Abstract: In today’s data-driven financial environment, organizations increasingly rely on analytical tools to make informed decisions under uncertainty. This study examines the relevance and application of decision tree models in financial management, focusing on their capacity to enhance accuracy, interpretability, and efficiency in decision-making. Decision trees provide a transparent and systematic framework for evaluating multiple financial alternatives, quantifying risks, and predicting outcomes. Through a desk review approach, the study synthesizes existing literature and empirical findings on the use of decision trees in budgeting, capital budgeting, portfolio management, fraud detection, and credit decision-making. It further traces the evolution of decision tree algorithms from traditional models such as ID3, C4.5, and CART to modern ensemble techniques like Random Forests and XGBoost, emphasizing improvements in explainability through SHAP and LIME frameworks. Findings show that decision trees offer robustness, flexibility, and high interpretability, making them valuable for regulatory compliance and ethical financial practices. However, challenges such as overfitting, instability, and limited data generalizability persist. The study recommends that financial institutions integrate ensemble-based decision tree methods and invest in high-quality data management systems to improve model performance and reliability. Overall, decision tree models remain indispensable tools for achieving transparency, accountability, and strategic decision-making in modern financial management.

Keywords: Decision Tree Model, Financial Management, Credit Scoring, Fraud Detection, Machine Learning