| STAGE |
DESCRIPTION |
COMMENTS |
| 1 |
Customer Data Integration |
Match Customers on Name, Address, etc.,
using Products like DataFlux |
| 2 |
A Single View of the Customer |
Consolidate data from different Sources
using Customer Hub Software
which Products like
Oracle's Fusion Middleware. |
| 3 |
Establish Customer Preferences |
Build an understanding of Customers, eg preferred Channel and Contact Times
which can be done simply by using SQL. |
| 4 |
Customer Profiling |
Extends previous work, and can be done simply by using SQL. |
| 5 |
Customer Services Analysis |
For example,
identify Customers receiving more than one Service,
and establish Patterns of Service Usage
This can be done simply by using SQL. |
| 6 |
Identify Good and Bad Customers |
Identify Bad Customers who are always in Debt and abuse the available Services.
using either SQL or Business Intelligence Products like
Cognos. |
| 7 |
Customer Propensity |
Identify Customers with a Propensity to demonstrate
certain kinds of behaviour.
For example, to Commit Fraud, Get into Debt or Need Help.
This requires sophisticated statistical software like SAS or SPSS. |
| 8 |
Data Mining |
More complex analysis leading to more valuable insights.
Following a Methodology like
CRISP, which is a Cross-Industry Standard Process for Data Mining.
and using Products like
SAS Enterprise Miner,
or SPSS Clementine
or Tanagara.
Dr. Tim Hoban, of Serco, has provided the following note :-
Tanagra is from the University of Lyon in France. It is available as a free download.
It is not designed for commercial use but can be used on real data to show the power of data mining.
It uses the same algorithms as the commercial products, but the interface is different.
|
| - |
Books |
Data Mining Techniques |
| - |
Books |
Data Preparation for Data Mining |
| - |
Consultants (UK) |
Dunn Humby - consultants in Customer Segmentation & Loyalty Retention. |
| - |
Consultants (USA) |
Data Miner Inc.. |
