The Data-Ink ratio is a concept first introduced by Edward Tufte—in his book the Visual Display of Quantitative Information. It can be summarized—in Tufte’s own words—by one of the five principles in the theory of data graphics:
Above all else show the data.
Or mathematically by the formula:
Data-ink ratio = data-ink / total ink used to print the graphic
Tufte splits ink used to display information into two categories: Data-ink and Non-data-ink. Data-ink is that portion of ink dedicated to represent the core of a graph—the measured quantities. Non-data-ink is ink used to draw unnecessary elements of a chart that do not contribute to clarifying the intended message.
To explain the concept lets take a really bad example. The chart below is inspired by actual graphs I've encountered during presentations I've sat through and delivered by professionals. So it’s real—no fiction here. If you’ve been long enough in business, I’m sure that you’ve come across similar graphs regularly. The problem with the graph below is the large amount of ink used that doesn’t contribute to our reasoning about quantitative information. We can completely remove this ink without reducing the amount of data communicated by the graph. In fact, removing this non-data-ink dramatically improves the legibility of the chart.