For instance, an e-commerce merchant who also has a "real" retail store may wish to correlate his online sales with his in-store sales. If such a merchant were to discover that his correlation was falling well below that of the one found here based on overall U.S. Census data, he may determine that he is not meeting his potential for sales in the e-commerce world. For example, if this individual found that his e-commerce sales were correlated with his in-store sales at a much lower level, such as a correlation of less than .4, this merchant may decide to invest more money into his e-commerce business in an attempt to increase his sales online. Knowing the overall correlation based on U.S. Census data provides a "goal" or standard of sorts that a merchant can compare him or herself to, thus allowing the merchant to determine that perhaps his lower sales online are related to his poor website functioning, as opposed to simply assuming that consumers do not purchase things online as often as they purchase things in person. Examining correlations could also help market researchers to determine which segments of the population prefer online shopping. By dividing a sample based on socio-economic status or income level, researchers could then compare correlation coefficients between the groups, examining whether or not wealthy individuals have a similar correlation between their e-commerce purchases and their in-store...

A correlation would be useful for this type of analysis because the purchasing rates of these groups would vary greatly based on income, but the correlation between their two types of purchases would not necessarily vary based on income. If a variance was found between the groups, then researchers could interpret this variance as an indication that income level may be a predictor of online purchases. It could potentially be concluded that online purchases are more "frivolous" purchases, and thus perhaps more common among those with higher levels of disposable income (Groebner, 2004).
There are a number of different areas in which correlations can be applied to help with the decision-making processes that occur within business settings. One simply needs to devise the right questions and then obtain the data for the variables involved and calculate the correlation. The quick and easy calculations could help add an extra level of assurance to the decision-making process, providing statistical evidence to help back up one course of action over another.

References

Groebner, D.F. (2004). Business Statistics: A Decision-Making Approach. New Jersey: Prentice Hall.

U.S. Census Bureau news. (2010, February 16). Retrieved March 16, 2010, from the Census Bureau Web site: http://www.census.gov/retail/mrts/www/data/pdf/09Q4.pdf