Why bother with data visualisation?
In many organisations, ALL data is presented to analysts and stakeholders in a big grid. Or perhaps a pivot table, if you’re lucky.
Sometimes, a grid is the appropriate way of communicating data. Usually, this is when you have a set of clearly defined and understood Key Performance Indictors (KPIs) which don’t need to be explained, and can be easily consumed from a single page report.
In many other scenarios, a grid of data will just make your reader’s head hurt. Particularly in the current data climate, with the volume of data available for analysis increasing beyond our capacity to process it.
This is where data visualisation starts to become useful.
Using data visualisation techniques will:
- Allow your organisation to better absorb the complex data sets being created
- Show trends earlier and clearer
- Increase the clarity – consumers will understand what you’re trying to say at first sight (if you’ve done it right!)
Sometimes, the visualisation can be something as simple as a chart, showing how a measure has changed over time. Other times, it can be as complex as those created by Mike Bostock of the New York Times and the creator of d3.
Statistician and artist Edward Tufte has been working hard for years to show us ways of presenting large or complex data sets in ways that allow us to appreciate connections within the data, and generate questions to be answered by further detailed analysis.
I’m not advocating reaching for d3 each time each time you need to build a report – but the question of how the data should be visualised should be one of the first questions asked when a report is being compiled.
Data visualisation doesn’t replace text or numbers. At all. Each should be used together to ensure the position you’re trying to present is as clear and as useful as possible.