Do Think Carefully
When you’re sitting down to create a data visualisation, you might think the most important thing is to decide what kind of visualisation to use—would a bar graph grab more attention, or should you use the trusty pie chart? Those actually aren’t the questions to be asking. Rather, the most important question to ask is ‘What is my message?’
There is no purpose of a data visualisation for the sake of it. Your data visualisation needs to tell a story, and you need to understand what that story is from the outset.
Do Understand the Principles of Data Storytelling
Data storytelling is a way to make sense of information (usually numbers) so that your audience cares about it. With data storytelling, you explain why your information matters.
Data visualisations are actually a data storytelling technique—they bring data storytelling to life. However, if your visualisations can’t tell a story effectively, your audience will struggle to care about your numbers, and they won’t understand what the next best steps are to take.
Do Pick the Right Data Visualisation
Once you understand the point you want to make as well as how to incorporate data storytelling into your data visualisation, your next step is to choose the right kind of data visualisation.
How do you choose the right data visualisation for the data? That answer depends on your information, your audience, and your goal. Let’s say you want to show how sales have increased over a six-month period; the right data visualisation would be a line graph because your audience will be able to see how sales rose.
Don’t Make Your Data Visualisation Cluttered
It's not a good idea to illustrate a number of things within a single data visualisation. If you do, the visualisation will look messy and cluttered. As a result, you risk distracting your audience, if not repelling them outright.
The guiding principles of data visualisations are as follows: smooth and sleek. Keep them simple so that they grab your audience’s attention.
Do Use Colours the Right Way
The right colours can make your data visualisation stand out and attract the eye. Conversely, the wrong colours do the following:
- Make your data visualisation confusing
- Divert the audience’s attention
- Actually drive them away from your message
How do you choose the right colours for your data visualisation? Think about what you want to say. If you’re trying to show two contrasting messages (such as expense data from two different years), you want colours that contrast, that are easy to see, and that won’t give anyone a headache when they look at them.
Also, use colours logically. If you know that red and black typically have negative associations, use those colours to show downward trends. Green tends to have more positive associations, so it’s a good colour to use if you want to point out an upward trend.
‘Use colours logically—think about the associations colours have.’
It’s important to note that seven to ten percent of men experience colour blindness. They might not be able to distinguish between certain colours, meaning that your beautiful data visualisation will be less meaningful to them. Test your image with colour blindness simulators—there are free versions available online.
Don’t Use Bad Data
What do we mean when we say ‘bad data’? Bad data can fall into a few categories:
- It could be out of date
- It could be inaccurate
- It could come from an untrustworthy source
- It doesn’t make sense (for example, all segments of a pie chart have to add up to 100%)
Before you sit down to work on your data visualisation, check your sources—are they up-to-date and accurate? Can you trust this information, and does the information contradict something you already know to be true?
Do Make Data Visualisations Interactive
Interactive data visualisations capture your audience’s attention and demonstrate why your information matters.
Here’s a real-life example: in 2017, Enlighten Designs began collaborating with Microsoft to support its Data Journalism programme. Data journalism allows readers to understand the fuller story through relevant, accurate information. Through this collaboration, Enlighten assisted the Associated Press (AP) in creating an interactive map of how communities in the State of Virginia voted in elections.
‘Interactive data visualisations capture your audience’s attention.’
Viewers could see the election results as votes were counted. Moreover, the visualisation was customisable for AP’s subscribers, so voters across the US could see how their local communities voted.
Do Label Charts and Graphs
Whichever data visualisations you choose, make sure you label them properly.
For example, if you’re using a pie chart to show the breakdown of your customer base, label each segment carefully so that it’s clear that, say, 40% of your customers are CMOs, while 60% are CIOs.
Do Ensure the Data Visualisation Passes the Squint Test
The goal of the ‘squint test’ is to check the strength of your data visualisation design. Here’s how it works: squint at the page until you can’t read any of the text. Now, what can you still see on the page? Can you tell if the data visualisation is balanced, that the colours pop out, that it’s clearly organised?
There are times when our data visualisations won’t always appear as crisp and clear as they were when we created them—maybe it’s the fault of a projector, or there’s a problem with the screen. Performing the squint test is an excellent way to ensure that the overall design of your data visualisation is strong.
Enlighten Designs: Helping You Create Amazing Data Visualisations
Since 1998, Enlighten Designs has specialised in creating fantastic digital experiences, including stunning data visualisations. We’re a proud Microsoft partner, with extensive expertise in Power BI. To learn more, contact us.