(Image Source: MSN Bankbazaar.com)
Segmentation is the basis for any marketing campaign. Traditionally is it defined as dividing the market according to certain characteristics like Geographic regions or Psychographics, but customer segmentation should be based on benefit.
Currently clustering is a key tool that being but to do segmentation based on benefits. Customer segmentation with clustering tools has its own drawbacks. One major weakness of clustering is that’s it is un-supervised in nature. That is to say, customers are segmented using information which does not include customer response informant ion (or performance data on business aspects). In other words, it is not geared for best results.
Rather it relies on luck! Unless you find useful patterns in clusters.
A better customer segmentation method is Hotspot Profiling Analysis. Hotspot searches segments with highest (or lowest) responses (or performances). In principle, it does the similar job as clustering. But it is focused to maximize response (or performance) indicators and directed to search such segments. In general, you just get better results using clustering than random techniques. With hotspot analysis, on the other hand, you may be able to achieve best results. Generally, you can expect better results than clustering.
Hotspot profiling analysis drills-down data systematically and detects important relationships, co-factors, interactions, dependencies and associations amongst many variables and values accurately using Artificial Intelligence techniques such as incremental learning, and generate profiles of most interesting segments. This can be applied customer data to find profiles of most (or least) responsive segments that can be used to develop marketing plans.
Health Insurance Customer Segmentation
In insurance industry, profiling is very important in determining premium rates. Typically, insurers collect information available. However, analyzing thoroughly is not feasible since the number of variables is normally large. The following two examples demonstrate how hotspot profiling analysis can be used in profiling risky insurance policies out of dozens of customer variables.
An insurance company keeps health insurance records in its database: gender, age, education, smoking, drinking, sun activity, height, weight (=obesity level), claim payment, etc., as well as contact information. The company wishes to know which health insurance groups are at the highest risk, i.e., have the highest claim ratio. The following is a possible output of hotspot analysis;
In this case we can see the probability of this individual claiming is much higher given the set of characteristics that determine his risk exposure.
This methodology would go a long way in the Health insurer’s new policy of concentrating on the average premium per member rather than the number of members added.