E-learning was a vital tool for education during the global lockdown, as it was the only alternative for billions of students worldwide to continue learning amid the COVID-19 crisis. However, despite the importance and relevance of E-learning, there is a significant lack of studies that examine E-learning success under heterogeneity effects, such as gender, cross-cultural, or level of education. A clear and comprehensive framework that encompasses all necessary e-learning services is crucial to ensure their effective implementation, use, and impact on student learning and achievement. For this purpose, we used a new version of the original 3S-T model that measured the E-learning success based on self-student assessment through a survey study with 136 students from the Universitat Politècnica de València (Spain) and 81 from the Central University of Ecuador (Ecuador). This model that expands the existing theoretical framework of models related to E-Learning (Technology Acceptance Model (TAM), Extensions of TAM model, and Information Systems Success Model (ISSM) identifies a broad range of success predictors that help to achieve success, such as learning and academic outcomes. The success predictors include factors related to the surrounding conditions, the system features, the tutor's development, the student's performance, accessibility, interactivity, content quality, assessment methods, learning outcomes, and student engagement. This conference paper uses a multigroup Analysis (MGA) using Partial Least Squares (PLS-SEM) to evaluate moderation across both countries over the depurate 3S-T model, including the assessment measurement invariance of composite models (MICOM) procedure. [For the full proceedings, see ED665357.]