Lifi

LI_BFSI_01 - Life Insurance Sales
Capstone Project – Note 1
SUBMITTED BY
Gana Devraj
0
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
1) Introduction of the business problem 2
a) Defining problem statement

b) Need of the study/project
c) Understanding business/social opportunity
2) Data Report 2
a) Understanding how data was collected in terms of time, frequency and

methodology
b) Visual inspection of data (rows, columns, descriptive details)
c) Understanding of attributes (variable info, renaming if required)
3) Exploratory Data Analysis 5
d) Univariate analysis (distribution and spread for every continuous

attribute, distribution of data in categories for categorical ones) Visual
inspection of data (rows, columns, descriptive details)
e) Bivariate analysis (relationship between different variables, correlations)
f) Removal of unwanted variables (if applicable)
g) Missing Value treatment (if applicable)
h) Outlier treatment (if required)
i) Variable transformation (if applicable)
j) Addition of new variables (if required)
4) Business Insights from EDA 13
k) Is the data unbalanced? If so, what can be done? Please explain in the
context of the business
l) Any business insights using clustering (if applicable)
m) Any other business insights
1
Proprietary content. ©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited
PROBLEM STATEMENT:
a) Defining problem statement
Life insurance companies are faced by several problems such as low penetration rate, consumer
behaviour, harsh economic environment and financial illiteracy among others. The company agents,
whose efforts contributes to a larger of the companies' income also face difficulties such as instances of
rejection during their constant hustling, and networking. The excellence of agent performance has
remained a major problem which needs to be addressed.
b) Need of the study/project
This project will help the company to design appropriate engagement activity for their agents basing on
their performance.
c) Understanding business/social opportunity
Life insurance companies offers various opportunities to individuals, institutions and the economy. They
protect societies' wealth, maintain standards of living, provide a sense of security and peace of mind,
encourage loss mitigation, increase prosperity, and provide awareness of the reality of risks and their
consequences through civic education.
DATA REPORT:
 CustID - Unique customer ID

 AgentBonus - Bonus amount given to each agent in last month
 Age - Age of customer
 CustTenure - Tenure of customer in organization
 Channel - Channel through which acquisition of customer is done
 Occupation - Occupation of customer
 EducationField - Field of education of customer
 Gender - Gender of customer
 ExistingProdType Existing product type of customer
 Designation - Designation of customer in their organization
 NumberOfPolicy - Total number of existing policy of a customer
 MaritalStatus - Marital status of customer
 MonthlyIncome - Gross monthly income of customer
 Complaint - Indicator of complaint registered in last one month by customer
 ExistingPolicyTenure - Max tenure in all existing policies of customer
 SumAssured - Max of sum assured in all existing policies of customer
 Zone - Customer belongs to which zone in India. Like East, West, North and South
 PaymentMethod - Frequency of payment selected by customer like Monthly, quarterly,
half yearly and yearly
 LastMonthCalls - Total calls attempted by company to a customer for cross sell

 CustCareScore – Customer satisfaction score given by Customer in previous service call
Import Libraries
Load data
Inspect Rows
Data contains 4520 rows and 20 columns.
Data Information
Dataset has 20 variables
There are 5 integer data types, 7 float and 8 object types.
EXPLORATORY DATA ANALYSIS:
Missing values
Check missing values
'isnull()' method detects missing values. It returns a Boolean same-sized object indicating if the values
are NA. NA values, gets mapped to True values. Everything else gets mapped to False values.
'sum()' method followed by 'isnull()' adds up all the True instances per column to give a summary of
missing values per column.
We find no missing values in our new data frame
Outliers
Some columns contain outliers. However, we will not drop them since they may be correctly observed
values
Univariate analysis
Continuous variables
- Summary statistics
- The maximum agent bonus is 9608.00000 and the minimum is 1605.00000 and median bonus is
3911.50000
- The average gross monthly income of customer is 22890.309991
- Minimum age of customer is 2 and maximum is 8
- Minimum existing policies of a customer is 3.565363 and maximum is 6.000000
Categorical variables
1. Channel - Channel through which acquisition of customer is done
- Agent is the most used channel and online is the least
2. Occupation
- Most customers are salaried, few are freelancers
3. Education Field
- Most of the customers are graduates and few have MBA

4. Gender
Males are dominating
5. Designation
- Most people who subscribe life insurance policy are managers and the least are exe.
Bivariate analysis
1. Agent bonus vs occupation
- Agents receive more bonus from subscriptions of large businesses
2. Agent bonus vs Education field

There is slight variation of the bonus received by the agent from different education fields
3. Agent bonus vs zone
Most bonus come from people located North, West and East of India
4. Agent bonus, designation and payment method
Most bonus income are from Vp who pay monthly and yearly, and AVP who pay quarterly
Business insights from EDA
1. Is the data unbalanced? If so, what can be done? Please explain in the context of the business
- The distribution of class variables in our dataset is comparable hence balanced
2. Any business insights using clustering
There is a fair distribution of bonus basing on the occupation and customer care score
3. Any other business insights
- Life insurance company should train their agents to target the following persons:
 The large businesspersons

 The small businesspersons
 Salaried persons, especially managers
 VPs whose subscriptions are monthly and yearly
 AVPs who subscribe quarterly

Lifi

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lifi

Uploaded by

Copyright:

Available Formats

LI_BFSI_01 - Life Insurance Sales

Capstone Project – Note 1

a) Defining problem statement

a) Understanding how data was collected in terms of time, frequency and

3) Exploratory Data Analysis 5

d) Univariate analysis (distribution and spread for every continuous

4) Business Insights from EDA 13

a) Defining problem statement

b) Need of the study/project

c) Understanding business/social opportunity

 CustID - Unique customer ID

 LastMonthCalls - Total calls attempted by company to a customer for cross sell

Data contains 4520 rows and 20 columns.

EXPLORATORY DATA ANALYSIS:

Check missing values

1. Channel - Channel through which acquisition of customer is done

- Agent is the most used channel and online is the least

- Most customers are salaried, few are freelancers

- Most of the customers are graduates and few have MBA

Males are dominating

1. Agent bonus vs occupation

- Agents receive more bonus from subscriptions of large businesses

2. Agent bonus vs Education field

3. Agent bonus vs zone

2. Any business insights using clustering

3. Any other business insights

 The large businesspersons

You might also like