AUTOMATED CUSTOMER SEGMENTATION AI-POWERED LEAD SCORING FOR EDTECH

EdTech companies collect vast amounts of data, such as browsing behavior, email engagement, and other contact details, which can be leveraged through predictive analytics to estimate a lead’s purchase probability. This study investigates the use of machine learning for prospect scoring using a dataset of approximately 9,000 educational lead records. The objective is to enhance lead conversion rates by predicting the likelihood of conversion using historical behavioral data and engagement metrics. The problem is approached as a binary classification task, where supervised learning algorithms such as logistic regression, decision tree, and ensemble methods like random forest are applied. Purchase timestamps are used to define activity windows for converted leads, ensuring fair data representation. The models are evaluated using accuracy, precision, recall, and ROC-AUC. Among them, logistic regression achieved the highest accuracy and interpretability, while random forest provided valuable insights through feature importance analysis. The results demonstrate that machine learning-driven lead scoring can effectively prioritize high-potential leads, optimize marketing and sales strategies, and offer actionable business insights through visual analytics for decision makers.