Welcome to my first tutorial on the best practices for visualisations. Whether you’re creating your first visualisation (viz) on Tableau or on Power BI, these key tips and lessons are applicable to you. We use visualisations to better understand data by mapping out relationships or trends and highlighting outliers or patterns. This three-part series of blog posts will aim to answer the following questions:
- Part 1
- How do you embark on your first visualisation?
- What are pre-attentive attributes? Why are they important?
- Part 2
- How could you use pre-attentive attributes effectively to create meaningful visualisations?
- How do you cater for audiences with accessibility needs?
- Part 3
- Which chart types should you use, and when?
Embarking on your first visualisation (What? Who? When? Why? The Four Ws)
When you start building your visualisation, you’re excited to try all the options; you might have not even planned the story you are trying to tell. So, you dive right in and begin using all the colour options available (there’s so many to choose from), then you play around with the shapes, the sizes by the end you might end up choosing the most “interesting” looking chart types with the most “exciting” looking elements.
This is a terrible idea.
Stop and think!
Think about the story you are trying to tell.
What questions do you seek an answer to? Who is your audience? Why are you creating a dashboard? What is your origin? Is it exploratory or explanatory? Are you predicting future trends, showcasing the present situation or relating events of the past?
You need to understand the story you want to show your audience and the message you are portraying. This brings me to the next part about planning a clear layout for your dashboard.
You’re making a dashboard that you are the expert of – well, you should be – but your viewer might not understand the data or the visualisations without you creating clear guides such as legends, headings and text (clean and concise text – not an essay).
Sketch it!
Draw out a plan of the dashboard you want to create—envision the placement of the charts.
You’ll also need to ensure that you use colours, sizes, and chart types effectively to demonstrate the insights you find in your data exploration.
It’s important to understand some basic ‘visualisation best practices’ to ensure that what you present is accurate, easy to understand, and conveys the story you are trying to tell. To effectively tell the story, begin by understanding pre-attentive attributes and how it affects our minds.
What are pre-attentive attributes, and why are they important?
Pre-attentive attributes refer to the specific visual properties that our brains can understand or draw conclusions from—instantly and instinctively. This could be through colour, shape, size, orientation, and position; however, some of these are better to use than others.
It is important to be aware of the attributes in order for you to better understand how your viewer may interpret your dashboard. Which will help you better relay the insights your charts are supposed to be showing.
Since -once again- you might know the data and where the patterns, insights and trends may be, but the viewer might need you to guide them. To show them what you found, and the meaning behind them.
This is particularly pertinent when dealing with high-level clients that might not have the time to try and look through all the data you are presenting. As the saying goes, “time is money” and this is especially true in business. Instead, you need to help your viewer quickly grasp the message, or the outcomes in as little time as possible and as precisely as possible.
So, let’s look at an example of what I mean by this. Look at the image below. All the birds are coloured the same, the position and orientation are all the same—how easily and instantly can you identify all the ducks?
How did you find it?
Did you find three ducks?
Was it quick and efficient?
You probably needed to look carefully row by row to spot the ducks, which is not the most efficient for someone trying to understand the results in a timely manner. It might even be possible that you missed a count if you tried to do this quickly – which would affect the accuracy of the viewers estimation.
Now, look at this next example.
Colour hue allowed you to instantly count the number of ducks in a single glance.
Here is another example, using grouping to help count the different types of birds.
Having all the birds grouped makes it easier to count the different species quickly. As you can see, our brains are better at processing information if it is presented in specific formats.
In Summary
To conclude this blog, let us recap what we have learned. We have discussed beginning your first visualisation, pre-attentive attributes, and their importance in data storytelling. You know that it’s important to plan your dashboard by answering key questions from the Four Ws.
It is vital to sketch your dashboard layout, while thinking about the charts you will use and the legends, text and colours that might be needed to guide your storytelling. Pre-attentive attributes are key – they guide our brains to understand visuals; they are important to think about so that the story you tell can be quickly grasped and business decisions can be made efficiently.
In Part 2 of my blog post, I will continue to discuss ten of the pre-attentive attributes that have been identified by Tableau in more detail and how you can use pre-attentive attributes effectively to create meaningful visualisations. I will also discuss how to cater for audiences with accessibility needs.
See you in the next blog if you want to learn more about data visualisation best practices and become a better data storyteller.
“To be a data storyteller is to be a creator, a maker, a constructor. We do not just make information, but new and enthusiastic visions of how things are and how they might be.”
— RJ Andrews, author of Info We Trust
