Looks v Brains; Getting a balance

I saw an interesting graphic on twitter afternoon. It was popped up by Gavin Heaton and was by the agency We are Social.

They have produced an impressive compendium of Global Digital facts, and they had tweeted a chart from that deck that contained a summary of several measures for regions around the globe. It piqued my interest, and I had a close look. I found it difficult to make comparisons between regions so I created a data set, and had a little play in tableau to see what I could find

Let me compare thee to…

I think the design of the original graphic is very clean and elegant, but as the central design is around a map, I found it very hard to compare measures across each region. As a viewer, or consumer of the information, I want to be able to compare each measure, but I had to dart around the regions. It wasn't easy to complete the task I wanted to do - which is normally to try and see some sort of pattern or outlier or correlation. 

X Marks the spot

I love it when I get a data set that enables me to plot points on a map. What I've found though is there are a few choice situations where it’s actually right to use one. Plotting a small table on a region may be handy if I don't know where that region is, but in this example I know where all these regions are located. So this becomes a design redundancy, and clouded my ability to answer the “So what?” question inside of me…

So, I took the figures from this chart and had a play around to see what was interesting about the numbers. Mobile access / connections are of interest to a lot of people at the moment, so that was the first place I looked. I initially created a dual axis bar/line graph, so you can see regions that over or under perform in terms of mobile connections compared to their population.

But I think looking via a scatter plot makes the differences clearer. I could encode mobile connections, internet users, and population in the one chart.

You can clearly see that developing economies are the ones that have a higher proportion of mobile connections.  And from there you can go on and make all kinds of connections, or seek out other data sources;

  • that lack of infrastructure means this is an ideal technology...
  • how does this impact the development of digital tools in these regions...
  • East Asia has the largest proportion of internet users, and I wonder why their mobile connection proportion doesn't reflect others that are geographically close? 
  • what can we learn from these regions experiences in developing tools in more developed economies... etc etc

I am sure in the comprehensive deck that We Are Social has developed  contains all these kinds of observations (check it out here - it's likely to be a great reference and interesting read if you work in digital, insights, or have an interest in these things). I guess I really wanted to see what sorts of things emerged from the data if you allowed easier comparison between the numbers.

One of the things I have to constantly ask myself when I design an information graphic for a client is if the audience can easily do the task they need to do (compare data points, understand a dynamic, see a trend, etc)? Are they left with a "so what?" or an "Aha!"? (And what's an Aha for me could be a "yeah, whatever" for others - so making sure I test is also important too)

It's hard juggling these requirements with the demands of making something visually appealing. There is a LOT of debate in the industry about how far you go down the "visually appealing" path. Some practitioners are all "Just that facts m'aam"...but I tend to align myself with more moderate commentators - that an aesthetically appealing design is very important, because without an appealing aesthetic, you may be overlooked by your audience. I don't think my new chart is better looking than the original graphic.  I can't deny, there is something very compelling about a map! It's a tricky balance, folks, and a topic I hope to explore here again.