Let's start simple. In this chapter of his book «Fundamentals of Data», Claus Wilke explains what we use colors in data vis for. Keep in mind the difference between a qualitative (= individual hues) and a sequential or diverging color scale (=gradients); the next articles will build on that knowledge.

After learning that we need both nice distinctive colors and nice gradients for our data visualizations, let's start with reading more about distinctive colors. Here's a long article I wrote about how to pick more beautiful ones.

In the previous article, you learned to extend your color palette by using non-pure colors or changing the saturation or lightness. Today we'll limit ourselves again: How can (or, actually, should) you change your data visualizations so that colorblind people will be able to decipher them, too? 

Instead of reading something today, here's an exercise for you: Using the great tool Viz Palette, spend a minute or two to come up with seven colors that don't have any color conflicts in the Color Report at the bottom of the page, not even for lines or small points. Then check the Color Report for colorblind people (at the top, «Color Population»). Spend the rest of your 10 minutes trying to adjust the colors so that they have no color conflicts for anyone. (It's likely you fail. That's ok.)

Grey is the most important color in data visualization. In this nice and short overview of ways to use grey when communicating data, Andy Kirk explains that thinking about the color for highlighted data points is as important as thinking about the color for non-highlighted ones. (If you don't know it already, check out visualisingdata.com after reading the article! It's one of the most important blogs in the data vis scene.)

Before moving on to learn more about gradients, here's a reminder that colors convey meaning. In this article, I explain why people use pink and blue to communicate gender data, why we shouldn't do this, and which alternatives exist. 

Samantha Zhang from former tech company Graphiq, before getting acquired by Amazon in 2017, wrote a short, easy-to-read article about how they designed gradients for their data visualizations. Don't miss the explanation of rule 2, «Follow natural patterns of color». 

When applying gradients to a choropleth or heat map, you can use different kinds of interpolations, or «stops». In this last article, I ponder how to use them for a good compromise between drawing attention to the facts that you want to draw attention to and using the data in an honest way.

Thanks for joining me on this colorful journey into the world of data visualization! I hope it answered some questions and provoked new ones. To learn more, the best way is to experiment, play around, and ask for feedback, e.g. in the Data Vis Society Slack or on the Data is Beautiful Subreddit. If you're interested in more data vis principles, consider browsing around the Datawrapper blog for which I'm responsible. And don't hesitate to get in touch with me, either on Twitter (@lisacrost) or via email.