Andy on Enterprise Software

Seeing is believing

August 9, 2006

A worthy competition was recently held by the Business Intelligence Network to pick the best data visualisation solution to a series of problems presented. The first article in a series of five to discuss this (by Stephen Few) nicely illustrates how to meaningfully display data in an example, and perhaps even more usefully some instructive examples of how not to.  It is perhaps appropriate that the winning solution was from Spotfire, a company that has carved out a strong niche in visualisation, initially in the pharmaceutical industry and recently on a broader basis.

I have written previously that visualisation software is not a mass market, but it can have a major impact in areas where smart visual displays add value e.g. in trading, life sciences or oil exploration.  Stephen’s article also shows up something very illuminating: having a flashy tool is less improtant than thinking carefully about design.  When Powerpoint came out presentations everywhere started to sprout absurd clipart and pschedelic colours, but eventually it was realised that there was still a place for graphic designers.  Similarly in visual analysis it is rarely the case that a more sophisticated tool is needed – even Excel has quite decent charting abilities.  However application designers put too little thought into thoughtful design that is intuitive to use. Before rushing off to buy the latest tool, try reading the following book: “The Visual Display of Quantitative Information” by Edward Tufte, which elegantly illustrates best practice in making sense of complicated data sets in ways that are meaningful and easy to understand.  It should be mandatory reading for anyone designing a system involving the visual display of information.

Next time you see an incomprehensible or misleading graph presented by your IT department, smile sagely and buy them a copy of this book.

1 comment so far

Hey Andy – an old friend of mine, who’s very smart, once told me “People don’t care about data models, they care about what they can do with data – and most of all they care about how they can see it in great ways”.

So what’s up with this “flashy tools” stuff?

Now that said, thanks for your nice comment on Spotfire.



Leave a comment
Your e-mail address is for administration purposes and is never displayed.

(required)

(required but not displayed)