Professional blog

Problem Statement

AssignmentTutorOnline

Introduction

The bank marketing case study will help you understand how to apply the concepts of EDA and Optimisation in real-life business scenarios. In this case study, you will need to analyse the dataset of a marketing campaign conducted by ‘Bank of Corporate’ during 2017. You are expected to derive the essential insights using EDA and identify the optimised solution for a marketing campaign in the future.

Business Understanding

You work as a Business Analytics Consultant at the Bank of Corporate. The bank provides financial services/products such as savings accounts, current accounts, debit cards, etc. to its customers. In order to increase its overall revenue, the bank conducts various marketing campaigns for its financial products such as credit cards, term deposits, loans, etc. These campaigns are intended for the bank’s existing customers. However, the marketing campaigns need to be cost-efficient so that the bank not only increases their overall revenues but also the total profit. You, being a consultant to the bank, need to apply your knowledge of EDA and optimisation on the existing data to analyse the patterns and provide the best-optimised solution for the future marketing campaign.

In 2017, the bank conducted a telemarketing campaign for one of its financial products ‘Term Deposits’ to help foster long-term relationships with the existing customers. The dataset (provided below) contains the information about all the customers who were contacted during this year to open term deposit accounts.

What are term deposits?

Term deposits also called fixed deposits, are the cash investments made for a specific time period ranging from 1 month to 5 years for predetermined fixed interest rates. The fixed interest rates offered for term deposits are higher than the regular interest rates for savings accounts. The customers receive the total amount (investment plus the interest) at the end of the maturity period. Also, the money can only be withdrawn at the end of the maturity period. Withdrawing money before that will result in an added penalty associated, and the customer will not receive any interest returns. This kind of investment deposits provides the required funds to the bank for lending money to the corporates or individuals at a higher interest rate than what is paid to the customer.

Business Objectives

Your aim is to identify the target customers for the term deposits from the pool of the bank’s existing customers. You should also capture the key driving factors (or driver variables) behind the successful conversion of a customer, i.e., the customer opening a term deposit account with the bank. Using this information, the bank would optimize investment in its future marketing campaign.

EDA

In this case study, you will use EDA to understand how customer attributes and campaign attributes influence the successful opening of a term deposit account with the bank.

• Customer attributes include all the information pertaining to the customers, such as bank account number, age, job, marital status, education, the cash balance in the account, information about any loan default, availing housing loan or personal loan, etc.
• Campaign attributes include all the information related to the calls/reach-outs made by the bank to the customers during the marketing campaign. The following attribute information is provided in the dataset given below:
  • The information of when the customer was last approached by the bank regarding any one of the product/services such as credit cards, loans, etc.
  • The total number of times that the customer was contacted by the bank for any of the products/services till date
  • The outcome of the last reach-out with the customer for any one of the products/services
  • The call duration with the specific customer for opening a term deposit account
  • The outcome of the call made for opening a term deposit account with the specific customer
• The customer attributes and campaign attributes are also mentioned in the Data Dictionary given below.

Optimisation

For its future marketing campaign, the bank has allocated a budget of ₹1,50,000 and has also decided to segment the customers based on their marital status and educational background. Also, notably, the cost incurred by the bank for a one-minute call to any customer is ₹10. Considering all these factors, you, as a consultant, need to provide the analysis to the bank regarding the number of calls to be made to each customer segment(‘Customer segment’ is explained below) such that the total number of customers opening the term deposit account is maximised.

Customer Segment:

In the dataset, the customers are segregated based on their Marital status and educational background. ‘Marital status’ and ‘educational background’ has three categories as follows.

Each combination of marital status and educational background is considered as a customer segment.
Examples: ‘Single – Bachelors’ is considered as one segment. ‘Single – Masters’ is considered as one segment. Similarly, ‘Married – Masters’ is considered as one segment, ‘Married – Doctorates’ is considered as one segment and so on.

The Conditions of the Bank:

The bank is concerned about the overall customer diversification. It wants to ensure that it reaches out to all the customer segments. For this, it has provided you with the following information to include in your analysis:

• From each customer segment(‘customer segment’ as explained above), at least 50 customers need to be contacted.
• The total number of calls made to each customer category should meet the minimum number of calls as mentioned in the following table:

• The total number of conversions of the following customer categories should match a minimum number as mentioned below:

Main Objective of Optimisation Problem

Within the conditions given above, you need to estimate the number of calls needs to be made for each customer segment such that the total estimated no of converted calls for the future marketing campaign is maximised.

Instructions

You need to obtain the following details from the EDA analysis performed on the 2017 marketing campaign data which help you to calculate the estimated cost of the future marketing campaign for each customer segment.

1. For each customer segment, calculate the total average call duration for converted calls based on the data for 2017.
2. For each customer segment, calculate the total average call duration for unconverted calls based on the data for 2017.
3. For each customer segment, calculate the conversion rate based on the data for 2017 (conversion rate = Number of converted calls / total number of calls made). 
4. Based on the estimated no of calls for each customer segment and the conversion rate that is calculated in the above step, you can estimate the ‘no. of converted calls’ and ‘no. of non converted calls’ for the future marketing campaign.

Using the total average call duration for both converted and unconverted customer calls(can be calculated using points 1 and 2 of instructions), and the estimated ‘no. of converted calls’ and ‘no. of non converted calls’ for the future marketing campaign(can be calculated using point 4 of instructions), calculate the estimated cost for each customer segment.

Data Understanding

Download the dataset given below. It contains complete information about all the customer calls made as part of the marketing campaign for term deposits during 2017.

Data Set

file_downloadDownload

You can access the data dictionary which provides the meaning of the variables of the data set from the link provided below:

Data Dictionary

file_downloadDownload

Results Expected

• Provide a step-by-step approach and details of the analysis performed to derive the insights in the form of a PowerPoint presentation.
• Identify the missing data, duplicate data, outliers, spelling inconsistencies, and use the appropriate methods to deal with these issues.
• Identify the possible derived columns and, if required, convert any specific column to a required format for ease of analysis.
• Explain the results of univariate, segmented univariate, bivariate analysis, etc. in business terms. In addition, identify the insights that will help explain how a specific variable is playing a key role in converting the customers to open the term deposit account with the bank.
• Include visualisations and also summarise the most important results in the presentation. You are free to choose the graphs that explain the numerical/categorical variables.
• Derive the required values based on the EDA of the dataset and find out the optimized solution for the future marketing campaign.

You can use the following editable template to present your analysis, the major business insights of EDA, and also the results of Optimisation.

PPT Template

file_downloadDownload

You are expected to submit your analysis in a zip file with four different Excel workbooks along with the presentation. The four different excel workbooks can be classified and named as follows:

1. cleaned_dataset: This workbook should consist the dataset after you have made all the required modifications to it and it is ready for analysis.
2. univariate_analysis: This workbook should include the results of univariate and segmented univariate analysis of the dataset.
3. bivariate_analysis: This workbook should consist of the results of the bivariate analysis of the dataset.
4. optimization: This workbook should include the results of the optimized solution.
5. bank_marketing_presentation: This is a PDF (converted from the edited template PPT) where you will provide the details of the approach and major insights of the analysis.