Beyond Insight
How to value data graphics
Good data graphics are expensive. Their design must be budgeted by people who are not smitten with the “magic of graphs.” Was your last bar chart worthwhile? Can you give a dollar-amount to the value it created?
Our craft is fragmented across society. As a consulting data storyteller, I’ve charted science, business, healthcare, academia, and entertainment. As an amateur historian of information graphics, I appreciate our craft’s trajectory, potential, and failures. My observations have me spooked: Today, there is a startling lack of recognition of the value that charts create.
Let’s take a step back and consider the point of charts.
The popular efficacy of data graphics goes something like: “Valuable charts spark insights that inform tasks, like decision-making.” This perspective reduces charts with the language of robotic efficiency: speed-to-insight, recollection of a learned fact, clarity of presentation.
Prevailing attitudes are incomplete (at best). They do not recognize a newspaper diagram that explains how a tragedy unfolded on the other side of the globe. They do not understand why a CEO might show a graph of old data in a high-stakes presentation. They have no patience for a weird new type of graph, perfectly fit to its data.
Data graphics suffer from what British advertising executive Rory Sutherland calls the doorman fallacy. It would be a mistake to define your doorman’s role as merely ‘opening the door’ and then replace him with an automatic mechanism. That narrow view misses the doorman’s other functions: taxi-hailing, security, vagrant discouragement, customer recognition, and signaling the status of the building.
The doorman fallacy would have us replace a colorful graphic with a narrow substitute, such as a statistical table or summary sentence. The substitute might convey the same bare-minimum takeaway as the colorful graphic, but it will not encourage the same meaning in the mind of its reader. (Further, the substitute might falsely portray precision and certainty that do not exist.)
Charts, like doormen, serve a multiplicity of functions. Under-recognized benefits of data graphics include attracting attention, building trust, and unifying a team. Perhaps it is understandable why these features are not often explicitly acknowledged: They are hard to distinguish. Many are impossible to quantify.
We can’t apply the same analytic rigor we use while processing our data inputs to understand our targets, people. But the immeasurability of the real benefits of data graphics—such as creating meaning and influencing behavior—should not dissuade us from seeking better understanding. I believe that these mysterious traits are the ones that matter most.
Story time.
In early 2020 I became a special advisor to the U.S. federal government (“expert, data visualization”). One of the reports I developed tracked the development of Covid-19 diagnostic tests. Across their lifespan, my charts created a variety of values.
As a social tool, the mere promise of a chart helped urgently coordinate and align our organization. It was a carrot (let’s make a pretty chart together . . . ) and stick (. . . because the White House needs to see it now). My charter afforded me access to sensitive data, helped me inspect that data, and motivated experts to teach me about the nuance of that data. Suddenly, a disparate batch of federal employees, with little prior relationship, were racing forward, unified.
As an information tool, my blue charts created context for what was happening regarding diagnostic tests. Early on, you could count the units they showed on your fingers. You might also anticipate that things were about to get busy. The chart also reassured its audience that a centralized function (i.e. RJ) was tracking the action closely.
The chart’s design evolved as time went on and the data mounted. I do not completely understand how it was after I sent it to leadership. We can intuit that one of the chart’s early functions was as proof of work: It showed that things were happening. It likely helped attract needed discussions about its topic, diagnostic tests: Having a chart to show is one way to call attention to a point of view. We might fantasize that the chart influenced a particular decision, but that is mere speculation.
The strongest indication that the chart was worthwhile was that I was instructed to keep making them. And then, I was asked to start making additional charts in the same style to show the development of other tools needed to fight Covid-19. The requested expansion was fuzzy feedback about the original report’s worth, but the signal was clear: something was working. My single-chart offering became a package of graphics that was updated, by hand, seven days a week.
After some months I handed-off the daily production of these charts. The printed graphics eventually became like wallpaper or furniture: A familiar part of daily life for their readers, but nothing that stood out from the background noise. After nearly a year of service, they were discontinued.
Going forward.
If we better recognize the ways data graphics are worthwhile, we might find that the biggest value of data graphics, even insightful charts, doesn’t have much to do with insight at all.
I seek a better sense of how data graphics create value. In the following weeks I will use this newsletter to explore the value of charts. We will look at data graphics with a few lenses. For example, an economic lens will help us see charts as attention-goods, information-goods, and cultural-goods.
If you have some thoughts about how data graphics create value in your life, or ways to think about this topic, I’d love to hear from you. Further, I appreciate your consideration of becoming a paid subscriber to this newsletter. All material support of my writing will help speed this series into being.
Onward!—RJ
Notes:
For the “magic of graphs” read Henry D. Hubbard’s preface to Graphic Presentation (1939): link.
I am a consulting data storyteller à la Sherlock Holmes, consulting detective. See my consulting work at infowetrust.com and histories at visionarypress.com
Read Alchemy by Rory Sutherland, especially if your field prides itself on being scientific, rational, or logical