This simple, and in some ways counter-intuitive, principle is one of the main tenets of the philosophy of the famous designer Dieter Rams. He is known for his industrial design masterpieces, characterized by an extremely minimal appearance paired with unbeatable functionality. Looking at his watches or electronic shavers, one gets surprised by how simple and pure their shapes are. Using them, you discover how wonderfully they work and how every detail seems to be designed to serve the general purpose of the product.

How is this possible? How is it possible to create a fully functional product by removing features and components rather than adding them? This can be done by following the second part of the principle: better.

Dieter Rams' philosophy is to work around the object, working out its purpose, deciding which components are essential and what can be actually removed. The focus is dedicated to the essential components, as they clearly emerge once the additional ones are removed. 

How does this translate into data visualization? We can answer the question by looking at the following plots:

I hope that there is no doubt which is the wrong plot (if any doubt arises, the titles should help). We can focus our attention on why the second plot is awful.

The two plots show exactly the same data, but while the first one lets you immediately focus on what is relevant, the smooth increasing trend, the second one has a prominent background, pointless shadows, and meaningless 3D effects.

Why can't our brains simply focus on what is relevant in the second plot, avoiding looking at the useless details we just listed? Because our brain is crafted to try to acquire a general sense of what it gets in front of it and is not able to process all the information at once. Have a look at the additional reading referenced at the end of the chapter for more on this. Since our brain is not able to immediately focus on the trend, it starts looking around the plot, looking at the background, the shadows, and the 3D effects. 

We can bet that if confronted with such a plot, our audience will remember it, but we can also bet that they will probably not remember its main message, which is the trend (no, it is not the papyrus background).

So, less but better—when creating a plot, do not simply add all you can add to it, but focus on what is strictly essential to communicate your message. Is that vertical grid needed on your time chart plot? Probably not, so you can remove it. Have you got any good reason to mark the y axis labels in bold? Probably not, and you should probably make them as plain as the x axis labels. It may require some practice, but focusing on what is really relevant in your data visualizations will dramatically improve their appearance and effectiveness.