The main thrust of this study was to determine whether a genre-based instruction improve the writing proficiency of Iranian EFL learners. To this end, 30 homogenous Iranian BA learners studying English at Islamic Azad University, Bandar Abbas Branch were selected as the participants of the study through a version of TOEFL test as the proficiency test. The selected participants were 15 females and 15 males who were randomly divided into two groups of experimental and control. The both experimental and control groups were asked to write on a topic determined by the researcher which were considered as the pre-test. The writing of the students were scored using holistic scoring procedure. The subjects received sixteen hours instruction--the experimental group using a genre-based pedagogy and the control group through the traditional methodology which was followed by a post-test--the subjects were, this time, asked to write on the same topic which they were asked to write before instruction. Their post-writings were also scored through the holistic scoring procedures. In analyzing the data, t-test statistic was utilized for comparing the performances of the two groups. It was found that there is statistically significant difference between the writing ability of the participants who go under a genre-based instruction and who don't. The study, however, didn't find any significant role for gender.