Profiling is a description of a customer or set of customers that includes demographic, geographic, and psychographic characteristics.
Customer profile is needed when you are trying to market your product.
Government insurance and pension scheme as PMSBY, PMJBY and APY are new security based bank products introduced to market.
Profiling banking customers - Insurance and Pension Products
1. Project Report On Profiling Banking Customers For
Government’s Insurance And Pension Plans Using
Analytics
By,
Suryakumar T
14MBA1073
VIT University
2. INTRODUCTION
• Profiling is a description of a customer or set of customers that includes demographic,
geographic, and psychographic characteristics.
• Customer profile is needed when you are trying to market your product.
• Government insurance and pension scheme as PMSBY, PMJBY and APY are new security
based bank products introduced to market.
• Customer profiling helps to target customers for the above banking security products.
Customer
Profiling
Portrait of
Customers
Group of
similar
goals
customers
Decision
Making
3. ABOUT THE INDUSTRY
• Banking industry is a financial services company or financial intermediary that creates credit by
investing, storing and lending money to a borrower.
• Basic products of Banking Industry:
Consumer & Corporate Banking
Pension and Insurance
Mortgage Loans and Investment Banking
Private Equity
Savings and Credit Cards
• From these banking products profiling of the customers is made for Insurance and Pension
products in our study.
• Here from the banking industry, Indian Bank is considered for the profiling study.
4. OBJECTIVE
• To analyze the data obtained from the bank customers to find relation/association and classification.
• To profile the group of customers who are potential customers of the government introduced insurance and
pension plans.
Background:
• The government introduced insurance schemes are PMSBY and PMJBY, while pension scheme is APY.
• The main motive for the introduction of these schemes are for providing social security for all common
people (BOP).
PMSBY
• Accidental Insurance
• Coverage: 2 Lakhs
• Premium: 12 Per annum
• Age: 18 to 70
PMJBY
• Life Insurance
• Coverage: 2 Lakhs
• Premium: 330 per annum
• Age: 18 to 50
APY
• Pension scheme
• 1000 to 5000 per month
• Age: 18 to 40
5. REVIEW OF LITERATURE
• ROL is made to gain knowledge about the existing study.
• Findings:
Decrease in the rural banking still, more untapped market potential among rural millions (Rural
insurance plan )
▫ “The decline is due to the reason that the banks are trying to increase the credit of their finance system”
Reaching BOP through Unit linked insurance plan.
▫ “Gradually, the banking sector has realized the importance of serving the “bottom of the pyramid.”
Micro Finance through Self Helping Groups(SHG) promise to deliver poverty alleviation and
results in Financial Inclusion.
India’s life insurance market is an emerging market as the ratio of life insurance premium to GDP
is only around 4 per cent and penetration level is about 26 per cent.
Direct relationship between Life Insurance and economic growth
• Ideas Formed:
Need of insurance plans – BOP upliftment (85% of population) & rural upliftment and Economic
growth
Impact of SHG on microfinance
6. RESEARCH METHODS
• Instrument Design: Questionnaire form is designed to measure the variables and characteristics of
interest.
• Sample size: 125 sample data is collected for the study of project.
• Sampling Technique: Non-Probability sampling technique
• Period of study: 7 weeks (11th May 2015 to 27th June 2015)
• Research design:
Descriptive analysis
Inferential analysis
Predictive analysis.
• Statistical tools applied:
Chi square test – To find the association between two variables
Cluster analysis – For segmenting and classifying the
Regression
• Software tools used: MS Excel, SPSS 21.0 and Tableau 9.0
7. ANALYSIS AND INTERPRETATION
The analytical report of the data is made to profile the customers for insurance and pension schemes.
Hypothesis:
Ho – There is no significant relation between two variables.
H1 – There is significant relation between two variables
Variables Considered:
Dependent variable : ‘Applying for Insurance’ and ‘Applying for Pension’
Independent variable:
INSURANCE PENSION
Gender Gender
Salary earned per month Salary earned per month
Knowledge about the schemes Knowledge about the schemes
Age Group Age Group
Organization Employed Organization Employed
Perception of Cost of living Perception of Cost of living
Reason for selection Reason for selection
Expected Pension
8. INSURANCE SCHEME –Result and Findings
• From the descriptive statistics of the insurance scheme, we are able to find out that 53.6% of people are
applying. This shows that nearly half of the customers are willing to apply for the insurance schemes.
