The Mining Data Mart - Data Mining Techniques in CRM: Inside Customer Segmentation - page 178

• Where: The specific store/department/channel concerned.
• How: The payment type, for example, cash, credit card, and so on, of each
purchase.
Data obtained from the point of sale, when appropriately mined, can lead to
a clear identification of customer characteristics. A specific problem that retailers
have to tackle in order to track the transaction ''trail'' of each customer is the
''personalization'' of the purchase data. A form of identification is required to link
transactional data to the customers. This identification is usually made possible
through a loyalty scheme. A loyalty scheme's main purpose is to tighten the
relationship with the customers and to promote loyalty through rewards and bonus
points. Furthermore, it also provides a means of collecting customer information
and of relating the transactions to the customers.
The scope of a mining data mart for retailers is to provide an integrated view
of all customer attributes, allowing customer behavior to be monitored over time.
A key concept in retailing data is the ''transaction.'' Hereafter the term
transaction will refer to each purchase event. A transaction can be thought of as a
purchase basket or a discrete order and typically involves the purchase of multiple
product items.
The information contained in the retailing data mart should capture all
transactions and all customer-related information, covering, at a minimum, the
following data dimensions:
• Registration data, collected during the loyalty scheme application process.
• RFM attributes, overall or for each product category. More specifically:
- The time since the last transaction (purchase event) or since the most recent
visit day.
- The frequency (number) of transactions or number of days with purchases.
- The value of total purchases.
• Relative spending in each product category/subcategory.
• Preferred stores/departments.
• Time and day of customer visits, also including special occasions (summer,
Christmas, etc.) and sales periods.
• Payment (cash, credit card, etc.) type and channel of purchase (store, phone,
Internet).
• Campaign contacts and responses.
• Use of vouchers and reward points.
• Recorded complaints and results.