|
|
|
|
|
Identifying a critical mass of customers having similar behaviour is the first step in any analytical frame work. Segmentation of the customer base using multiple variables – demographic, psychographic and behavioural – is a standard analytical approach used in financial services industry to identify the customers with similar behaviour. Clustering using statistical techniques will pre-process large data sets related to customers and identify the segments with high similarity of the customer attributes. Once the segments are built from customer data, they can be analyzed for opportunities and risks. Unlike other tools which do not allow the user to interactively build the segments our segmentation module provides two options:
Automatic segmentation of the customers based on statistical algorithms
Interactive segmentation using domain expert’s judgment and statistical algorithms
|
| |
Clustering tree - It shows how the customers are segmented based upon the risk factor. The High rick customers are further segmented into High risk Group 1 and high risk group 2.
Once the best possible grouping of the customers into segments having similar behaviour within the group and distinct behaviour between the groups is done, the segments can be analyzed and profiled. | 
Click to enlarge
|
| |
|
Rule based analysis of segments
This module will analyze the segments for risks and opportunities within the segments. The analysis will be based on the user supplied analysis rules. There are rules to define opportunity and risk. Based upon this rules, segments are profiled to identify them. |
Click to enlarge
|
|
| |
| Solution Methodology |
| |
The Solution methodology is based on industry standard CRISP-DM methodology.
The methodology consists of five stages: |
| |
1. Data Preparation – Collect data from various sources and make it compatible for modelling
2. Building Analytical Models – Using the data, we select relevant variables that make sense from business
perspective and build models
3. Testing – The models are tested for accuracy and ability to meet business objectives.
4. Implementations of the solutions –The solutions are deployed at the business environment.
5. Training – Business users are given training to use the solution. |
| |
| Benefits |
| |
Seamless Data Integration: ABIBA’s systems ETL tool automates the data integration processes.
Interactive Solution:
Better understanding of Customers:
Identification of Opportunities and Risk:
Predictive capabilities:
Supports Enhanced Analysis: |
|
|
| |
 |
|
 |
Customer churn is one of the most pressing issues for telecommunications service providers. |
| |
read more... |
|
 |
|
|
|
|
| |
 |
|
 |
Response modeling is a generic analytical technique to support a number of business... |
| |
read more... |
|
 |
|
|
|
| |
 |
|
 |
The telecommunications industry loses in excess of billions annually in the form of bad debt. |
| |
read more... |
|
 |
|
|
|