Sometimes the data sets that seem the simplest can provide some serious challenges. This week’s data consisted of 20 records: city names and square meters. That was all. The community took various approaches to visualize the data and during Viz Review we noticed that there were actually a few who struggled with the dataset and the information it held. Nevertheless we were treated to a number of great visualizations, including many bar charts which turned out to be the easiest way to represent the data.
LESSON 1: ITERATE TO IMPROVE YOUR VIZ
Makeover Monday is not a competition and we have designed the weekly ‘flow’ of the project in a way that gives everyone a chance to submit, review, get feedback and make changes as the week goes on. Something that stood out this week in particular was how many people kept making small changes to their visualizations until they told the story more effectively.
The first one of those was actually Andy. He published his visualization and when I looked at it my first thought was ‘huh?’. In that moment my face probably resembled this emoji 🤪. A couple of people in addition to me questioned his representation of the data. Andy noticed that he had misunderstood the article and made changes until his viz was showing the right stuff.
Another person who went through a few iterations until he arrived at his final version was Rodrigo.
We originally gave Rodrigo the following feedback during viz review:
- The title is inaccurate: These are not the world’s largest cities
- What is a wage/square metre gap? How is it relevant?
- How are the cities sorted? Without understanding the relevance of this ‘gap’ concept, the sort order seems strange
- The subtitle doesn’t make things any clearer.
He went and made a bunch of changes resulting in a completely new viz.
I still had more feedback because I’m a big fan of getting the details right and because Rodrigo has participated so many times that he’s happy to focus on the really small things as well. That’s what takes his vizzes to the next level, getting that detail right that can go unnoticed but makes a big difference.
- Move the cheap places to the lower half and the expensive ones to the top by flipping the y-axis
- Change the title because it still isn’t correct.
As you can see, taking feedback on board and making a few changes helps you get to better visualizations.
Taking the time to assess the feedback, what the results of the changes are and how to beat incorporate it will pay off in the long run.
It is also a great way to practice stakeholder interactions as business users you design visualizations for in your dashboard will more than likely provide you with feedback as well.
Consider the feedback and discussions on Viz Review and the social media channels as a way to practice the (often more difficult) conversations you may be having in your world environment with stakeholders.
LESSON 2: SHAPES AND ICONS
This week I noticed quite a few visualizations which used icons or shapes to represent data points. Most of the designs unfortunately were much harder to understand and make sense of than if the authors had chosen simple bars or squares.
I understand the desire to get creative and try something out, but it should not make the data even harder to understand than the original viz.
One thing Andy reminded everyone of during Viz Review was that these weekly challenges are called ‘Makeover Monday’ for a reason. The idea of a Makeover is to create something that communicates the information more effectively than the original.
There is certainly room for creativity and for bending, breaking and ignoring rules. Just keep in mind that your designs, no matter what direction you take them, should still communicate the information in a way that is accessible for your audience.
The challenge with shapes, especially irregular shapes like houses, prison cells, shapes of people, money bags and others that appeared in vizzes this week, is that judging their size relative to each other is incredibly hard. Comparing the length of a bar or size of a square in comparison is easier. Even circles would be easier to compare in this case.
What if you want to use icons or shapes but still build an effective and informative visualization? Why not use the icon or shape just once as part of your design and then stick to easier visual representations of the data, e.g. through bars, in your charts?
See what works for you. By all means be creative but also remind yourself of the purpose of your visualization and what you want to achieve with it. If you want to inform and engage your audience with your insights, then, I’m afraid, simplicity and easy to understand charts will work best, even if they don’t leave much room for adventurous designs.