Info Aperture is a blog about information design by Kate M.

Telling the "whole story:" October 2019 SWD Challenge

Telling the "whole story:" October 2019 SWD Challenge

October SWD_FINAL-01.png

For this month’s SWD challenge we were asked to improve upon this chart:

Original SWD Challenge chart.

Original SWD Challenge chart.

This month’s challenge had two steps:

  1. How would you improve the actual chart to help people understand the data?

  2. What visualization might you use to help people compare revenue.

As a viewer, I’m bad at math, and would love to see the totals added up for me. So that was my first addition. I’m glad I did this because I immediately saw that our sales revenue does not add up to 100%. Doing a little bit of basic algebra, I was able to define an “others” column which accounts for the missing revenue. Next, I started looking at the individual columns. I immediately stood out to me the vast differences in # of accounts. Our highest account tier (C) had signed 406 MORE accounts then our lowest account signing tier (A+), and yet, this was not reflected in their % of revenue. This interested me so to further investigate, I assumed the distribution of our account $ was normal and added a new column looking at the average account $ of each tier. This was super interesting. We had vast differences between our tiers. Which to me indicates the original chart is only part of the story. I think it would be a mistake to assume each one of these tiers were approaching new accounts in the same way. What are the strategies of each tier? Maybe Tiers A & A+ are selling to larger companies or are specifically designated to find larger accounts, while our other tiers, like B, C, D are going after the “smaller fish.” Both have their values in business, especially since the sales revenue of these smaller accounts (>$11,000) still generated 54% of all the revenue.

My chart (all of my additions are in red).

My chart (all of my additions are in red).

I really enjoyed seeing everyone else’s approach to the second step on twitter this month. I like how some people “cut right to the chase” and presented their major finding in their graph heading, using color to connect their data with their words. I also enjoyed seeing the different trade-offs and decisions people made. Sure, a stacked bar graph makes accounts/revenue share more easily comparable, but it might be harder to see each Tier’s individual contributions and the nuances of their account signings.

I wanted to practice my storytelling skills by leading people through the chart data with an alluvial diagram. I decided to focus more on the tier differences and the # of accounts and the (assumed) average account amount. I chose this diagram for a few reasons:

  1. It’s pretty- Maybe this isn’t the most important thing in a conference room with a bunch of decision-makers in a multi-million-dollar company, I wouldn’t know, I’ve worked in the public sector my entire career, but personally, I place high value on a beautiful looking visualization. I also think it’s kind to your audience, no matter who they are to give them something nice to look at.

  2. Alluvials are typically used for finances so I felt like this was the right application of the graph.

  3. I liked the idea of having people be able to compare the revenue amount by the thickness of the tier lines as we went through several levels of analysis. Which keeps that comparison factor at play even as we look at averages and account #s.

I used icons to keep people oriented despite the line thickness changing order as we go through the analysis, more revenue, accounts, etc. are always on the left while less is on the right. I achieved this ordering by making three separate alluvials in RAWgraphs and stacking them, making sure our tiers lined up. Finishing touches were done in Illustrator. I also took the time to comprehensively annotate and point out certain things about the data.

While I don’t think this would make a good PowerPoint slide, I think it would be great as a printed handout. This might be “old school” but it gives people something to look at and take with them. If you ever have the opportunity to attend an Edward Tufte workshop (do it) you’ll find he spends a good chunk of time discussing how paper is still our “highest resolution” and that PowerPoint is a “crutch.” Anyways, If I were presenting, I think I’d have separate slides for each level and use isolated coloring when I was talking about specific tiers but would still give people the handout.   

Thanks for the opportunity SWD!

Making Every one Count: Fun with small data sets

Making Every one Count: Fun with small data sets

You Are What You Absorb: uncertainty in counting calories

You Are What You Absorb: uncertainty in counting calories