Your skills and attributes
The skills required for most effectively displaying information are not intuitive and rely largely on principles that must be learned.
It’s relatively easy to start creating charts and diagrams. After all there are many software applications available, some are likely to already be installed on your computer, ready to let you work with data right away. There are also a lot of useful resources – books, blogs and tutorials which can help you to learn. But like any other skill, in order to get good at it, you have to practise and begin to discern what works and doesn’t work for you and your customers.
It helps if you are generally interested in data and curious about what the data can show you. Ideally you should have an eye for spotting patterns and be able to analyse as well as synthesise. It is also useful to be open-minded to what you discover and accept it if the data contradicts any pre-conceived ideas you might have. Once you understand the data yourself, be a storyteller and communicate the story hidden in the data in the most accessible way.
If you are unsure where to start, find some good examples and try to recreate them. You could try to get input from others – either other data analysts or from the audience you are working for.
Consider your audience
Sometimes when working on data analysis and creating charts, it is possible to become so focused on the data and technology that you risk losing sight of the fact that the data reports are being designed for someone else to consume. The best data visualisation is the one that meets the audience’s need, and identifying that need is often a big challenge.
To start with, consider the level of the visual and statistical literacy of your audience. Take into account their mix and background. If it is likely that you will have novice users, you will need to stick with simpler, more common charts, include annotations, explanatory text and avoid business jargon and abbreviations. On the other hand, if you are dealing with well experienced groups, keep their interest and buy-in going with more innovative approaches. Don’t forget that your work can have educating impact on your customers and as time passes, you might need to ‘up your game’ to keep them satisfied and interested in your work.
Dialogue with end users
Make sure that you are clear on the actual data requirements, agree the scope, criteria and data definitions. Sometimes your clients may be unclear themselves or unaware that the data could be interpreted in a number of ways. Your data knowledge and expertise should help to explain what could be presented and the best way of presenting it.
According to Tukey (1962), “it’s better to approximate an answer to the right question than provide an exact answer to the wrong question”. It is therefore important to make sure that you and your client share a common understanding of data definitions and apply the same criteria. For example, when your boss asks for statistics related to international students, you both mean the same thing as this category could be interpreted in a number of different ways: students paying overseas fees, born in a foreign country or studying abroad at franchised institutions.
Before starting to work on creating charts that are intended for regular use by your clients, it can help and save a lot of time in the long run if you include them in the design process. It’s good to consider user testing and feedback on any prototypes or mock designs you can prepare in advance.
If appropriate, allow your user to edit the visualisation. Interactive online charts work well, especially when users can use filters, drill up or down to see different levels of detail, and add new data items. Some users have a preference regarding specific charts and display formats. This can be discussed at the design stage and can also be made available to them via an interactive online tool.
Be careful about being distracted by all the possible ways of showing the data and don’t forget that you may be dealing with an audience of statistical novices. Your goal should be to minimise the hard work needed to understand the data, and choose the visualisation which will do the best job of interpreting it.
It’s difficult to balance the ‘wow’ of a visually striking presentation (that tends to sell well) with the ‘aha’ moment when the user is able to easily and simply understand the problem at hand.