Download CRM-Market Basket Analysis (MBA)-Notes-Business Administration and more Study notes Business Administration in PDF only on Docsity!
Market Basket Analysis (MBA)
Market Basket Analysis is one of the most common and useful types of data analysis for marketing. It is an algorithm that examines a long list of transactions in order to determine which items are most frequently purchased together. The purpose of market basket analysis is to determine what products customers purchase together. It takes its name from the idea of customers throwing all their purchases into a shopping cart (a "market basket") during grocery shopping. Knowing what products people purchase as a group can be very helpful to a retailer or to any other company. A store could use this information to place products frequently sold together into the same area. Direct marketers could use the basket analysis results to determine what new products to offer their prior customers. The strength of market basket analysis is that by using computer data mining tools, it is possible to find out, what products consumers would logically buy together. Once it is known that customers who buy one product are likely to buy another, it is possible for the company to market the products together, or to make the purchasers of one product the target prospects for another. If it's known that customers who buy a sweater from a certain mail-order catalog have a propensity toward buying a jacket from the same catalog, sales of jackets can be increased by having the telephone representatives describe and offer the jacket to anyone who calls in to order the sweater. By targeting customers who are already known to be likely buyers, the effectiveness of marketing is significantly increased. This is the purpose of market basket analysis – to improve the effectiveness of marketing and sales tactics using customer data already available to the company. 4.7.1. Methodology
The input to a Market Basket Analysis is normally a list of sales transactions, where each column represents a product and each row represents either a sale or a customer, depending on whether the goal of the analysis is to find which items sell together at the same time, or to the same person. In order to perform market basket analysis, it is necessary to first have a list of transactions and what was purchased in each one. For example, consider the transactions of convenience store customers, each of whom bought only a few items :
Transaction 1: Frozen pizza, cola, milk Transaction 2: Milk, potato chips Transaction 3: Cola, frozen pizza Transaction 4: Milk, pretzels Transaction 5: Cola, pretzels Each customer purchased a different basket of items, and at first glance, there is no obvious relationship between any of the items purchased and any other item. The first step of a basket analysis is to cross-tabulate the data into a table, allowing to see how often products occurred together. For these five convenience store purchases, the table looks as follows : Frozen Pizza Milk Cola Potato Chips Pretzels Frozen Pizza 2 1 2 0 0 Milk 1 3 1 1 1 Cola 2 1 3 0 1 Potato Chips 0 1 0 1 0 Pretzels 0 1 1 0 2
The central diagonal of the table shows how often each item was purchased with itself. Though this is significant for figuring some reliability statistics, it does not show how items sell together. In the first row, out of the people who bought frozen pizza, one bought milk, two bought cola, and none bought potato chips or pretzels. This hints at the fact that frozen pizza and cola may sell well together, and should be placed side-by-side in the convenience store. Looking over the rest of the table, there is nowhere else that an item sold together with another item that frequently. Hence, this is probably an actual cross-selling opportunity. Compare this to the second row of people who bought milk, one bought frozen pizza, one bought cola, one bought potato chips, and one bought pretzels. It seems milk sells well with everything in the store. There is probably not a good cross- selling opportunity with milk. This makes sense for a convenience store. People often come to a convenience store for the purpose of buying milk, and will buy it regardless of anything else they're looking for. In the real world, there would usually be more than five products, and would always have more than five transactions to look at. As a result, the distinction between products that sell well together and products that do not would be much sharper. Hence,
catalog merchants get the same benefit, by conveniently organizing their catalog or
Web site so that items that sell together are found together.
Outside of the store environment, basket analysis provides different benefits. For a direct marketer, it is far preferable to market to existing customers, which are known to buy products and have a history with the company. The company already has these people in its database, and knows a significant amount of information about them. After running a basket analysis, a direct marketer can contact its prior customers with information about new products that have been shown to sell well with the products they've already bought. Even when making sales to new customers, telephone representatives can offer buyers of a product discounts on any other products they know sell with it, in order to increase the size of the sale. Market basket analysis has uses even outside the realm of marketing. It can be useful for operations purposes to know which products sell together in order to stock inventory. Running out of one item can affect sales of associated items. The reorder point of a product should be based on the inventory levels of several products, rather than just one. Basket analysis can be used in any case where several different conditions lead to a result. For example, by studying the occurrence of side effects in patients with multiple prescriptions, a hospital could find previously unknown drug interactions about which to warn patients. There are several advantages to market basket analysis over other types of data mining. It is undirected. It is not necessary to choose a product that you want to focus on in order to run a basket analysis. Instead, all products are considered, and the data mining software reveals which products are most important to the analysis. In addition, the results of basket analysis are clear, understandable association rules that lend themselves to being immediately acted upon, and the individual calculations involved are simple.
4.7.3. Limitations
Though an useful and productive type of marketing data mining, market basket analysis does have a few limitations.
offering promotions such that the buyers of one item get discounts on another they have been found likely to buy, sales of both items may be increased. In addition, basket analysis can be useful for direct marketers fo r reducing the number of mailings or calls that need to be made. By calling only customers who have shown themselves likely to want a product, the cost of marketing can be reduced while the response rate is increased.
Other application areas
Although Market Basket Analysis conjures up pictures of shopping carts and supermarket shoppers, there are many other areas in which it can be applied. These include: Analysis of credit card purchases. Analysis of telephone calling patterns. Identification of fraudulent medical insurance claims. Analysis of telecom service purchases. Note that despite the terminology, there is no requirement for all the items to be
purchased at the same time. The algorithms can be adapted to look at a sequence of
purchases (or events) spread out over time. A predictive market basket analysis can be used to identify sets of item purchases (or events) that generally occur in sequence : something of interest to direct marketers, criminologists and many others.
