Not always. More often the purpose of a chart seems to be to:
- Make the presenter look smart, or at least cool.
- Make the presenter’s product look better than the competition, despite the facts.
- Accentuate (fabricate) the positive and obfuscate the negative.
- Hide the fact that the presenter doesn’t know what’s going on.
Raymond backs up his assertion with a counter-example which I presume looks like this, with barely distinguishable green gradients and a useless legend.:
You can improve the information transfer by removing the gradients, flattening the awful 3D perspective, and replacing the legend with appropriately located data labels. What do you know, slices one, two, and four are the same size.
A further improvement is made by changing the pie chart to a column chart. And what else do you know, slice two is not the same size as one and four.
So pie charts exemplify chart suck, we knew that. So what are charts really for?
You can use a chart to selectively and interactively display one set of data among many.
You can use charts in a Graphical Approach to Solve a Simple Physics Problem.
You can use charts for a lot more purposes, and better purposes, than described in the four bullet points at the top of this post. If you’re careful, you may even demonstrate Raymond’s thesis that The purpose of charts is normally to make information easier, not harder, to understand.