Do Not Lose Your Audience’s Attention Using a (too) Colourful Visualization
Main steps to strategically use colours in your visuals.
The main purpose of data visualization is to help viewers quickly understand a big volume of information presented and easily remember it.
I said quickly understand a big volume of information, so the last thing we want is to distract and confuse the audience from the story your visualization is trying to tell.
Let’s check immediately some best practices for better visualizations.
Too colourful is not the way to go
The right use of colour will help highlight patterns and trends and make key points stand up from the big amount of data presented to your audience. But highlighting everything available in the visualization won’t help them focusing on the key points.
Make (only) important elements stand out: highlight the most important aspects of your message.
Avoid unnecessary usage, use colour only where appropriate. Most data should be in neutral colours like grey with one (or a limited number of) bright colour reserved for directing attention to significant or atypical data points. Using grey for “less important” chart elements will help bold colours stand out more. Grey, not black, to allow for greater contrast.
Look at these 2 bar charts, where is your attention going in the first chart? Too many info shown, nothing highlighted.
Of much more impact is the second one, where the colour is used to simplify data and effectively communicate the message. The audience will focus directly where we want them to look at, without spending time to make comparisons between bars, as we’ve already highlighted the important aspect for them (the “quickly understand” bit we were talking about in the opening line). The legend embedded in the title is helping too.
Use colours consistently
If you have certain variables that you repeat across different charts, you should use the same colours where possible. A sense of consistency helps viewers to relate, facilitating the process of interpreting the data.
Choose the right colour
Colour can be used to group data points of similar value and to reduce the extent of this similarity:
- divergent colour palettes allow the emphasis of a quantitative data display to be progressions outward from a critical midpoint of the data range. Technically a diverging palette could have as many colours as you’d like but usually range only between two contrasting colours.
- sequential data classes are logically arranged from high to low, and this stepped sequence of categories should be represented by sequential lightness steps. Low data values are usually represented by light colours and high values represented by dark colours.
The tools I like the most for selecting colour combinations are:
- coolors.co that allows you to pick up a ready palette or to create one
- viz-palette
Not everyone can see all colours
Roughly 8% of the world population is colour blind. This means 1 person every 12 won’t see the colours as expected. To make coloured infographics accessible to everyone, avoid use of combinations of red and green.
I found this interesting post to better understand colour-blindness and how this impacts how our visuals are perceived.
If you want to learn more about how to use colours in data visualizations and how to create effective visuals, I highly recommend to
- read Storytelling with data. A data visualization guide for business professionals by Cole Nussbaumer Knaflic
- check out this post written by Eva Murray for Forbes
- and follow Ben Jones, who shared his knowledge in this Data Visualization nanodegree.
Key points
A good set of colours will highlight the most important points and make patterns more visible, while a poor one will just distract from the visualization purpose.
Leverage colour selectively as a strategic tool to highlight the important parts of your visual and focus your audience’s attention. So use it wisely, use it consistently and consider colour-blind issues.
