3 Steps for Effective Segmentation Using Data Mining for CRM in Banking

The banking industry is going through a rough patch across Europe and US. The effects of this are visible in the emerging economies as well. In this challenging time banks need smarter strategies and systems to survive and grow. Also with increasing competition, especially in the emerging markets, CRM systems can prove to be a great competitive advantage.

Customer segmentation is the first step towards building an effective business development strategy. Data mining and analytics integrated with Banking CRM system plays an important role in this. Customer base comprises of variety of customers with distinct needs depending on demographic, economic and behavioural conditions. Banks need to identify common characteristics and form distinct groups to handle their differentiated needs. In order to provide personalised products, rewards and incentives a refined approach is needed that would segregate the customers in terms of life-stage, psychographics and profitability.

Data mining techniques can be used by applying filters on various fields captured during transactions, account opening and running statistical models on the data collected. This will need to collate and analyse data from the core banking system and various supporting systems including CRM systems, credit card systems, etc.

The major objectives of segmentation:
  • Customised product offerings 
  • Customised and priority services 
  • Improve relationship with profitable customers and cut resources spent on loss making customers 
  • Better offering to new customers based on the intelligence gained from the existing customer segment they belong to 
  • New product development and bundling as per the customer segment profile
Steps for successful customer segmentation:
1. Segmentation based on contribution to profit - The objective of segmentation should be to retain them as they have a major contribution to the bank’s profit. Their attrition will drastically decrease the bank’s profit. This is done based on the marginal revenue contribution. Customers can be segmented in 3-5 categories. Mostly pareto’s principle will hold true, i.e. 80% of the revenue will come from 20% of the customers. This can be labelled as Tier 1, Tier 2, Tier 3, Tier 4, Tier 5 and so on.  
2. Further segmentation based on the customer assets with the bank and frequency of transactions - This is done based on the average periodical balances maintained with the bank depending on the product holding. Customers with high balances and high loans can be categorised accordingly. Depending on the frequency of transactions and balances inactive customers are identified and targeted to increase their activity.
3. Further dividing the above segments based on product ownership, channel utilisations and type of transactions - This is more refined segmentation to design customised marketing techniques. Depending on the product ownership i.e. monthly average.
  • Balances with respect to product 
  • Number of transactions
  • Value of the transactions 
  • Alternative channel usage
The types of transactions and products include-
  • Deposit transactions 
  • Withdrawal transactions 
  • Lending v/s Asset Balances 
  • Type of loans 
  • Type of investments
The above can be further divided depending on the life stage of the customer.

This will result in multi-dimensional segmentation as below.

Tiers
Other Factors
Profitability
Age
(Years)
Lending v/s Asset Balances
(%)
Alternative channel usage
(%)
Number of transactions
(Monthly Avg.)
Value of Transactions
(Monthly Avg.)
Tier 1





Tier 2





Tier 3





Tier 4





Tier 5






The Marketing Process for the above segments
The above segmentation gives information on behavioural aspects it does not give information on the qualitative aspects like wallet share, satisfaction level, attrition risk etc. This can only be gained through marketing research. This data will give a very intelligent insight on the strategy to target these segments. For instance customers with high deposit balances and higher age should be offered fixed income product, while those with healthy savings account deposit and age group between 25-40 should be offered mutual funds, equity and insurance products. Customers should be managed throughout their life cycle and helped to graduate to a healthy financial portfolio and in the process making them more profitable for the bank. A large part of young customers might have a new and single product relationship with the bank. They should be encouraged with loyalty programs and credit card or loans depending on their income levels.

Segmentation with the help of data mining from various existing systems is a very important exercise and a must for effective business development. Making this intelligence available to the customer facing teams and marketing team in the Banking CRM system can prove to be a great tool to increase cross selling and up selling capability. A 360 degree view of the customer with effective segmentation can be a important competitive advantage for the banks.



1 comment:

  1. Thanks for the blog! Data mining is the most effective tool especially for bank or loan sector. Once again thanks for sharing this blog!

    ReplyDelete