The Mining Data Mart - Data Mining Techniques in CRM: Inside Customer Segmentation - page 161

• Outgoing traffic to identify outgoing usage patterns (number and traffic by call
and destination type) over time.
• Incoming traffic to identify incoming usage patterns (number and traffic by call
and origin type) over time.
• Financial information such as profit and cost for each customer (billed revenue,
interconnection cost, incoming revenue, etc.).
• Purchases of handsets and accessories (e.g., hands free, Bluetooth, etc.).
• Value added services (VAS) such as news, music, etc.
• Campaign contacts and responses.
• Recorded complaint outcomes and results.
• Billing, payment, and credit history (average number of days till payment,
number of times in suspension, average time remaining in suspension, etc.).
A period of two years is proposed as the time frame for the data mart in order
to support advanced data mining applications.
Finally, the data mart should be updated on a continuous basis to include the
most recent view of each customer. The database diagram in Figure 4.4 shows
an example of an indicative schema which depicts the established relationships
between the tables of CDRs, customers and contracts, outlining some of the main
dimensions of information that underlie the mobile telephony domain.
The proposed data mart presented below refers tomobile telephony consumer
(residential) customers. Business accounts typically includemultiple lines in a single
Figure 4.4 An example of a database diagram for a mobile telephony data mart.