DATA ANALYTICS

CAPSTONE
PROJECT

DONE BY: JOEL LIM

 TABLE OF CONTENTS

PROBLEM STATEMENT
An overview of the
01 02 EXPLORATORY DATA ANALYSIS
Explore various means of data
business problem. visualization.

DATA INSIGHTS
Detailed data analysis for
03 04 RECOMMENDATIONS
Provide some recommendations
business insights. for management’s action plan.
PROBLEM STATEMENT
A business manager of a consumer credit card portfolio
is facing the problem of customer attrition. He would
like to analyze the data to find out the reasons behind
the high attrition rate of the credit cards as well as the
low credit card transaction values. He would also like
to leverage the same on how to retain his customers
who are likely to drop off.
 10,127 rows

Dataset Description 19 columns

0 missing value
S/No. Column Name Description
1 CLIENTNUM Client number. Unique identifier for the customer holding the account.

2 Attrition_Flag Internal event (customer activity) variable - if the account is closed then 1 else 0.

3 Customer_Age Demographic variable - Customer's Age in Years.

4 Gender Demographic variable - M=Male, F=Female.

5 Dependent_count Demographic variable - Number of dependents.

6 Education_Level Demographic variable - Educational qualification of the account holder (example:

high school, college graduate, etc.)
7 Marital_Status Demographic variable - Married, Single, Divorced, Unknown.

8 Income_Category Demographic variable - Annual Income Category of the account holder (< $40K,
$40K - 60K, $60K - $80K, $80K-$120K, $120K+)
9 Card_Category Product Variable - Type of Card (Blue, Silver, Gold, Platinum).
 10,127 rows

Dataset Description 19 columns

0 missing value
S/No. Column Name Description
10 Months_on_book Period of relationship with bank.

11 Total_Relationship_Count Total number of products held by the customer.

12 Months_Inactive_12_mon Number of months inactive in the last 12 months.

13 Contacts_Count_12_mon Number of contacts in the last 12 months.

14 Credit_Limit Credit Limit on the Credit Card.

15 Total_Revolving_Bal Total Revolving Balance on the Credit Card.

16 Avg_Open_To_Buy Open to buy Credit Line (Average of last 12 months).

17 Total_Trans_Amt Total Transaction Amount (Last 12 months).

18 Total_Trans_Ct Total Transaction Count (Last 12 months).

19 Avg_Utilization_Ratio Average Card Utilization Ratio.

EXPLORATORY DATA ANALYSIS

There are 8,500 existing

customers which accounts for
83.9% while 16.1% of them have
closed their accounts.
EXPLORATORY DATA ANALYSIS

In terms of gender distribution, there are more female customers. 43.7% of the existing
customers are female while 40.2% of them are male. The remaining customers have closed
their accounts.
EXPLORATORY DATA ANALYSIS

In terms of marital status distribution, there are more married customers. 42.4% of the
existing customers are married while 34.9% of them are single and 6.69% are divorced. The
remaining customers have closed their accounts.
EXPLORATORY DATA ANALYSIS
3,128 of the customers
are Graduates followed
by 2,013 with High
There are 9,436 Blue Card
customers compared to 555 Silver
School. Card, 116 Gold Card and 20
Platinum Card customers.

Most customers have dependent count of 2 to 3.

Majority of the customers
(3,561) earn less than $40K
while 1,790 of them earn
between $40K to $60K.
EXPLORATORY DATA ANALYSIS
Most customers have 3 products with the bank. Most customers are 2 to 3 months inactive in the
last 12 months.

Most customers have 2 to 3 contacts in the last Majority of the customers have 36 months of
12 months. relationship with the bank.
EXPLORATORY DATA ANALYSIS
The mean age of our customers is 46.3 years old. Our customers have a total mean transaction
count of 65.

Our customers have a total mean

transaction amount of $4.4K.
EXPLORATORY DATA ANALYSIS
Our customers have a mean credit limit
of $8,632.
Our customers have a mean total
revolving balance of $1,163.

The mean average open to buy credit The mean average card utilization ratio
line is $7,469. is 0.275.
EXPLORATORY DATA ANALYSIS
Most attrited customers
Most attrited customers have 2 to 3 dependents.
aged 43 to 50 yrs old.

