When related data is displayed in Visual Designs, new discoveries are possible, turning the raw data into actionable information.  Visual Designs allow the brain to discover meaningful relationships that may be used to prioritize a business plan and actualize objectives.

Graphical representations that display correlated data (traditional charts and graphs) have been around for a long time.  At an elemental level, the purpose is typically an effort to discover information unknown by looking directly at the data elements.

Consider the example below from Anscrombe: Four sets of data each of which has the same statistical measures. All four sets of data have 11 rows (n=11), the mean of the x‘s is 9, and the mean of the y‘s is 7.5. Each set of numbers has the same regression line Y = 3 .5X.

These four columns have the same statistical measures.  If we were going to analyze these numbers statistically, we may make the same conclusions to all four sets of data.

From a statistical perspective, these four data sets appear the same. However, when displayed visually, the data sets appear entirely different:

In the above example, using a data visualization may well lead us to different conclusions (and hence actions) in
respect of the above four data sets. 
 
Note that this example utilizes only one traditional type of picture – a graph. Such a traditional method for turning volumes of data into information, however useful, may well be considered obsolete when placed within the current age of large amounts of data stored in enormous data warehouses.

Source:  F.J. Anscombe, “Graphics in Statistical Analysis”, American Statistician, 37 (February 1973), 17-21