Colour is a part of our everyday lives, we choose clothes with colours we like and it guides us on our journeys to work everyday. Think about your drive, walk or ride to work every morning. We stop at the big red stop signs, walk when the green man says and stop when the red man says. Due to the importance of colour in our lives it becomes an an important part of our dashboards and data visualisation.
So let’s talk data and colours!
How many colours should I use?
Stephen Few states, “use colour only when needed to serve a particular communication goal” (Few, 2008). Therefore, colour should only be used sparingly.
A dashboard should not be a rainbow of colour, colour should be serving a purpose as we need to justify the selection of every choice made in our dashboards. Stephen Few clarifies this by stating “use different colours only when they correspond to differences of meaning in the data” (Few, 2008). Accordingly, you do not need a different colour for every chart type you use.
Myself personally I only like to use 7-8 colours maximum. My colour palette will contain a header colour, main colour, secondary colour, font colour, figure ground colour and comparison colours (data is always black and I do not include the colours of company logos). An example colour palette below.
What colours should I choose?
When building dashboards I always use our clients company or website colours. I believe picking colours from your clients or company website to be the easiest way of creating your colour palette.
By using your clients or company colours it provides an easy way of providing context to your dashboard. As a client could have multiple websites/brands which differ in colour, and by matching the colour of the website to your dashboard, you are allowing users to preattentively process your dashboard with the correct website.
The downside of this approach is that you are limited to the colours of your company or client website. This can lead to issues when your client/company website uses saturated colours or if they reuse colours for their sub-companies or brands.
So what do we do if our company or client is using saturated colours?
Ideally we should avoid saturated colours as they cause unnecessary strain on the users eyes. “Areas of strong colour and high contrast can produce after images when the viewer looks away from the screen resulting in visual stress from prolonged viewing” (MacDonald, 1999). An example of this, is when you stare at the sun and continue to see the sun after you look away.
Recently I found a way to overcome this limitation by altering the background of my data studio reports from white to light grey. By doing this I was able to significantly reduce the strain on my users eyes which was caused by the saturated red in the report. So if you are in this situation consider using a grey background to reduce the strain on your users eyes.
E.g. Stare at both red squares and you will find the red square is bearable to stare at with a grey background.
Another issue with using saturated colours is that information doesn’t stand out as much in your reports. To overcome this limitation, reduce the number of colours in your report and use a contrasting unsaturated colour, e.g. for a report where the main colour is red the contrasting secondary colour should be an unsaturated red such as a peach or salmon.
What happens if your company or client doesn’t have different colours for their sub companies or different brands?
One way of achieving this is ensuring your dashboard can be differentiated by more than just colour, as the appearance of your dashboards should be different by the variety of different visualisation techniques (see my other blog about Data Visualisation Techniques for Great Dashboards) you use to acquire user attention.
However, by using different colour palettes for your dashboards you are allowing users to preattentively process your dashboards with the correct brand or company.
So if you are in this tricky situation you should flip your colour palette, by swapping your colour header with your main or your main with a secondary colour and so on. This provides enough separation between your dashboards so that users can preattentively process your dashboards apart.
E.g. Swap the main colour teal with the secondary colour red.