Small and medium enterprises (SMEs) increasingly rely on digital marketing to enhance brand visibility, drive engagement, and boost conversion rates. However, given the diversity of marketing channels and the variability in audience behavior, predicting campaign success remains a challenging task for these businesses. This study aims to address this problem by developing a machine learning model to predict the success of digital marketing campaigns based on historical data. The dataset used in this research comprises 10,000 campaigns, with features such as ad spend, engagement metrics, conversion rates, and audience reach. After preprocessing the data to handle categorical variables and standardizing numerical features, a Random Forest Classifier was trained on the dataset. The model was evaluated using accuracy, log loss, and ROC-AUC metrics, focusing on differentiating between successful and unsuccessful campaigns. Results showed that the model achieved an accuracy of 90% in predicting successful campaigns but struggled with accurately identifying failures, as evidenced by a lower ROC-AUC of 0.48. Key predictors of success were identified, including engagement metrics, conversion rates, ad spend, and audience reach. This study provides valuable insights for SMEs by highlighting which digital marketing factors contribute most to campaign success, offering a data-driven foundation for resource allocation and strategy development. However, the class imbalance in the dataset limited the model’s ability to predict failure cases accurately, suggesting the need for advanced techniques such as synthetic data generation to improve future models. These findings have significant implications for SMEs aiming to optimize their digital marketing strategies by focusing on the most impactful variables.