|
 |  |
 |
 Overview Success Stories
|  |
Chordiant Decision Management—Maximizing Response Rates A large Dutch bank wanted to improve its process of selecting prospects for a mailing campaign. Chordiant Decision Management was called to the job. The ChallengeThe bank wished to maximize the response rate for a given select rate. The data available contained 300 attributes for each customer. Chordiant Decision Management was used to analyse the data in order to filter out the attributes most effective in predicting whether they were inclined to respond. The Solution Chordiant Decision Management used the most predictive customer attributes to develop a so-called spectrum scoring model, assigning scores to individual customers. A clear insight was gained into the consequences of different selection strategies. How It Was Done The model was developed on the results of a pilot mailing of 9,000 cases and validated using the results of a prior operational mailing of 77,721 cases. Of the 9000 pilot cases, 860 responded, a response rate of 9.56%. Of the 77721 customers addressed by the validation mailing, 6097 had responded, accounting for 7.84%. During the project, several models were developed. These models were analyzed using the Chordiant Decision Management's model analysis module to select the best applicable model. The selected model scored the cases and grouped the scores into categories, each with its own response rate, given a certain select rate. The Outcome The Chordiant Decision Management model offered a clear insight into the response rates for different select rates of mailing. For example, only 52% of the customers would have to be addressed in order to reach 75% of the responders. At a select rate of 76% over 90% of the responders would be reached. Figure 1 depicts the cumulative response rate over the different categories. 
The ResultDuring the project, several models were Chordiant Decision Management succeeded in adding value to the process of selecting prospects for a mailing. More responders could be obtained out of a smaller selection group, and thus at lower costs. |  | |
 |
| © 2000-2008, Chordiant Software, Inc. All rights reserved. Legal/Disclaimer | Privacy Policy | Site Map | |  |
| Solutions | Retail Banking | Card Services | Lending | Insurance | Telecommunications | News & Events | About Chordiant | |
