Correlation is a strong tool to find something hidden in the data.

You may effectively find the bottelneck, root-cause, effectiveness or relationship among thda data sets, which will definitely contribute to your business performance.

One of the advantages of correlation is the fact that you can get the result in fice seconds when you use Excel. In that case you can easily do as many trials as possible until reaching a reasonable answer in a short time.

Following sample shows the data of the sales revenue and marketing expense of each week.

You may want to know whether the expense is effecively and efficiently used to maximize the revenue. But many people waste the money without knowing its effectiveness (i.e. correlation to the sales revenue).

After the Excel function “CORREL”, select the two data sets, with “,” in between.

Then you will get a correlation coefficient in the cell (-1 to 1).

In this case the coefficient is 0.71. There is no official threshold to judge correlation. But in most cases, this is the common sense:

-1.0 ~ -0.7: Strongly correlated (opposite direction)

-0.7 ~ -0.5 : correlated (opposite direction)

-0.5 ~ 0.5: weak / not correlated

0.5 ~ 0.7: correlated

0.7~ 1.0: Strongly correlated