Here is an interesting chart from Economist, ironically titled The weight of the world. Can you tell what is wrong with it?
That’s right. The coloring of dots is unnecessary. The color scale dots add an extra layer of complexity to the chart. They are redundant.
So there you go. A simple but effective rule. Don’t add redundant colors or extra layers to your chart. Your users are smart, so let them decode the chart.
Make awesome charts every time…
Check out these additional charting principles and case studies to make awesome charts.
4 Responses to “Don’t make your charts heavier than they should be – The weight of the world chart [case study]”
I hear ya on not adding too much, but actually I was just looking for something like this. (sort of)
I have a scatter graph with Store #'s and data points of sales per labor hour with multiple stores. Some but not all stores are designated "Training stores" . As of right now, all the little "diamond shapes"(diamond shapes are just what management wants to see and they serve the same purpose as the "dots" on the above chart), anyway, all the "diamond shapes on the scatter graph are the same color. I'd like to make the training stores a different color. Any thoughts ?
Thanks very much
~Keith
@Keith
I would use 2 or 3 series for this
I would use a separate series for each Store Type
Then a 3rd series for the line if you have a line joining all the points
You will need to setup formulas to extract the data for each series separately
If you want specific help can you please post the question at the Chandoo.org Forums
http://forum.chandoo.org/
attach a sample file for a quicker more targeted response
If anything, if they want to focus on the differences from the averages they should have gone with a residual plot.
Very good written article. It will be helpful to anybody who utilizes it, including me.
Keep doing what you are doing - can'r wait to
read more posts.