Whilst I was struggling through some designs which involve fairly complicated data, Anna Debenham recommended I have a look at ‘A Practical Guide to Designing with Data‘ by Brian Suda. I was really impressed with the previous Five Simple Steps books I’d read, Designing for the Web, Hardboiled Web Design and Designing the Invisible, so I bought it in both digital and paperback straight away.
The first thing I’ll say about this book is that it isn’t what I expected. In a time where infographics are being used (and over-used!) all over the place, I had expected the book to cover more of these visualisations. Instead, I found an incredibly solid foundation covering simpler and more straight-forward ways of displaying data through charts.
Parts 1, 2 and 3 were fascinating. Brian covers the reasons we use charts to express data and the visual design of charts with depth. I really loved the focus on simplicity and clarity, although I sometimes struggled to relate to the importance of the amount of pixels, and reducing the amount of pixels, in a chart. (This seems relevant to ink in print, and maybe slightly relevant to pixels on a small screen.)
Brian Suda is clearly a man of logic. Throughout the book he kept reminding me of Tuvok from Voyager. Except where he threatens to hunt down people who use doughnut charts, I can’t really think of a Star Trek character who would do that… Brian continually emphasises that expressing data through charts is about real maths. It’s about making the data as easy to read and understand as possible. Brian was clear that any chart that doesn’t do that either ineffective or misleading, or both. In fact, he’s fairly scathing about infographics and any decorative elements used to describe data, frequently dismissing them as ‘chart junk.’
This did mean that through Parts 4 and 5 of the book, ‘Common types of charts’ and ‘Not so common charts’, the conclusions are mostly that each type of chart is often created for a specific type of data, and if it’s not used for that type of data then it’s useless. This results in Brian frequently saying “you should really use a bar chart for this type of data.” If I’m honest, and whilst I’m sure that Brian is absolutely right, I found this a little bit prescriptive and left me feeling limited when I now design with data.
The last, and surprisingly short, chapter on ‘Everything and the kitchen sink’ covered a few more interesting chart types and I’d have loved to have read more about these. What I’m secretly hoping is, that now Brian’s covered the foundations of designing with data in this book, he’ll release another one with some more exciting and unusual examples to help us get more innovative with data design.
Overall, I’d definitely recommend Designing with Data. I feel as though I’ve learnt a lot from it but I found the first half of the book a lot more inspiring than the second. I’m becoming a firm fan of the Five Simple Steps books, they seem to be making a name for themsevles in thoughtful and interesting books really exploring design theory more than any other publisher.