|
 |  |
 |
 Overview Success Stories
|  |
Chordiant Decision Management—Optimizing Sales Channels In order to reduce costs, a large international publishing company used Chordiant Decision Management to optimize its sales channels The ChallengeOur customer is in business-to-consumer advertising. In the existing situation, prospects were segmented by ‘slicing-and-dicing’ prospect information and historical sales efforts and results. These segments were then used for planning the prospects and targeting the sales force, but although they seemed sensible, the resulting sales showed that improvement in revenues could be achieved by better targeting. To interact with its prospects, our customer primarily uses representatives, outbound / inbound call center, direct mailing and public advertising. These not only differ in costs, but also in hit rate (the percentage of sells that were made), capacity, and ‘intimacy’ with the potential customer. Furthermore, the prospects themselves also differ considerably from each other. Chordiant Decision Management was given the task of finding the optimal channel / prospect combinations. The Solution Instead of segmenting the prospects by their characteristics, Chordiant Decision Management was used to develop a scoring model. Based on historical sales results, a model was produced to generate a score for each specific prospect. Considerable cost reduction appeared to be achievable by selecting the highest scoring prospects. How It Was Done Similar to segmentation, the characteristics of the prospects were used as input to the model, but the combination of characteristics is used more intelligently. For a specific region, a model was developed and its resulting sales were analysed. Higher score indicated a higher propensity to buy. 
The OutcomeA preliminary study showed that 61% of the revenues were generated by 20% of the prospects. Figure 1 shows the cumulative numbers for volume (percentage of prospects), revenues and sales (conversion rate), ordered by score. The Result The sales channels are now arranged by assigning a specific channel to a specific score range. Models for other regions are being developed and implemented. Also, models for predicting the revenues per prospect are being developed. These will be combined with the propensity-models to provide segmentation on both propensity and value. |  | |
 |
| © 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 | |
