Enterprise Level Systems
The likelihood to respond / relative future value scoring method described in the
Drilling Down book is based on one of the most fundamental and predictive customer behavioral scoring
algorithms in database marketing. In fact, most other predictive modeling
techniques will find the
core RFM variables as highly predictive of
future customer value. For example, both a top-down
regression model and a bottoms-up data-mining training session will find Recency
as very highly predictive of customer response to promotional efforts.
For this reason, more advanced users can take the
Drilling Down methods and models explained in the book and use them to mimic the
functionality of enterprise class systems that analyze customer behavior and
customize the website experience
based on this behavior. All you have to do is create the loyalty
scores in your customer database and use a simple
Perl script to modify the website presentation based on loyalty score.
For example, you could create an anti-defection campaign where best customers with
falling loyalty scores could be automatically presented with a discount or gift.
If you are unfamiliar with these systems and how
they are used, you might visit the vendor sites for more detailed information.
Below is a list of resources, not in any particular order, where you can learn
more about the use of predictive customer modeling in enterprise class systems.
"How To" Article:
Measuring the ROI of CRM
Overview of Issues and Software
Broadbase (now a part of Kana)
NetGenesis (now a part of SPSS)