The main purpose of the research was to investigate the effectiveness of genre-based approaches in teaching academic writing. The study was motivated by the researcher's perceptions about university students' difficulty in acquiring the essayist literacy of the academy, and the fact that very little empirical research had been conducted on the effect of genre-based writing interventions. The following questions guided the research: (1) Can genre-based approaches be justified theoretically? (2) How effective are genre-based academic writing interventions? (3) Which is more effective: a narrow-angled or a wide-angled approach? The theoretical framework combines foundational principles of Systemic Functional Grammar, Constructivism and Critical Literacies. A mixed methods design was used, including a survey of writing tasks, genre analysis, discourse analysis, and a quasi-experimental comparison of pre- and posttest essay ratings. The survey of writing tasks indicated that the academic essay was the written genre most frequently required by humanities departments, and that argumentation, discussion, explanation, description and analysis were the text types featuring most prominently in writing prompts. Since the materials of the Department of Historical and Heritage Studies contained the largest number of essay-length tasks, the subject-specific intervention was focused on students of history. The cross-disciplinary group included students with Economics, English, History, Philosophy, Political Sciences, Psychology and Sociology as majors. A genre-based presyllabus, comprising exploration, explicit instruction, joint construction, independent construction and critical reflection, was customized for the subject-specific and cross-disciplinary groups. The syllabus gave prominence to the use of rhetorical modes, logical development of an argument, and engagement with other authors. The statistical analyses of the essay scores show that the narrow-angled and the wide-angled genre-based interventions were effective. Although the size of the improvement on the four dimensions of the scoring instrument was not equal, the overall improvement of the students in each of the groups is statistically significant. Despite the more modest overall improvement of the students in the cross-disciplinary group, their mastery of stance and engagement exceeded that of their subject-specific counterparts. Even though both interventions were effective the subject-specific group performed significantly better than the cross-disciplinary group overall (p = 0.043). Their performance was also more consistent across the four dimensions of the scoring instrument. The results of the opinion survey indicate that students from both groups were generally positive about the effect of the respective interventions on their academic writing abilities. The only significant difference is the subject-specific group's more positive evaluation of the transferability of the skills they acquired. The more pronounced skills transfer was probably facilitated by the subject-specific group's deeper level of engagement with source materials and more opportunities for practising content-based writing. Main limitations of the study include the small sample size and non-parallel presentation of the two interventions.