Lee Feinberg is a master of data visualization. Clad in glasses, paisley shirt, and humility, he… frankly, he reminds me of my older brother, but that isn’t much help to you, is it? Let’s start over. A couple weeks ago, STUDIO had the pleasure of inviting DECISIONVIZ President and resident data whiz Lee Feinberg into our Hoboken office for an informal how-do-you-do. An official partner of both Tableau and Alteryx, he has sailed the world of communicative data. He knows the names of seventeen kinds of visualizations you’ve never heard of. His house exterior is color-coded by material, his socks are organized in a datablick; and I’m clearly making things up, but you get the idea.

The saying goes: when you’re a hammer, every problem looks like a nail. This might lead a wizard like Lee to assert his particular brand of magic -- marrying data to pictures -- as the solution to any problem, but Lee’s magic doesn’t work like that. Yes, it would be of some help to DecisionViz’s clients if Lee and his cohort were to walk in, immediately help visualize a dataset in a way that uncovers a new trend, and walk out. In fact, to many entities, that would be of immense help. After all, you call DecisionViz when you want help visualizing your data such that you can make a newly informed decision. Right? So I had a fairly good idea of what we would get from a half-hour conversation with the data viz guru. And then he started talking about storytelling, and the scales fell, one by one.

History graphs and decision trees are two solid representations for Choose Your Own Adventure books.

“Data” is a word we use to ascribe purpose to numbers. Data has meaning, direction, and intent. Data speaks. In the same way that sentences organize words, and books organize sentences, effective data visualizations organize their constituent parts in a meaningful way. This seems obvious, but before Lee will dig into visualization, he takes a step back: what data do I have available? What data do I wish I had available? What trends do I think I might find? What connections have I considered? What other connections may exist? Do I have a hypothesis to test, or a result to demonstrate? Where are my words and sentences, and what is the story I’m telling?

While accurate, this chart is not terribly helpful. Just kidding, it's perfect.

A technology-first approach to data visualization -- whether it’s shoehorning data into the flavor-of-the-month, dynamic, combinatory, who-knows-what-graph, or it’s falling back to a pie chart because, damn it, I know how to make and read a pie chart – can be devastating. Misrepresenting data is akin to hollering a delicate and beautiful poem through a megaphone, backwards, underwater. No one wins. At best, you’ll confuse your audience. At worst, the meaning you were attempting to convey will be lost, perverted, or ignored. So the next time any of us think, “Hey, I should graph this,” or, “I wonder what our sales numbers look like,” let’s pledge to invoke Lee Feinberg. Let’s work to define our problems before we attempt to solve them. Let’s treat our data with the respect it deserves, and allow it to tell us the stories it hides in the folds of spreadsheets and annual reports.

My fellow people-who-deal-with-data-and-things, let us wield the tools of visualization responsibly: as the bindings and language of our stories, and not as nice-to-haves and afterthoughts. Otherwise, we'll all find ourselves feeling a bit like...

Max Baehr

Charts and graphs are fun, too!