In marketing, there is a term called Symbiotic Marketing in which two complementary products are bundled for promoting sales, e.g. washing machine and washing powder, rice cooker and rice, etc., but this kind of sales promotion is limited to products with a known symbiotic relationship. Yet, statistics can help break through such sales model and combine different categories of products for sale.
One of the well-known examples is the co-marketing of diapers and beer. Walmart, an American department store, collected data on weekend purchasing records of 3-person families with a new-born baby for analysis, and found out a significant correlation between purchases of diapers and beer in terms of customer and time of purchase. On weekends, new fathers would go to buy diapers at the supermarket and give their wives a break at home, and very often they would get some beer for themselves as well. Based on these results, the company decided to place beer next to the diapers, and move the higher-priced diapers that were not selling well close to the beer. The fathers tend to shop for convenience and may not compare prices of diapers. As a result, the business reported a 30% increase in sales. This unprecedented success is largely attributed to the collection and analysis of statistics. Diapers and beer are completely unrelated by conventional thinking. Without the help of statistics, it would be impossible for Walmart to discover the valuable patterns from the data.