Most attrited customers

Most attrited customers are Married and Single.
are Graduates.
EXPLORATORY DATA ANALYSIS
Most attrited customers Most attrited customers
have income level of less are Blue card and
than $40K. Platinum card holders.

Most attrited customers

have 36 months on book. Most attrited customers
have 2 to 3 products with
the bank.
EXPLORATORY DATA ANALYSIS
Most attrited customers
Most attrited customers
have 2 to 3 contacts with
have 2 to 3 months
the bank for the last 12
inactive with the bank.
months.

Most attrited customers Most attrited customers

have low credit limit of have total revolving
less than $5K on the credit balance of less than $500
card. on the credit card.
EXPLORATORY DATA ANALYSIS
Most attrited customers
Most attrited customers
have approximately 35 to
have average open of less
55 transactions.
than $2K to buy credit
line.

Most attrited customers Most attrited customers

have total transaction have average utilization
amount of $1.5K to $3K. ratio of less than 0.1.
DATA INSIGHTS
There are 3,403 clients who earn less than $40K under
the Blue card category followed by 1,675 clients in the
next income category. Management shall be concerned
about this group of customers which represents about
33.5% of the clientele base.
 DATA INSIGHTS

6% of the attrition rate comes from the customers in the lowest income category. Even
though the attrition rate is consistent with the income level of the customers, Management
can look into the various product and marketing strategies for each income level to better
serve their customers as well as to reduce the attrition rate of the clientele base.
DATA INSIGHTS
Avg utilization ratio is observed to be high for the existing
customers age group 65 to 67 years old.

Transaction amount is observed to be high for the existing

customers age group 45 to 50 years old.
DATA INSIGHTS

Avg utilization ratio seems to be the lowest for customers

who has 2 products with the bank, more than 4 contacts and
0 month inactive for the past 12 months .
DATA INSIGHTS

Transaction amount is observed to be high for the

Graduates while avg utilization ratio is high for the
Post Graduates and Graduates since they belong to
the mass affluent group of customers.
DATA INSIGHTS
The average Credit Limit is shown to be the highest
for the males especially for the highest income
category. The female customers seem to enjoy a
much lower credit limit compared to their male
counterparts. If they can have a higher credit limit,
they should be able to transact more with higher
amounts. Management can also review the
Unknown income category female group of
customers which may impact their credit limit.
 RECOMMENDATIONS
1. To focus on the high attrition rates of credit cards for the following characteristics of
customers:
❑ Blue Card holders
❑ Age Group of 43 to 50 years old
❑ Marital Status (Married and Single)
❑ Education Level (Graduates and High School)
❑ Income Level of less than $40K
❑ Number of dependents of 2 to 3
❑ 36 months of relationship with the bank
❑ Average of 3 products held by the customer
❑ Around 2 to 3 months inactive with the bank for the last 12 months
❑ Around 2 to 3 contacts for the last 12 months
❑ Low Credit Limit of approximately less than $5K
❑ Total Revolving Balance of approximately less than $500 on the credit card
❑ Average Open of approximately less than $2K to buy Credit Line
❑ Approximate of 35 to 55 Transactions and Transaction Amounts of $1.5K to $3K
❑ Average Utilization Ratio of less than 0.1
 RECOMMENDATIONS
2. To review the current product offerings to make them more attractive in order to increase
the retention ratio of the existing customers and reduce the attrition rates of the outgoing
customers.
3. To review and enhance their promotional marketing strategies for certain groups of
customers especially the Blue Card and Platinum Card holders (affluent customers) which
have the highest attrition rates.
4. To improve on the overall card benefits for the mass affluent card holders to better meet
their expectations so that this group of customers would think twice before closing their
accounts with the bank.
5. To find out the reasons as to why there are some “Unknown” entries under the ‘Income’
Category which will make data analytics more challenging since such dataset may be
filtered out for analysis.
6. To review the existing credit limit for the females which are much lower compared to their
male counterparts.
7. To conduct more market research on the more affluent customers to improve on the
mean total transaction count as well as the mean total transaction amount.
THANK YOU!