The increasing adoption of asynchronous learning formats in higher education presents new challenges for monitoring student engagement and performance, particularly in mathematics courses that require consistent practice and timely feedback. While Moodle, a widely used open-source learning management system (LMS), offers a wealth of data on student activity, its potential for learning analytics (LA) remains underutilized in practice. This paper explores the foundations of a learning analytics framework tailored to an asynchronous undergraduate mathematics course designed for engineering students. Using Moodle-generated activity logs, quiz data, and session patterns, the study examines how student engagement behaviors correlate with academic performance. The research identifies typical user profiles, such as consistent performers and deadline-driven crammers, and highlights signs of superficial engagement—such as repeated quiz attempts without prior content access. The findings offer actionable insights for improving course design, fostering deeper engagement, and enabling early interventions. This study contributes a practical, evidence-informed approach to integrating learning analytics within Moodle and emphasizes the value of aligning analytics frameworks with pedagogical goals in asynchronous STEM education.