, it’s important to not only acknowledge sustainability as business value but also do it in a way that respects the intelligence of the audience. People can sense when a message is more about marketing than reporting.
I’m glad you mentioned that, because audiences need to parse a lot of messages in 2024. During an election year, what should audiences look for in terms of data visualizations—and what questions should we ask ourselves to make sure that we’re practicing good information literacy?
Mary Reddy: I look for—and want to hear more about—context. When we see data visualizations based on polls, are those polls focused only on a certain geographic area or age group? If you’re polling a certain age group, are you doing it online or by telephone? Is that age group more likely to respond to one polling method over the other?
If you’re looking at a particular data visualization, ask yourself if it’s providing that context. Has it followed the basic rules of chart integrity? Do you see distortions in the chart? Do you feel that there may be some cherry-picking of data going on? Look for statements that describe the methodology. Considering the source is another way to detect a bias in the data reporting. We need to better educate ourselves so it becomes instinctive for us to look critically at any data that’s presented to us.