7 Best Practices for Using Color in Data Visualizations

Devon Tackels

Content Marketing Manager, Sigma

Does viewing a red light raise a person’s heart rate? Could a certain shade of blue reduce impulsivity? Color has been shown to have a small yet direct effect on human biology and psychology. Color also affects the way our brains process information. Using color strategically can increase memory, aid pattern recognition, and attract attention to priority information.

These capabilities combine to make color a powerful tool for data visualization. In this post, we’re looking at the best practices you can implement to make your data visualizations more effective.

Why color use in data visualization matters

But first, why does the way you use color in your visualizations matter so much? The goal of data visualization is to help viewers quickly digest information and remember it. While other design principles have a role to play (including the use of white space, contrast, grouping, etc.), color is one of the easiest to apply to data visualization. Using color strategically helps viewers understand the meaning and impact of the information presented — and remember the most important details.

On the flip side, color used poorly can distract from the story your visualization is trying to tell. Rather than aiding understanding, it will confuse people. Let’s dive into the best practices for better visualizations.


Bring your data to life with Sigma. Build unlimited data visualizations with a free 14-day trial

Best practices for data visualization colors

There’s no one right way to use color, but we can take what we know about how the brain is influenced by color and apply it to visualization design to get better results.

 Do use color to create associations

As with all design used for communication, good data visualization design harnesses common conventions and uses them as shorthand. For the same reason that UX designers always use a cart icon to indicate the button e-commerce shoppers should click to complete a purchase, data visualization designers use colors to trigger associations and streamline understanding. For example, you may use orange to represent safety performance, deep green to represent profit, or light green to represent environmental sustainability. Color palettes can also create associations in the viewer’s mind, such as the colors of a country’s flag communicating data related to that country.

 Do use a single color to show continuous data

Along the same lines, be sure to use a single color in various saturations (or a gradient) to communicate amounts or numbers of continuous data. Using one color will help viewers to quickly grasp that they’re viewing increases or decreases in a single metric, such as the unemployment rate or an infection rate over time, for example.

 Do use contrasting colors to show comparison/contrast

When you’re comparing or contrasting two metrics, using contrasting colors will help viewers intuit that you’re differentiating between the two. You might be showing the difference between the conversion rates on Facebook ads vs. Instagram ads, for example. In this particular case, you might use contrasting colors that are also associated with the two different platforms — light blue and pink-purple.

This Sigma dashboard uses a contrasting color palette to differentiate information. 

 Do use color to make important information stand out

When you’re trying to highlight something important, such as data relevant to a particular county or zip code, a bright or saturated color can help it stand out. You may choose to use gray for less-important variables and a deep red or orange for the most important variable, for example. Or you could use muted colors for the less-important ones and a bright color for the most important one.

 Don’t pick colors that aren’t easily distinguishable

We’ve all been frustrated by charts or graphs that leave us squinting to determine what numbers are relevant to what variable. You want viewers to be able to interpret data at a glance. For this reason, the best colors for data visualization are easily distinguishable.

 Don’t Use Too Many Colors

Because the brain struggles to process many different things at once, using a limited color set in your visualizations will improve speed to insight. Just as in the famous supermarket jam experiment where 97% shoppers were so overwhelmed by the 24 jam choices that they failed to purchase any, your visualization viewers can be overwhelmed by too many colors. Try to stick to seven or fewer colors in a single visualization, the maximum number of items that the brain can hold easily at one time.

 Don’t Forget Accessibility

Not everyone has the same visual ability. A broad array of color vision deficiencies may affect a person’s ability to distinguish between certain colors. While accessibility is a big subject with many considerations, you’ll want to be aware of the colors and hues that may cause issues for people with visual challenges. Here’s a resource to help.


Join our weekly webinars  everyFriday at 11am PST to learn more analytics best practices. 

Common data visualization color palettes

Beyond best practices, there are three types of color palettes common in data visualization that you will find useful. Like associations, these palettes are based on common conventions, making them easy to interpret.

Qualitative — Qualitative palettes are those in which each color is distinct from the others. This type of palette is ideal for visualizations displaying categorical variables, those that are unrelated to one another.

Sequential — Sequential palettes use a single color in a variety of saturations or a gradient. A sequential palette clearly communicates information in ordered, numeric values, such as dollar amounts over time.

Diverging — A diverging palette shows where variables sit on a spectrum, such as cold to hot. This palette reflects the data by using one color on one end of the spectrum and a different color on the other end, with a neutral color in the center. The colors in between the neutral center and end of each spectrum are gradients in between neutral and the end color (usually light to dark) on either side.

Even with the “rules” of color in data visualization, there’s always room for creativity. The guiding principle of visualization is to use every element to aid in communication. You’ll find that there are, in fact, many ways to communicate information using color.

Still have questions? Learn more about data visualizations in our Definitive Guide to Data Visualization.

Wanna bring your data to life?

Sigma can help: Schedule a demo or start a free trial today