The Mining Data Mart - Data Mining Techniques in CRM: Inside Customer Segmentation - page 177

Examples of Derived Measures Used to Enrich the MCIF

An indicative list of more complex KPI examples for mobile telephony is as
follows:

• Percentage (%) of peak calls: outgoing peak calls as a percentage of total
outgoing calls.
• Percentage (%) of working day calls.
• Percentage (%) of SMS calls: outgoing SMS calls as a percentage of total
outgoing calls.
• Percentage (%) of on-net outgoing community.
• Percentage (%) of the total number of calls made to the three most
frequently called numbers.
• Ratio of total outgoing to total incoming calls.
• Percentage (%) of outgoing voice calls with a short duration ( < 60 seconds).

• Number of distinct months with roaming calls.
• Outgoing usage delta: compares the total outgoing calls (or minutes) of
the most recent period, for example, the last three months, to those
of the previous three months and denotes the respective percentage
difference.
• Flag of Internet usage.
• Coefficient of variation of outgoing calls: a measure of the weekly varia-
tion of outgoing calls, defined as the standard deviation of the weekly
number of outgoing calls divided by the weekly mean. It requires
weekly aggregates of call data.
• Number of different service (call) types.
• Number of rate plan changes (divided by the respective tenure).

THE MINING DATA MART FOR RETAILERS

As in any other industry, retailers need to continuously monitor the relationship
with their customers. Fortunately, all customer transactions are recorded in detail
at the point of sale (POS), logging information on:

• What: The detailed universal product code (UPC) of the items purchased.
• When: Time/day.