It is crucial to choose the right data visualisation for end users. When you are dealing with a complex dashboard, you need to transform the data into instant insight. My role at Satsumas is to understand how to unlock the information beneath the data, either via main business intelligence (BI) technologies or open source solutions, for example the D3 library. Below are some examples of my research based on data from the Australian Bureau of Statistics.
Labels in charts should be short to leave as much space as possible for the measures. However, sometimes they are too long, and in this example, it is better to illustrate the offences with icons that are easy to understand.
Illicit drug offences
Prohibited and regulated weapons and explosives offences
Abduction, harassment and other offences against the person
Robbery, extortion and related offences
Visual representation of figures is a very powerful way for comparison. Alongside the donut chart that gives a sense of proportion by state, each of the top 5 offences is illustrated by icons. One icon equals to 1%, each row is always made of a maximum of 10 to facilitate counting if necessary. With this layout, you can either look at the figures for values or the realestate used by the icons to visually compare them.
(Click on the pie chart to create multiple selections)
Visualising a lot of data is often difficult and identifying trends or patterns is even harder. In this chart, bars for male and female are overlapping to immediately identify which one is higher per age. The animation through the years also reveals an important, non-natural increase of the population between 20 and 40 years of age over the last 10 years. It would take much longer to interpret these information with classic data visualisation.
(Use 
and 
arrow keyboard to go through years or click on these icons)
When I am working on data visualisation, I always try to remove, what I call "layer of interpretation", as much as possible. KPIs, figures or charts can easily become misleading or being misinterpreted, if you do not give their context or the right visualisation. Using the 5 WHYs is - as always - an essential tool to understand end users and their needs.
If you want to talk about data visualisation, please do not hesitate to contact me.
LinkedIn
Behance
