Data Mining (UPDATE)

Source: http://dssresources.com/faq/index.php?action=artikel&id=41

This article was quite interesting. It was about the relationship found between the sales of beer and diapers using data mining. Dan Power, the author of the article, is the editor of DSSResources.com, DecisionAutomation.com, and PlanningSkills.com.

It is said in the article that a pattern was found in a Midwest grocery store, that men bought diapers on Thursday and Saturday and they usually bought beer as well. Now there are several versions of this story told by different sources. And also the reasons of why beer and diapers go together are different. One source said that men usually would buy diapers on Thursday and then buy beer for the upcoming weekend to have for social events, etc. Another explanation was that when men were told to get diapers they bought beer as a reward for themselves for their good deed of running a quick errand for their families.

It is said that when the stores realized and found this pattern that they were able to capitalize on it by moving the beer closer to the diapers on Thursdays and Saturdays.  The legend of the beer and diapers story is true, there was a correlation of these purchased items. But what was really the main story was that stores changed their way of selling products. “They went from a traditional methodology of selling what was bought by consumers to a merchandizing methodology of buying what was sold”.