Losing the Plot

Charts that actually work can be hard to find.

There have never been so many charts around. More pie charts than you can shake your mouse at. Bar charts, flow charts, tree maps, radar charts, venn diagrams. And we’re software-empowered. So what’s the problem?

Well, I’ll get to the point: The majority of them don’t really work, and on top of that they’re very hard to decode. Sometimes I wonder if as much as 95% of the current output is dubious. Some of the remaining 5% is designed for scientists or specialists in a certain field, and although these charts are a mystery to me, I respect their role. And then there are the ones that do work: Ranging from the good to the truly inspired. People have found ways of presenting data that we could only have dreamt about in the last century. But it’s a very small percentage of the total output.

So let’s take a deep breath, and remember what we’ve known all along: Making infographics is a skill that has to be learnt. You don’t become Rembrandt when you buy a paintbrush. Nor do you become a data visualizer when you get your hands on a chart-plotting program. I truly believe that bringing information design to everyone is a great thing, but I‘m worried about the craft of understanding and presenting data being lost under an avalanche of infographic over-enthusiasm.

Page designers are sometimes compounding the problem. Now let me be quite clear: I greatly value the work of designers. I‘m always telling infographic makers to respect designers and learn from them. We can all improve our graphics with a designer’s sense of the use of space, typography, color, and many more essential components of good design. But… some designers have a habit of falling in love with ill-conceived infographics because they look great in a page layout. And as a general rule, making charts to match layouts, in style, size and color does not help to produce something that actually works.

How do we keep things on an even keel? We have to identify the nonsense graphics as exactly that, and be careful about posting examples of infographic weirdness on blogs with comments like “Cool” and “Innovative.” Every day I get a few emails with links to charts that are “really cool.” Often, the chart does indeed look cool or downright beautiful, but upon further inspection turns out to be a very bad way to present the data. We should all look at these graphics much more carefully, and ask ourselves if they work as visual explanations. Do they turn the light on?

There is a counter argument to the idea that charts should be checked by the infographic police. Obviously infographics need pushing into new forms. We cannot stand still in any area of design or graphics, so some risks need to be taken. And I don’t want to slam the lid down on experimentation. We just need to know in our own minds when we’re looking at developmental work as opposed to serious explanation. Clearly distinguish between charts that inform and charts that just explore the idea of informing.

Maybe you’re thinking that the dinosaur has spoken. And, yes, maybe I’m wrong about it all. But in my personal opinion, if anyone thinks that a large amount of the charted stuff out there is much use to mankind, they must be from somewhere outside this solar system.

Hey, that rant was fun. Now I feel much better. Where’s that pie chart I’m working on…

Footnote:Some simple questions to ask of any chart that is aimed at the average reader:

  1. Can it be easily understood by someone who is not a graphic designer?
  2. Is the amount of effort needed to decode the chart definitely worth the payoff? And most importantly:
  3. Is the data presented accurately, fairly, and of course, clearly?

John Grimwade is graphics director at Conde Nast Publications.