A complete end-to-end Bank Analytics project analyzing loan data and credit/debit transactions. Includes dashboards in Excel, Power BI, Tableau, SQL-based analysis, and a detailed presentation. Datasets are confidential and not uploaded.
This Bank Analytics Project provides a comprehensive analysis of customer financial behavior by integrating two major datasets: a Banking Loan Dataset and a Credit/Debit Transaction Dataset. The project aims to uncover patterns that influence loan approval, customer spending habits, transaction behavior, and financial risk levels. Through detailed dashboards created across Excel, Power BI, Tableau, and SQL-based data exploration, the project delivers powerful insights into customer segmentation, loan performance metrics, and transaction trends. The analysis helps identify high-risk customers, seasonal transaction spikes, spending categories, and overall loan portfolio health, supporting data-driven decision-making for financial institutions.
This project utilizes a multi-tool analytics approach to ensure a complete and professional analysis. Power BI and Tableau were used to build interactive visual dashboards and uncover deeper insights through advanced charts and drill-downs. Excel supported pivot-based dashboards and summary analytics for quick reporting. SQL was used to explore, clean, filter, and analyze the datasets through efficient queries. Together, these tools provide a full end-to-end data analytics workflow, starting from raw dataset exploration to interactive dashboard creation and insights presentation.
The project is based on two structured datasets. The Banking Loan Dataset contains information such as customer income, age, employment type, loan amount, credit score, marital status, and final loan approval status. The Credit/Debit Transaction Dataset includes detailed transaction logs such as customer ID, transaction amount, transaction type (credit/debit), timestamp, merchant category, and monthly spending behavior. Due to confidentiality guidelines provided by my mentor, the original datasets cannot be uploaded to GitHub; however, the project documentation, dashboards, and analysis results have been included in the repository for review. Combining both datasets provides a richer understanding of customer financial behavior, expenditure patterns, and their impact on loan performance.
The analysis reveals several meaningful insights. Customers with stronger credit scores and stable income patterns demonstrated higher loan approval rates. Spending behavior from the transaction dataset showed that debit transactions peak during weekends, while credit card usage increases during festival months. High-risk customers were identified through a combination of low credit scores, inconsistent spending trends, and previous loan defaults. The dashboards also highlight top spending categories, seasonal transaction variations, loan repayment behavior, and demographic patterns influencing financial decisions. These insights can help banks optimize loan policies, improve customer targeting, and enhance financial risk management.
Future improvements to this project include integrating machine learning models to predict loan approval or customer risk scores based on historical behavior. Additional enhancements may include building an automated ETL pipeline, adding real-time transaction monitoring, expanding the dataset with customer feedback or demographic details, and embedding the dashboards into a web application. Incorporating Python-based predictive analytics and developing a unified financial customer scoring system will further strengthen the project's analytical capabilities and practical use for financial institutions.