My entry for the Forbes’ Graph Makeover Contest

I stumbled accross Naomi Ribbins challenge on a blog entry at Forbes on the re-design of a simple pie chart. I’m usually quite a fan of challenges (I try to maintain this contest page updated with current/past challenges in dataviz) and decided to give it a shot.

The challenge is twofold, but I’m only adressing the first one here: redesign of a social media referral trafic pie chart. Below is the orginal graphics to re-design:

It’s actually not too controversial (beside the choice of a pie chart than any other graphical representation). Naomi didn’t really express any reason for picking up this particular chart (it issued from a series of many other ones in Eloqua’s report on marketing data).

My suggestion to the redesign is as follows:

I wanted to emphasize the streaming nature of the data to depict (Internet traffic). I used a bar chart because I think this is better for comparison than pie chart. Each color circle in the charts do not represent an actual value (such as a percent or visitor on the websites) but their simultaneity provide a sens of the proportion of the flows (% of the traffic share).


After a short amount of time, the accumulation of the circles create the bar charts and the oldest circles (at the bottom) shrink and disappear to leave room for the new ones (coming from the top). The resulting chart clearly shows the difference of traffic between each site (Facebook being consistently dominant). It also shows the drop of LinkedIn traffic between B2B and B2C companies.


I used our toolkit called Visual Sedimentation to desgin those charts (still at an alpha stage but worth a try!).


Below is a static version of the chart:


Posted in Uncategorized