It’s called Iron Viz and it’s feeding time! Each year, Tableau Software hosts a series of ‘feeder competitions’ to select three worthy competitors from around the globe to compete in the ultimate data viz contest, Iron Viz, in front of thousands of data viz fans at the annual Tableau conference.
I’m determined to be on that stage, to proudly represent all Australian data rockstars and show the world just what we can do.
For the last two weeks, I’ve spent every spare moment working on my submission for this first feeder round. I put all my effort into telling a story, rather than showing off every trick in my toolkit. The result is this: Failure to act is failure to feed.
Two sources were used as the basis for this submission:
- United Nations – Department of Economic and Social Affairs, Population Division: 2015 Revision of World Population Prospects
- United Nations – Food and Agriculture Organization, Statistics Division: Food Balance Sheets
Between these two data sets, I had the makings of a really interesting story: I had population estimates from 1950 to 2015 and population projections from 2015 to 2100, food production levels from 1960 to 2011, as well as kg / capita / year food supply for this same period.
From these actuals, I wanted to derive future production and future consumption forecasts. I wanted to see for myself how well our world was equipped to deal with our future population projections.
What started as two simple tables quickly grew into a complex web of forecasts, scenarios and calculations*. I extrapolated the amount of food required to feed the world’s growing population. I calculated a range of scenarios: from no change, to incremental changes of 0.5%, 1.0%, 1.5%, 2.0% and 2.5% growth in food production. I was all consumed by the massive numbers that were presenting themselves, numbers I could not fathom and numbers that meant so much more than the characters on my screen.
I had to move past the empty characters, facts and figures to a story that would allow the weight of these numbers to be felt by those who would view this work. It was time for the story.
In exploring the data, I quickly became aware of a rapidly growing problem in Africa: the projected growth rate in population was far outstripping their local, as well as our global, food supply capacity. No other region came close to the levels of projected population, and therefore levels of food required, as I was seeing in Africa. For this reason, I decided to aggregate the regions displayed in the viz to Africa, Asia and “Rest of the World”. The sheer scale of the challenge in Africa required this aggregation in order for any perspective to be visible within the data visualizations.
Whilst my initial calculations and forecasts had focused on the amount of food consumed and required for the future population, I was struggling to feel what this meant. I was juggling numbers in the order of hundreds of thousands of metric tonnes. I’ve never needed to lift anything that weighs hundreds of thousands of metric tonnes, I had no point of reference by which to judge these numbers. I needed to turn these numbers into something I could understand. I chose life.
What if I were able to tell this story in terms of human life? What if I were able to talk about these empty numbers and characters in terms of mouths fed and mouths that we would fail to feed?
I had found my story.
At the end of the day, this was an exercise in data visualization – a submission to a data viz contest where I would hopefully demonstrate my skills and take my place in the Iron Viz stadium. However, the content and context of this story needed to be my guiding principle and the viz needed to be sombre, stark and subdued.
With food and food production being central to the story, I sought inspiration from the world around us when selecting the palette for this viz. Van Gogh’s “Farmers planting potatoes” captured some of the mood and landscape I wanted to portray.I used this piece to build my palette. Now I needed to build my visualization(s).
Initially I was tempted to build schmick geographic maps using the Tableau Mapbox connector, and I was in love with my outputs – but I remembered something a wise man once said “don’t use maps just because you can”. I had honed my story to focus on three regions, and three regions didn’t need a map. The map was scrapped.
I wanted to show some tricks, so I decided to create a visual of stacked blocks – with each ‘block’ representing 500,000,000 people. This way I could show the number of ‘blocks’ we can feed, and those we can not. Unfortunately ‘blocks’ lose their punch when they are so tiny that they are almost indecipherable from one another. The blocks were scrapped.
I was torn. I needed a visualization that would pack a punch, whilst at the same time be humble and allow the focus of this viz to be the story within. I dusted-off the old faithful line chart. With some minor tweaks, such as highlighting particular years, annotations and as much subtlety as I could muster in the various layers of formatting, I had my charts.
I had a story, I had some charts; I just needed something to weave it all together.
I turned to the work of Cole Nussbaumer Knaflic to guide me through these most important final steps. My “ah ha!” moment came when I came across her post that encouraged us to lead with story. In my quest to ‘reduce clutter’, ‘keep it clean’ and ‘visualize the data’ I had forgotten to speak. I had forgotten to share the words and the phrases that were swimming in my head. I needed to thread my story together with words – I needed my words to translate the devastation and the catastrophic conditions were facing if we did not commit to act.
Tableau’s invested a lot of time and effort in the development of ‘story points’ – but I’m not their biggest fan: they take up a lot of space, they are limited in their formatting, and in the world of beauty that can be created in Tableau – they feel generic. I wanted the effect of story points, but I didn’t want to use story points. So I made my own.Enter paper, pen, parameters and some unthinkable ‘pop & swap’ prowess. I like to refer to this process as ‘playing tetris’ – I need to plot every move and every possible consequence until I get the final sequence that I then replicate by way of nested layout containers, calculated fields and multiple worksheets to push and prod my story into place. The goal here is to convince you that you’re looking at one simple, clean viz. The reality is that you’re looking at a championship tetris match – masterfully hidden and skillfully played out with a simple slide-bar control.
This literally took a couple of hours – one false move and tetris won. Fortunately for me, Iron Viz allowed me to work in the comfort of my own home – rather than on the ‘floor’ at a client site – where I can pontificate, curse, scream and ultimate celebrate in private! I won. Game over – tetris!
The viz is done. It’s over to the judges now. I wasn’t expecting to find the story I did – I wasn’t expecting the enormity of the problem to smack me in the face like it did. I hope this submission does two things: firstly convince someone do something – we can’t afford to do nothing, and secondly take me to Austin – although that wish feels pathetically selfish now.
On so many fronts, we will have to wait and see.
* All calculations are defined in the ‘Supporting Information’ tab of the published viz.
Join my next masterclass!
Register now to join my next Tableau for Business Masterclass - a free webinar packed with tips to take your Tableau skills to the next level!