As I responded to the final edits on The Visual Organization, I couldn't help but reflect a bit. We've come so far since the 1990s with primitive Excel bar and pie charts. Now more than ever, it's easy to create visually compelling graphics that tell a story. Heat maps, tree maps, choropleths and other cool graphics don't require IT's involvement anymore.
But adding more isn't always tantamount to making a dataviz better.
Consider a relatively simple business question: Who are the highest paid employees at Google?
Of course the answer to this query can be represented graphically, and a recent HBR article does just that. The dataviz below is certainly pretty:
But that original, simple question has become much more complicated when viewing the infographic above – unnecessarily so. From the piece, it is "colorful...but nearly incomprehensible. This is what can happen when the tools that allow people to create new content – music, websites, data visualization – become available to the masses."
This begs the rhetorical question, Is this the best way to visually represent the Google compensation data? I would say no – not even close. While not as sexy, the following chart (based on the same data) isn't nearly as pretty, but it's something vastly more important: it is pretty effective:
Source: Junkcharts.com
Play around on Junkcharts and you'll find no shortage of just plain bad, ugly or confusing data visualizations. These are the types of graphics that fail to tell a story and/or fail to tell a story well.
Where to start? Start with the premise that simple is good, a point echoed by Jason Lankow, Josh Ritchie and Ross Crooks in their 2012 book Infographics: The Power of Visual Storytelling. The authors demonstrate how even very simple formatting can make certain data stand out at the expense of other data.
Simon says: subtract
Data visualizations are often effective not because of what they include, but because of what's omitted. There's absolutely no reason to overly complicate things. Remember that the goal is never to make art. Rather, you're trying to communicate information and enable better data discovery and, ultimately, decision making.
Beyond that, don't confuse your cons. Adding more for the sake of adding more often confuses. The goal should be to convey information.
Feedback
What say you?
