A bell curve is a graphical depiction of a normal distribution. Bell curves have distinct characteristics which enable predictions and inferences. The highest point of a bell curve is the average for the thing being measured. Things which fall into a normal distribution and can be shown as a bell curve obey the 68/95/99.7 rule. This specifies that 68% of the measures are found symmetrically distributed within one standard deviation of the mean, 95% are within two, and 99.7% are found within three.
This allows for a prediction that for any phenomenon that is normally distributed, 68% of the measures you take will fall within one standard deviation of the mean. For example, if measuring the daily website traffic of similar B2B brands, any given day’s web traffic measurement will fall within one standard deviation of the average on 68% of days.
Of course, in this example, changes in a company’s marketing, introduction of new products and the like, may cause changes in the traffic that fall outside of one standard deviation, or are even outliers (outside of 3 standard deviations).
