The purpose of the paper was to examine factors that predict students' continued usage intention of web-based learning management systems (LMS) in Tanzania, with a specific focus at the School of Business of Mzumbe University (MU). Specifically, the study investigated major predictors of actual usage and continued usage intentions of e-learning system, and challenges of using the e-learning system. The case study research methodology was used in this study. This study was conducted at the School of Business of Mzumbe University because most of the faculty members use the LMS to deliver e-courses, assignments, and announcements. Data was collected through questionnaires and interviews. Questionnaires were physically distributed to 300 third year undergraduate students, with a rate of return of 77%. A total of 20 faculty members were also interviewed in order to supplement and corroborate results from questionnaire survey. Exploratory Factor Analysis (EFA) was used to assess the validity and reliability of the measurement model; while multiple regressions were used to examine the research hypotheses by using SPSS. The research model for this study was formulated based on UTAUT, and it is comprised the following constructs: effort expectancy, self-efficacy, performance expectancy, social influence, facilitating conditions, actual use, and continued usage intentions.The results show that actual usage was determined by self-efficacy, while continued usage intentions of web-based learning system was predicted by performance expectancy, effort expectancy, social influence, self-efficacy and actual usage. Challenges for using web-based LMS were related to ICT infrastructure barrier, LMS user interface was not user friendly, weak ICT policies, management and technical support, limited skills, lack of awareness, resistance to change, and lack of time to prepare e-content and use the e-learning system.The study findings are useful to e-learning managers and university management to identify important factors and develop appropriate policies and strategies to encourage long-term usage of e-learning systems for future studies and lifelong learning.By using UTAUT in the context of continued usage intentions and the integration of an additional construct (“self-efficacy”), the extended UTAUT model fits very well in the webbased learning systems in Tanzania, in particular where such studies are scant. The findings can be used in other institutions with similar conditions in investigating the continued usage intentions of e-learning systems.