Data Visualization Lessons from a Bad Chart

My favorite bad chart sat nearly undisturbed for years in the kitchen cabinet of my former workplace. It was an almost ordinary looking bar chart adorning the side panel of a big yellow box of Lipton tea.
It was the kind of chart that would be overlooked by most, but for some reason it always drew me in. I thought about the chart the other day and started to consider the lessons it can teach us about misleading visualization.

The Chart

Thanks to Google image search, I get to share that chart with you.
Look at it for a few, let it sink in. Also remember the context, the side of a big yellow box of 100 tea bags.

First off, what the heck is a flavonoid?

I had my guesses. Perhaps it was just some kind Flavor Flav inspired Dominoes Noid rehash…

Probably not.
But that’s really beside the point. For most of the chart’s readers, are they going to have any clue what a flavonoid is or is not?
Talking about something nobody knows like it’s something everybody knows is a jerky way to create any kind of visualization.

Wait, why is broccoli on that list?

Comparing tea, coffee, and fruit juices makes sense. But why broccoli?
This is another subtle way to mislead, with weird comparisons. Even broccoli has more flavonoids than coffee. We can do that kind of thing for almost anything.
Like, look at how awesome I am at baseball…

Ok, so really, what is the point of the flavonoid chart?

Flavonoid’s are plant pigments found in leafy greens. Those leafy greens are good for us, so maybe it has something to do with the plant’s chemistry?
Who knows…
They just wanted a chart on there that says,

Hey, I’m tea and I’m really good for you. Look, we all know I’m way better for you than coffee and this chart proves that. But not only that, I am also way healthier than these other things you already believe are healthy.

Oh, and bonus, Flavonoid sounds way tastier than other chemical names.

Why I’m sharing this with you?

Deceptive charting is more than just bad 3D pies, log scales, and awkward double axes. The choices we make on what to visualize and how we visualize can easily change the message. So when you’re creating your own charts don’t follow the Lipton Example.
p.s. Lipton made a second version of this chart but took away the broccoli. It’s fancier looking (see how the tea leaves go beyond the picture frame!) but not nearly as cool.

***

Bonus Ugly Charts

Turns out, if you want to find really bad chart and visualization examples, just do an image search for “Flavonoid charts.”
You’ll find awesome pie charts…

Super cool 3D Pie Charts

Snazzy picture annotated bar charts…

Chemical models…

Brand new crazy chart types I’ve never seen before. Like this one which I now will refer to as the whack-a-mole chart…

And this chart. Which might be my favorite of the bunch.

Why is there a spreadsheet in the background? Is that the grid? It doesn’t hit the x axis! But wait, those 3d bars have different vanishing points. And what’s causing the glare?

And finally, the following keyed me into the concept of a flavonoid better than pretty much any other chart or visual I found in the search…

Interested in creating the kinds of charts that actually communicate and engage? I offer design consulting services/capacity building through freshspectrum.com.
I’ll also be re-launching DiY Data Design, my digital information design workshop. You can hop on the waitlist now.