With the advancement of information technology, Digital Game-Based Learning (DGBL) has emerged as an effective approach to enhancing students’ motivation and academic performance. Previous studies have shown that DGBL increases learner engagement and comprehension, particularly in programming education, where it helps reduce anxiety related to abstract concepts. However, empirical research on the effects of different learning sequences within DGBL remains limited. This study involved 66 students from a Digital Media Design department, who participated in a 17-week instructional experiment. Participants were divided into two groups: the “ExplorationFirst” group (Group A, n = 26) and the “Theory-First” group (Group B, n = 40). The course included pre- and post-tests, a learning satisfaction questionnaire, and qualitative interviews. The findings revealed: 1. Different learning sequences in DGBL have a measurable impact on students’ programming learning outcomes. 2. ANCOVA results showed a medium effect size favoring the exploration-first group in post-test scores, suggesting a potential advantage in practical application. 3. In terms of learning satisfaction, the exploration-first group scored significantly higher in the “output quality” dimension and reported greater overall satisfaction. 4. Qualitative findings indicated that the exploration-first sequence was more effective in enhancing learning motivation and engagement, while the theory-first sequence supported deeper conceptual understanding for some students. These results highlight the importance of appropriate sequencing in DGBL to improve learning experiences and outcomes.