How to Measure Correlation with CORREL

CORREL measures how strongly two columns move together, from −1 to +1 — ad spend vs. sales, visits vs. revenue.

You will learn to

• Calculate a correlation coefficient
• Compare two possible drivers
• Correlate against a calculated ratio

How it works

CORREL(array1, array2) returns the Pearson correlation coefficient: +1 means the two move perfectly together, −1 perfectly opposite, and 0 means no straight-line relationship. Values above about 0.7 (or below −0.7) are usually called strong.

Correlation is a quick first test of whether one thing might drive another, but it is not proof of cause — both might depend on a third factor such as season. You can also correlate calculated arrays, e.g. =CORREL(B2:B11,D2:D11/C2:C11) to test ad spend against revenue per visit.

=CORREL(B2:B11,D2:D11)