I feel almost compelled to share the newest and flashiest environmental data visualizations with you, but I’m just not sure this new one from GE and the folks at Information Is Beautiful accomplishes its goal.  Looks like it’s time for a ReViz, where I rant for several paragraphs about a visualization and then try my hand at improving it.

carbon impact

Now, don’t get me wrong – there’s a lot of interesting and relevant information in this graphic, like the fact that two minutes on a cell phone emits more carbon than watching TV for an hour.  Surprising, right?

Trouble is, those “Aha!” moments are not that easy to come by.

31 Flavors will make you sick.

If you ever go to a gelato shop in Italy and try to order a licorice/mint chocolate chip combination, you’re likely to hear the phrase “Non si sposa” – “it does not marry.”  That’s how I feel about this visualization.

The designers have done something truly noteworthy by compiling the most recent and comprehensive figures to estimate the carbon impact of over two hundred products or activities.  It’s wonderful that they’ve shared those results with the public, and even smarter that they opted to visualize the data, taking advantage of the massive computer in your skull.

Unfortunately, this visual format does not marry well to the type and volume of data they’re presenting, and the interactive element only serve to make things worse.

Order vs. Chaos

Two hundred data points is a lot for your brain to process, so adding a bit of structure – clean lines, ample white space, inoffensive colors and movements – would go a long way toward helping you make sense of what you’re seeing.

Instead, the designers present each element as an icon, arrayed in concentric rings around a central element.  Adjacent items may be roughly similar in scale, but they certainly aren’t sorted according to impact.

I spotted the “Average World Citizen” with 7 metric tons on his chest and was surprised to see the “Average North American Citizen” sitting below and to the right on the chart.  How could an American citizen be lower than the world average, I wondered?

He can’t.  The American has 28 t on his chest, revealing the near random scatter of these icons.

Which brings me to my next criticism – clicking on an image does not shift the image fluidly on a scale of impact, it blanks the screen and then causes everything to re-arrange around the new middle element.  What could be more visually disruptive?

Clicking on the North American causes the world citizen to move down and to the left, while other elements scatter around the page.  It’s almost impossible to make any meaningful multi-level comparisons without a sheet of paper handy (e.g. try to put in order: Music festival, space shuttle flight, Yellowstone mud volcano).  A structured, and static axis would go a long way toward making this data more useful.

And unfortunately, the little pictograms carry almost zero information on their own – you’re forced to read tiny, low contrast type below the object to find out what it is, then tiny type from the center of the object to know its impact.  If you want pictograms, at least scale them with the carbon number so that the viewer can visually estimate impact, or better yet, put the elements in order so that things on the left are lower and those on the right are higher.

ReViz: Making it work

So what’s to be done? As a last resort, I guess you could use a boring old bar chart if the primary goal is to convey data, but I understand this graphic is meant more for edutainment than scientific study.  Still – finding the nimble balance between excitement and effectiveness is at the heart of any successful marriage.  Si sposa!

Providing structure (like a bar chart) allows the reader to more effectively navigate the data on her own, but how do you make it interesting so she’ll take the time to actually dig in?  Why not tell a story to get her involved?

Imagine that the world passed a comprehensive climate treaty and you wanted to throw a party (yes, I realize this is a stretch!) What do you serve?  Being the environmentally conscious host, you’ve got to choose between wine and beer so that you don’t offend your guests.  Is it a faux pas to serve a French Cab?  Will all of that CO2 bubbling from your Bud Light bring an ironic and early end to your party?

Create a real-world situation.  Ask a question that has a non-obvious answer to draw the mind into engagement.  By doing a bit of journalistic work up front and visually highlighting the results, the reader is paid off with an instant insight: domestic wine has a higher carbon impact than imported beer.

But better than that, she might get to wondering, “Well, that wine bottle is a bigger volume, so ounce for ounce, it may actually be better than the beer.”  She’d also notice orange juice on the list and really get to wondering about the environmental impact of a screwdriver!

The fact is, you as the editor have made the topic less daunting and more personal.  Rather than clicking around to see random pictures and numbers scatter across the page, the reader can now apply at least one piece of information to her real life. And isn’t that the whole point of visualizing data?