Purpose: This study examined the integration of task and game-based learning into an online Test of English as a Foreign Language (TOEFL) preparatory course taught at two Indonesian higher education institutions (HEI) over a period of 20 online class meetings totaling 40 class hours. Methodology: Using a single-case experimental design, it engaged undergraduate students enrolled in a TOEFL preparatory course in March 2021 at Politeknik Perkeretaapian Indonesia Madiun (N = 48) and Universitas Muhadi Setiabudi, Indonesia (N = 48). The study used multiple t-tests and one-way ANOVA with "Cohen's d" to determine the degree of impact on students' total outcome on pretest and post-test, as well as their scores on particular skills. Findings: Multiple paired-sample t-tests revealed significant improvement in students' overall scores (t (95) = 15.35, p < 0.001) and in specific-skills scores for Listening Comprehension (t (95) = 10.32, p < 0.001), Structure and Written Expression (t (95) = 5.90, p < 0.001), and Reading Comprehension (t (95) = 5.63, p < 0.001). Independent t-tests and one-way ANOVA revealed significant variations in students' overall and individual skills ratings in study programs. However, the percentage of students who achieved the required TOEFL scores for admission to both institutions was significantly different. Significance: The findings have major implications for both the lecturers at the two institutions and the designers of TOEFL preparatory programs. This study sheds light on an online TOEFL preparatory course that teaches students through task- and game-based learning. To run this program successfully, teachers' creativity in generating materials (tasks) and incorporating digital technologies is essential.