Based on preliminary observations made, it was found that the ability to write poetry of class VIII students of SMP Negeri 13 Pekanbaru was still lacking. The lack of students' ability to write poetry is caused by three factors, namely factors from the teacher, students, and the environment. The factor of the teacher is the lack of use of techniques and media that vary in learning to write. Factors of students, namely students find it difficult to express ideas that are in their minds, difficulty finding the right words to be arranged into poetry, and the difficulty to imagine. The factors of the environment are the lack of facilities and infrastructure owned by the school. Based on the explanation above, this study examines two problems, namely (1) how is the improvement in poetry writing ability of eighth grade students of SMP Negeri 13 Pekanbaru after learning through clustering techniques with photojournalism media, and (2) how is the behavior of class VIII students of SMP Negeri 13 Pekanbaru in learning poetry writing uses clustering techniques with journalistic photo media. This study uses a classroom action research design conducted in two stages, namely cycle I, and cycle II with research subjects of class VIII Pekanbaru State Junior High School 13. Data collection is done by using test and non-test techniques. Non-data was obtained from observations, interviews, student diaries, teacher journals, and photo documentation. Data analysis includes quantitative and qualitative data. From the results of the first cycle, and the second cycle that students have experienced an increase in each aspect of writing poetry, this is evident from the acquisition of each aspect of writing poetry such as the suitability of the title with the content in the first cycle. increased by 6% to 82.7. For aspects of the choice of words or diction, the increase in the first cycle obtained an average value of 76.0 and the second cycle became 80.5 with an increase of 4.5%. Figurative language aspects of the average value of students in the first cycle 63.2, then in the second cycle increased by 17.3% with an average value of 80.5. For the versification aspect the acquisition of the average value in the first cycle was 60.2 and in the second cycle it was 80.5 with an increase of 20.3%. The latter for typographic aspects of the acquisition of the average value of the first cycle of 72.7 and in the second cycle became 76.7 so as to obtain an increase of 04%.