![color palette from image 1.5 color palette from image 1.5](https://visme.co/blog/wp-content/uploads/beach-color-palette-hex-codes-pastel-colors.jpg)
That way, if we are interested in telling a story about one data point, we can do so quite easily. By using grey as the primary color in a visualization, we automatically draw our viewers’ eyes to whatever isn’t grey. This is because the absence of color, not the excessive use of it, helps paint a picture and tell a story.
![color palette from image 1.5 color palette from image 1.5](https://www.color-hex.com/palettes/11458.png)
It would be a waste of time to present to them all of the exploratory work you did, which is why your presentation should make use of color to focus on your findings.Īs practitioner Andy Kirk puts it, visualization practitioners in this stage of presentation should make grey their best friend. If you’ve already explored, analyzed, and probed your data, you now need to deliver those insights to someone else (a supervisor, a client, or a curious friend). This kind of color use would fall into the category of explanatory visualization, as opposed to its exploratory counterpart. One use of color is to draw attention to a data point of interest. You should ask: why am I using color? 1) Color to differentiate So, how can you use color correctly? It depends on the purpose of your visualization, and, as a corollary, the purpose of color. Rather, the (far too frequent) abuse and misuse of color is. So, how should you use color?Ĭolor is not the enemy. Color is one of the most important parts of our visualizations, yet their current use is far too often gratuitous and overwhelming. That’s why color should be used more sparingly and more thoughtfully. It’s indistinguishable, it’s confusing and you’re just off-loading the complexity and decision-making to your reader. Rather than trying to find that impossible 20-color palette, stop using color when you have so many dimensions. And if your chart presents 14 different data points all mapped to different colors, what kind of story is it telling? I really like this from Apple’s data visualization practitioner Elijah Meeks: That’s why its important that we stop uncritically asking how we can use color in our charts.Ī data visualization is nothing more than a pretty picture if it does not inform its viewer. The takeaway? When you emphasize everything, you end up emphasizing nothing. The reality is, however, that if you need more than a handful of colors in your chart, you can probably present your data in a different way. Often times, those creating visualizations will argue that they must include a 14 colors in their chart because the dataset has 14 data points of interest! It doesn’t help that the default settings of some of the most popular data viz tools (such as Excel) by default map categorical variables to colors. In the charts above, for example, it’s evident that there are far too many colors, with no apparent reason for the
![color palette from image 1.5 color palette from image 1.5](https://www.icolorpalette.com/download/schemes/c5aeb1_colorschemes_icolorpalette.jpg)
One of the most common errors I see is the overuse of color.
Color palette from image 1.5 software#
Some combination of 1) default software settings, 2) an obsession with pretty color palettes, and 3) a lack of emphasis on careful color consideration has led to a sloppy use of color in some of our most popular data visualizations. These examples illustrate my thesis: Too often, we ask how we can use color in our visualizations when we should be asking why we are using it.