The Best Data Visualizations Go Beyond Curb Appeal
A home should rise up to meet you. This idea, tweeted by Oprah, isn’t just about curb appeal. Sure, what you see from the street is part of an abode’s power to pull us in. But what really makes a home feel inviting is the story of its occupant. This story is what greets us at the door; it’s embedded in the walls, infused in color palettes, embellished by accents and underscored by negative space.
Good data visualization is a lot like a welcoming home: It goes beyond window dressing to present data in a way that “rises up to meet” its intended audience.
“Data visualization is not just about the data. It starts with visual storytelling, and it always has a targeted audience,” says information graphics expert Dona Wong. “As we develop a graphic solution, we have to take the perspective of our audience and approach the graphic solution with conviction and empathy.”
“Pretty” Doesn’t Drive Action
With access to today’s software, it’s easier than ever to analyze data and spit out decent-looking graphics. But, as Wong emphasizes, data visualization is not an end in itself, but a means to a very specific outcome. Despite prevailing attitudes, this outcome is not a polished chart, destined to be plopped into a PowerPoint slide.
“In the current state of data visualization, most graphics out there are merely data dumps,” says Wong, whose career dossier includes graphics director at The Wall Street Journal and business graphics editor at The New York Times. “Over time, people will become more educated and more demanding as graphics consumers.
“They will ask the basic question: Does this graphic help me make informed decisions? They won’t appreciate 3D charts that mask information.”
This illustration, adapted from Wong’s book The Wall Street Journal Guide to Information Graphics, depicts the three essential elements of good information graphics.
Hankering for real-time dashboards and other data-infused graphics is becoming an almost primal need for marketers. We may long for these graphic illustrations, but are our requests really reasonable – or even truly helpful? Analytics experts invest considerable time analyzing data and building visual stories. If a data visualization is built from marketing information you can’t act on, it probably doesn’t make sense to create it.
For example, why identify a customer segment if there aren’t any buyers for your product or service in it? Say the segment in question does have potential buyers: Are you equipped to target them? If not, a graphical breakdown of the segment probably shouldn’t be a priority for your analytics team. Maybe a more useful question to investigate would be, “Is a rich understanding of the ‘who’ really more important to your campaign performance than the ‘what, when and how’?” How much of your successful targeting efforts were based on segmentation vs. the timing of previous offers (e.g., day of week, time of day)?
Data Exploration Before Design
To ensure a data visualization will be useful and actionable, information designers start with analytical exploration.
“We analyze the data to find the story,” says Wong, who penned The Wall Street Journal Guide to Information Graphics in 2013. “Everyone talks about big data, but the real story is in the small data.”
Filtering out the noisy data and finding the relevant, “small data” is a big job. And this is precisely what Steven Alexander does every day at Dun & Bradstreet.
“My main job is to tell senior leadership a visual story about our marketing performance. It sounds simple, but it’s definitely not,” says Alexander, who is senior director of customer analytics and insights at D&B. “Building a dynamic dashboard isn’t as cut-and-dried as spitting out static financial results at the end of the month. It’s alive.”
Alexander’s use of the iconic line from Frankenstein is apt in more ways than one. In addition to working with “living,” always-evolving data, he often pulls information from disparate systems, connecting report fragments into one seamless and unified truth.
“Say our CMO is surprised by a blip in closed deals and wants to know what’s causing the uptick,” says Alexander, who uses Tableau software to build interactive data visualizations. “We drill down into lead volumes to see if it’s a question of better lead quality. Or maybe sales’ follow-up call volume has increased. We can then visually communicate the root cause to our CMO.”
Part Dr. Frankenstein, part Sherlock Holmes…Describe Alexander’s analytic art how you will, but to him, it’s data exploration for supporting decisions. His colleagues on the advanced analytics team pursue the same goal for Dun & Bradstreet marketing clients who need help with segmentation, targeting, lead scoring and identifying growth opportunities.
Often, analytic discoveries (and their subsequent visualizations) challenge what executives think they know – about their customers as well as about the overall market.
“When done right, a visual analysis allows the audience to compare, contrast and discern unexpected patterns and anomalies,” Wong explains. “For example, a heat map of mortgage delinquencies can tell us who borrowed money, where they live and whether they can pay it back.
“By comparing and contrasting different neighborhoods, investors can identify new patterns, which help them make informed decisions about the housing market.”
Visualized Data Meets the Decision Maker
If a data visualization is effective, its message should be clear, equipping audiences for immediate action. Wong is particularly passionate about this because it’s deeply rooted in her studies at Yale. Under the guidance of Edward Tufte, she completed a two-part thesis. In the first portion, Wong focused on optimizing patient monitor displays in intensive care units; in the second, Wong mapped the location of emergency phones in relation to campus crime statistics.
“In those years, I learned information graphics are literally a matter of life or death,” Wong says.
Complicated graphics and unwise color choices make visualized data useless, leaving decision makers paralyzed with confusion. Thankfully, information designers can avoid this scenario with a clear data strategy.
“Data strategy is about evaluating the data to select the right chart form and the right color to display the information clearly,” she says. “Selecting a chart form isn’t like choosing a T-shirt. Every chart form has its own function.”
A sampling of Wong’s straightforward information graphic tips. Adapted from The Wall Street Journal Guide to Information Graphics.
Marketers aren’t the only ones who mess this up. Even data scientists are seduced by “curb appeal,” choosing colors and charts based on aesthetic preferences rather than on a data set’s characteristics. While data patterns are important to information designers, their depictions are more thoughtful and strategic. They think about how the information will be used to generate actual business outcomes.
“In my workshop, we use the same data to try to tell different stories. Depending on the data analysis and the chart form, you are either telling a story about market share or growth,” says Wong.
Here’s where a good data visualization brings it all home for marketing decision makers: the story and how it’s told. The true art and science of “seeing pictures with numbers” (Wong’s wording) lay in a clear, accurate and meaningful message that inspires action, the heart of which “rises up to meet” marketers.
When we see tangible business outcomes, we know we’re home.
The views expressed here are Wong’s own and do not necessarily represent those of the Federal Reserve Bank of New York.