Explore insights from Boston Institute of Analytics students on bank marketing analysis. Delve into predictive models and data-driven strategies for enhanced marketing effectiveness. Discover more at https://bostoninstituteofanalytics.org/data-science-and-artificial-intelligence/
3. Marketing Introduction
The process by which companies create value for customers and build strong customer relationships in order
to capture value from customers in return.
Bank marketing plays a crucial role in the financial sector by facilitating the promotion and sale of
banking products and services to existing and potential customers.
1.Customer Acquisition and Retention
2.Brand Awareness
3.Product Promotion and Cross-Selling
4.Market Expansion
5.Relationship Building
4. Term Deposit
A sum of money is kept for a fixed maturity and the depositor is not allowed to withdraw this
sum till the end of the maturity period. This is called as Term Deposits because they are kept
up to a particular term.
Key Takeaways
•A term deposit is a type of deposit account held at a financial institution where money is locked
up for some set period of time.
•Term deposits are usually short-term deposits with maturities ranging from one month to a few
years.
•Typically, term deposits offer higher interest rates than traditional liquid savings accounts,
whereby customers can withdraw their money at any time.
5. Data Description
This is a classic bank marketing dataset originally uploaded by UCI Machine learning repository. The dataset gives
your information about a marketing campaign of a financial institution in which we will have to analyse in order to
find future strategies in order to improve future marketing campaigns for bank.
>Features
1.age
2.job
3.marital status
4.education
5.default
6.balance
7.housing
8.loan
9.contact
10.day
11.month
12.duration
13.campaign
14 pdays
15.previous
16.poutcome
Data collection
Secondary Data Source:
https://www.kaggle.com/janiobachmann/bank-marketing-dataset
>Label
1.Deposit
6. Research Methodology
•Quantitative research involves gathering numerical data and analyzing it statistically to draw conclusions.
•Qualitative research focuses on understanding phenomena through in-depth exploration of perspectives and
experiences.
•Mixed-methods research combines both quantitative and qualitative approaches for a comprehensive
understanding.
•Approach: The chosen research design based on the research objectives, nature of the research questions, and
feasibility consideration.
Example: if the study aims to quantify the effectiveness of bank marketing campaigns, a quantitative approach may
be more suitable.
7. Exploratory Data Analysis
*There is no unwanted column present in given dataset to remove.
*No missing values found and no feature with only one value
*There are 9 categorical features
*Client with job type as management records are high in given dataset and housemaid are very less
*Client who married are high in records and divorced are less
*Client whose education background is secondary are in high numbers.
*Default feature does not play important role as it has value of ‘no’ at high ratio to value ‘yes’ which can be dropped
*Data in month of may is high and less in December.
*There are 7 numerical features
*There are no discrete variable counts in this dataset
*There are 7 continuous numerical features
*Client shows interest on deposit who had discussion for longer duration
*It seems some outliers found (age, balance, duration, campaign, pdays and previous has some outliers)
10. Presentation of Numerical Features
>it seems age, days
distributed normally
>balance, duration,
campaign, pdays and
previous heavily skewed
towards left and seems to
be have some outliers.
11. Model Selection
• Applying cross validation on Random forest and XGBoost to see which one gives us the
best score.
•K-Fold technique where dataset is sliced into 5 slices and gives us 5 scores.
•Bases on the mean value choose the best score.
•Fit the GridSearchCV objects to the training data to find the best hyperparameters for
each model.
•We print the best parameters and scores for each model.
•As per cross validation XGB classifier gives us the best score.
•Poutcom_success is the most important feature that would affect the bank decision as of
this dataset.
12. Conclusions
These models play a crucial role in targeting specific customer segments with personalized marketing
strategies, leading to improved customer satisfaction and loyalty.
Emerging technologies such as AI, blockchain, and big data analytics presents significant opportunities for
banks to enhance their marketing strategies and deliver more personalized and engaging experiences to
customers
The research is evaluated on different marketing channels employed by banks, including traditional
advertising, digital marketing, social media, and email campaigns. While each channel has its advantages
and challenges, banks can optimize their marketing strategies by leveraging a combination of channels to
reach and engage customers effectively.
By optimizing their marketing channels and strategies based on the findings of the research, banks can
improve the effectiveness of their marketing efforts. Targeted marketing campaigns tailored to specific
customer segments are more likely to retain customers and drive desired outcomes such as increased
sales and customer retention.
By staying attuned to customer preferences and adapting their marketing strategies accordingly, banks
can position themselves as trusted financial partners and attract and retain loyal customers.