• From the predictive analysis (Logistic regression) we infer the below details,
▫ Fitness of the model: From the Hosmer and Lemeshow Test, the significance value, ‘F’ is found. Significance
value, F = 0.834, hence it is a non significant chi square value and the model is found to be a “Fit model”
▫ Determination of variables: Cox and Snell and Nagelkerke R2 values are found. Nagelkerke R2 value = 0.536,
Hence the 53.6% of application status is determined by variables “Age”, “Income”, “Scheme knowledge”
and “Reason for selection”. Also the magnitude is found using Wald variable.
• From the Inferential analysis(Chi-square test) we infer that below segments apply for insurance using
significance test ,
Variables Segments
Age 31 – 50
Income Nil and less than 10000
Perception Costly
Knowledge Fully and partially
Working Sectors Unorganized and
Unemployed
Reason for selection Low cost and social
9. INSURANCE SCHEME –Results and Findings
• From the Segmentation and classification analysis (Cluster analysis), using significance test; we infer
that below segments apply for insurance,
• Overall from the above analysis we are able to profile the customers of insurance as below,
• The above listed customers are the potential and valued customers.
Variables Cluster1 Cluster2
Age 31 – 40 51 – 60
Income Nil 10000 – 20000
Perception Costly Costly
Knowledge Fully Partially
Variables Target Segments
Age Older population and middle aged
Income Fiscally fit and challenged
Perception Costly
Knowledge Fully and partially
Working Sectors Unorganized and Unemployed
Reason for selection Low cost and social
10. INSURANCE SCHEME –Dashboard (Tableau 9.0)
Maximum
Application
Maximum
Application
Maximum
Application
Maximum
Application
11. PENSION SCHEME –Results and Findings
• From the descriptive statistics of the insurance scheme, we are able to find out that only 10.4% of
people are applying. This seems to be a lower pension applying propensity.
• From the predictive analysis (Logistic regression) we infer the below details,
▫ Fitness of the model: From the Hosmer and Lemeshow Test, the significance value, ‘F’ is found. Significance
value, F = 0.996, hence it is a non significant chi square value and the model is found to be a “Fit model”
▫ Determination of variables: Cox and Snell and Nagelkerke R2 values are found. Nagelkerke R2 value =
0.421, Hence the 42.1% of application status is determined by variables “Age”, “Gender”, “Scheme
knowledge” and “Reason for selection”. Also the magnitude is found using Wald variable.
• From the Inferential analysis(Chi-square test) we infer that below segments apply for insurance using
significance test,
Variables Segments
Age 18 - 40
Expected Pension 5000 and above 5000
Knowledge Fully
12. PENSION SCHEME –Results and Findings
• From the Segmentation and classification analysis (Cluster analysis), using significance test; we infer
that below segments apply for insurance,
• Overall from the above analysis we are able to profile the customers of insurance as below,
• The above listed customers are the potential and valued customers.
Variables Cluster1
Age 18 - 30
Income Less than 10000
Perception Easy
Knowledge Partial
Variables Target Segments
Age Young adults
Income Fiscally fit
Perception Easy
Knowledge Fully and partially
Expected Pension 5000 and above 5000
13. PENSION SCHEME –Dashboard (Tableau 9.0)
Maximum
Application
Maximum
Application
Maximum
Application
Maximum
Application
Maximum
Application
Maximum
Application
Maximum
Application
14. CONCLUSION
• In this study we have finally profiled the customers both for the Insurance and Pension plans. Those
customers are,
• Marketing services is provided to the bounded customer:
Media channel like posters and newspaper ads for low income group peoples.
Placing contact centers like camps in rural areas
Reaching potential customers like SHG – Magalir kulzhu
• Further exploration:
When a new insurance or pension schemes are introduced with different eligibility criteria, the same model
can be used for the analysis.
Since it is being a fit model over the variables the same can be used for the association, relation and
segmentation.
Variables Target Segments for Insurance Target Segments for Pension
Age Older population and middle aged Young adults
Income Fiscally fit and challenged Fiscally fit
Perception Costly Easy
Knowledge Fully and partially Fully and partially
Working Sectors Unorganized and Unemployed -
Reason for selection Low cost and social -
Expected Pension 5000 and above 5000
Hinweis der Redaktion
Descriptive stats: Knowledge and low cost
Inferential statistics makes inferences about populations using data drawn from the population.
Significance level – 10%
Descriptive stats: Knowledge and low cost
Inferential statistics makes inferences about populations using data drawn from the population.
Significance level – 10%