I am a very big believer that visualizations created in a professional setting (where users derive insights and make decisions based on data) should be designed with analysis first. At work I'm heavily influenced by the teaching of Few and Tufte. If you are creating a KPI dashboard, you shouldn't include flashy images, or use charts that are hard to understand. Your managers/executives need to be able to digest the information quickly, and see areas of weakness and strength. Simple sparklines, bar charts, line charts and small summarized text tables are effective and easy to comprehend. You don't need bright "wow" colors, simple grey scales and two tone colors can be used to highlight items of interest. When it comes to work, Keeping it Simple can lead to better decision making. That being said, us as Data Lovers, don't always create visualizations in a professional setting. Sometimes we do it for fun, because we just love data. I've always said, that if you are creating visualization during your own time, for fun, you should design in a way that makes you happy. Me, I can't paint, I can't draw or take amazing photographs, but I can use Math and Tableau to create what I consider my form of art. Now, just because you are designing for personal, that doesn't mean you have to be overly "artistic" with your designs. If creating traditional dashboards makes you happy, than do that. For personal projects, they are personal, so do you. I really enjoy doing both. Living on both sides of the Data Viz spectrum. At work I'm Few/Tufte, at home I'm McCandless/ Holmes.
Talking about my "Viz at Home" life, I LOVE using Math and Tableau to create curvy lines, or big images of Michael Phelps thrashing through the water as the "King of the Pool". Designing in this style helps me create metaphors with the data. I use the data and other elements to tell stories that aren't readily apparent. Sometimes it works, sometimes it doesn't, but it is the exploration and the challenge of making the metaphor with data that I really enjoy.
In some cases, data is about more than sales, or profits. It can be about sports, or food or politics, etc. With this type of data, comes factors that are outside of traditional analysis. For example, I recently did a visualization on QB Air Yards vs Yards After Catch
Here, the metaphor I was going for was how the football actually moved in yards. The Arcs represent the Ball Flying through the Air (Air Yards) where the straight lines coming off of it represent the Yards on the ground (Yards After Catch). I could have used Stacked Bar charts with different colors categorizing Air Yards vs Yards After Catch, but this was a personal project, I wanted to do more with the data. I thought that creating the Arcs and the Straight Lines added to the value of the viz in terms of context. While it might not be as easy to interpret, it does give a deeper feeling of realism (At least in my opinion). There was a factor to this data that was more than just numbers. As I explore data sets for personal projects, I try to think about what the data look like in reality, as it is happening. For data on Cars top speeds, I might think of ways I could display the data so it looks like Cars are racing against each other, I might use images of Cars, or a race track to help push that metaphor.
Playing into that, I did another Air Yards viz that looks at data from 1992-2016. Once again, I tried to focus on the Metaphor of the ball flying through the air.
As before, there are more effective ways to visualize this data, but I wanted the data to "look" like a football flying through the air. To have the user imagine a QB throwing this ball from year to year, and allow them to compare how that changed. For me, I looked at Tom Brady and thought back to the Moss years, and all those Bombs he threw down field. The goal was to create a visualization that put the user on the field, or brought them back to a memory of a game or season where their QB had a crazy (or terrible) year. And maybe they can question what happened in that year, did the team have great Wide Receivers, did they change their offensive scheme? The goal was less "Compare the numbers" and more, "Imagine this as it happened".
Most recently I created a visualization for #MakeoverMonday on the Global Peace Index. I found the data really interesting, and thought of ways I could make it represent more than the ranks/indexes (Note-More does not necessarily mean better). While exploring/fiddling, I just happened to over hear someone on tv say, "It's crazy, it's just spiraling out of control" (Not exact quote, I was half listening). Now I can't for the life of me remember what this character was even talking about, but "Spiraling out of control" stuck with me. So I was looking at the data and thought, well what if I made "Peaceful" countries be together, like in a unified circle, and I showed "Less Peaceful" countries spinning off of them. Away from the rest, "Spiraling out of control" and into war. By modifying some geometric calcs, I was able to create the curvy/spiraling viz below.
Though it isn't best practice, and it can be kind of hard to understand, it plays into the metaphor/point I was trying to make. I know not everyone is going to "get it", and some will probably hate it, but that's ok. This goes back to the idea that designing "artistic/metaphorical" vizzes for fun, don't always have to be grey scaled bar charts. It can be whatever you want it to be, just make sure you have fun doing it. Who knows, maybe you create a new chart type, that while different, actually tells the story in a clear, engaging and beautiful way. That's what is great about doing personal projects, you can experiment and see what works and what doesn't. And designing to a Metaphor can help you think outside of the box, and creative alternative vizzes that haven't been done before.
If you agree or disagree, please share your feedback/comments. I would love to engage in more conversations about it, and learn more from others perspectives.
Thanks for reading!
Rody Zakovich
1 Comment
I totally agree with this Rody! I think knowing your audience is the most important question to answer before designing a viz. It dictates the style, the message, everything. I try to keep this in mind with every viz. My work related visualizations are much different than public beer-time vizzes. If I am doing it on my own time, it needs to be fun! Nice post!
Balas