Overload, clutter, and confusion are not attributes of information, they are failures of design
The use of graphs and tables are commonplace in communicating quantitative business information. In order to maximise their potential business analysts need to learn how to design graphs and tables effectively. There are many great examples of data visualisations, but there are as many if not more bad visual designs appearing in business reports, presentations and business intelligence dashboards.
The following section contains a few simple pointers to creating more effective information presentation. They are based on recommendations by well known data visualisation writers such as Stephen Few, Edward Tufte and many others in the field as well as our own experience.
- Select the right chart for the data type. This chart selector can help.
- Keep your diagrams simple – avoid unnecessary clutter and distraction.
- Maximise the data to ink ratio – only use extra colours, grid lines and labels if the graphics would lose the content when they are removed from the image.
- Use graphs to display information about data relationships – for example comparison, trend and proportion. Use tables when precise values are required.
- Pie charts can be difficult to understand because our visual perception is not designed to accurately assign quantitative values to two-dimensional areas. Use bar charts instead for a much clearer message.
- Help the reader by using explanatory titles which communicate the main focus of the visualisation.
- Label the axes and normally start them at zero, unless all data is clustered at high values.
- Use colour, size and position to improve understanding.
- Choose colours which can be seen by majority of the population. It’s probably best to avoid using both green and red in the same diagram as they both appear brown to people with colour blindness.
- Provide the source of information and dates to support credibility.