This report presents a comprehensive data science analysis of a Portuguese bank's marketing campaign. The primary objective is to predict which existing customers are most likely to subscribe to a term deposit, thereby optimizing marketing efforts. The analysis utilizes a dataset derived from phone call-based marketing campaigns, encompassing 41,188 rows and 21 attributes. The methodology includes data representation, cleaning, and exploratory visualization using ggplot2. Five machine learning models (Naïve Bayes, Random Forest, SVM, C5.0 and Logistic Regression) are implemented for classification, with the Naïve Bayes model achieving the highest accuracy. The report provides detailed visualizations of key attributes, such as education, marital status, and campaign outcomes. The conclusion highlights the significance of various attributes in building predictive models and recommends the Naïve Bayes model for identifying potential customers based on their demographic and campaign interaction data, enabling the bank to make cost-effective marketing decisions. The report also includes R code used for the analysis.