I got a cold, so my head feels a bit clouded and I take it easy at work. Most of the time I did work, I tested a new feature we’ll announce soon at Datawrapper with lots of different visualizations. I find testing surprisingly fun. Every little bug I find and that we fix before the launch makes Datawrapper better…and lots of data vis designers have a better time thanks to that. It feels like a satisfying and valuable task.
I also thought a bit about color connotations. Last week I had an interview with somebody who was afraid of messing these up – afraid of “weird associations” that people might have because of the colors he uses in his charts. Another person I interviewed talked about “leveraging color connotations”. She felt like she didn’t fully understood them, but wanted to to evoke emotions and convey meaning. And today I read a long article about colors in data vis that were mostly about color connotations, too, and which color in an orange-blue map signals e.g. “good”.
showing which colors are associated with which attribute in which color.](January%2024/Screenshot_2022-01-24_at_18.44.00.png)
Famos data visualization by Information is Beautiful showing which colors are associated with which attribute in which color.
And then there seem to be people and organizations who really don’t care that much. I think I’m one of these people. I wouldn’t use red for data points that are supposed to communicate “everything is ok!” and green for “attention, attention!”. And I would show a data visualization to a native coworker before I published it in a foreign country. But if we’re designing for an audience that was raised in the same culture as us, they probably have very similar ones (e.g. for me: (red = danger/bad, dark = negative/serious, green = environment, good). And these color connotations are so instinctive, that we probably notice ourselves that something is off when we’re using them weirdly.
I haven’t seen many charts that use colors in a very unintuitive way. I have a chapter on that in the book – but it’s not super duper long because of that.
Am I naive here? Is recognizing something like “my red column chart looks like one big ‘danger!’ sign although it shows how glass production is going up” not engrained as I like to believe?