The Mining Data Mart - Data Mining Techniques in CRM: Inside Customer Segmentation - page 134

combine all important blocks of information in a central repository that can enable
the organization to have a complete a view of each customer.
The mining data mart should:
• Integrate data from all relevant sources.
• Provide a complete view of the customer by including all the attributes that
characterize each customer and his or her relationship with the organization.
• Contain preprocessed information, summarized at the minimum level of inter-
est, for instance at a product account or customer level. To facilitate data
preparation for mining purposes, preliminary aggregations and calculations
should be integrated into the building and updating process of the data mart.
• Be updated on a regular and frequent basis to summarize the current view of
the customer.
• Cover a sufficient time period (enough days or months, depending on the
specific situation) so that the relevant data can reveal stable and non-volatile
behavioral patterns.
• Contain current and past data so that the view of the customer can be examined
at different moments in time. This is necessary since in many data mining
projects analysts have to examine historical data and analyze customers before
the occurrence of a specific event, for instance before purchasing an additional
product or before churning to the competition.
• Cover the requirements of the majority of the upcoming mining tasks, with-
out the need for additional implementations and interventions from IT. The
designed data mart could not possibly cover all the needs that might arise in the
future. After all, there is always the possibility of extracting additional data from
the original data sources or for preparing the original data in a different way. Its
purpose is to provide rapid access to commonly used data and to support the
most important and most common mining tasks. There is a thin line between
incorporating too much or too little information. Although there is no rule of
thumb suitable for all situations, it would be useful to keep in mind that raw
transactional/operational data may provide a depth of information but they also
slow down performance and complicate the data preparation procedure. At the
other end, high-level aggregations may depreciate the predictive power hidden
in detailed data. In conclusion, the data mart should be designed to be as simple
as possible with the crucial mining operations in mind. Falling into the trap of
designing the ''mother of all data marts'' will most probably lead to a complicated
solution, no simpler than the raw transactional data it was supposed to replace.
In the following sections we will present mining data mart proposals for
mobile telephony (consumer customers), retail banking, and retailing. Obviously,
the amount of available data, the needs, and the requirements of each organization
differ. Each case merits special consideration and design, which is why these