Market Basket Analysis (also called MBA and Association Mining) is a data mining technique that is based on the idea that if you buy a certain product, you will be more or less likely to purchase one or more additional items.
For example, if you shop at Amazon.com, then you’ve read their recommendations like . . . ,“ Customers who purchased this book also bought. . .”, . . .,“ Amazon also recommends. . .”, “ Buy these 2 books together and save . . . .” Amazon.com has been successful by using this data mining marketing strategy to generate additional sales through product promotion, cross-selling and product placement.
There are natural product affinities in the market place. For example, people who buy hot dogs frequently buy hot dog buns, ketchup and relish. Although some product affinities are obvious, others are often unexpected. A classic example involves an affinity between diapers and beer.
A number of large retail companies use market basket analysis to develop promotions and generate incremental sales. But while the analytics required for this technique are relatively simple, Market Basket Analysis requires tremendous processing capabilities to analyze transaction-level (individual market basket) data. As a result, most retailers can’t conduct this type of analysis.
THE MARKETING ANALYSTS have the servers, processing power and software needed to quickly define the likelihood of two (or more) items being purchased together. With our capabilities, we can identify the products that drive drive customers to your store and that should always be in stock. We can also analyze the contents of each basket and classify the shopping trip into a category: weekly grocery trip, special occasion, etc. Finally, we can perform store-to-store comparisons based on units sold per customer, revenue per transaction, number of items per basket, etc.