Feels right and wrong at the same time to write here. Right because has been over 12 months since I last posted, it was time. It’s incredibly embarrassing at the same time. It feels like I’ve neglected the blog for so long, even though I’ve been blogging over on InterWorks and interviewing lots of data visualization practitioners as part of the Podcast Your Data podcast. I have however, a need to reconnect with visualization a lot more than I did in 2018. Time as of late is something that I struggle with but I have a healthy list of areas and things I want to work on. With that in mind I’d like to share my first contribution to the #SWDChallenge.
It stands for Storytelling with Data spearheaded by Cole Knaflic from storytellingwithdata.com, Cole explains it below:
“The #SWDchallenge is a monthly challenge where you can practice and apply data visualization and storytelling skills. Think of this as a safe space to try something new: test out a new tool, technique, or approach. There’s no obligation—participate in one, a couple, or all. We encourage everyone to take part: all backgrounds, experience levels, and tools used are welcome. Each month has a different topic, announced at the beginning of the month. Participants have a week to find data, create and share their visual and related commentary. We’re focusing currently mainly on different graph types, but sometimes change it up with a tip to try or specific topic. This is meant to be a fun reason to flex your skills and share your creations.” Cole Knaflic – www.storytellingwithdata.com
The first challenge of January required us to use a visualization tool that’s unfamiliar to us. Repeating a visualization we have in a tool we know and try and repeat it in another tool and writing on our findings.
Since time is short and this was my first foray into the challenge I decided to keep it simple and use IMDB data and a chart in a Tableau, my data visualization tool of choice.
Visualization created using Tableau – data from IMDB
In the chart above I had a look at Budget for a movie vs its Revenue. I made a conscious decision not to make many changes to the scatterplot and not to spend too long on formatting. Aside from the colour of the dots, the rest of the visualization is all a default of Tableau. Since I’m pretty new to the other tool it would be unfair to compare defaults on one tool vs heavily edited and formatted views on another. The annotations work well out of the box, but I wish it formatted the numbers to Millions for instance, similar to the axis.
For the tool that I’m unfamiliar with I choose Flourish. I had the chance to hear from Anna Powell-Smith CPO at Flourish last summer and I’ve been curious about the product ever since. It was the perfect opportunity to get to use it.
Registering was super easy and since Flourish is web based I was up and running in less than 2 minutes with no installs. Flourish has a tutorial of sorts as soon as you open it but I skipped most of it. 😉 I thought i could just click on things until I got the results I wanted. Importing data is straight forward for the most part. My only criticism is that the tab stops responding when a heavy task is being carried out and it makes it seem like you either didn’t click the button or that it failed. A little bit of patience is required. I also think this is a reflection on myself since I’m not used to carrying out load tasks in a web based tool and I should expect some changes from an installed or local tool.
Once the data was loaded it was easy to see and change the order of the columns. To start visualizing it, first you pick a chart and only after, you get to the data. Which for me personally feels the wrong way around. Being used to grab data and just change chart types on the fly is something that I always cherish in Tableau as one of the best features of the product. While Flourish allows you to create very complex visualizations such as chord-charts it forces you to choose the chart in advance before exploring the data.
How to choose which measure should be used in the scatterplott were not very intuitive, it took me a good 5 minutes before I realized that I just needed to type the column letter in the boxes you see on the right side above in shades of pick and purple. Simple when you know it, but not intuitive.
Visualization created using Flourish – data from IMDB
The chart above was the final result and while I like some of the potential of flourish there are some things that I struggled with. Flourish arranges all the menus on the right hand-side and I saw a few interesting ones like animation which made think that I could at some point adding animation to a viz. Cool option for sure.
I managed to change the colours and add the title to tool tips. I tried to change annotations to mimic the ones from Tableau but it didn’t do anything. The biggest concern for me was the Y axis which I was unable to have it all showing no matter what I tried. I also pinged Flourish on twitter and hopefully they will come back and tell me what I did wrong.
UPDATE The team at Flourish came back and told me to change the margins to show the axis correctly. It was very simple again, just not super intuitive and the marks cover part of the axis label which is a bit odd. Flourish has also said that new templates are coming and this should be fixed soon, that way the axis will show as default.
Updated version after Flourish feedback
All in all I think flourish has a lot of potential and cool things going for it. But I didn’t feel like it was very intuitive. What I also noticed was that no matter the tool, formatting is such a pivotal part of software and something that most tools fail at. Tableau is great but it’s formatting pane is far from user friendly and straightforward. Flourish seems to have similar challenges. I really enjoyed this first challenge and look forward to trying out the next one in February.
Thanks for reading