The use of authentic materials for language teaching is common. Yet, there is not much information about what type of instructional materials benefits learners. The purpose of the study is to find empirical evidence of whether non-fiction authentic materials are more superior that fiction authentic material. The researchers used an experimental design to answer the research question. The population of the study was all students who took an essay writing course at University PGRI Wiranegara in the academic year 2019-2020. The sample was taken by using a systematic random sampling method which results in two equal numbers of students in two experimental groups. Group A was taught with non-fiction authentic material, while group B was taught with the fiction authentic material. The researchers used ANOVA to analyze the data in SPSS 22. The process of teaching and learning followed the writing as a process approach and was done in a blended-learning setting because of the pandemic of COVID 19. The result shows that there is no significant interaction between content type and proficiency level (F(2,30) = 1.347, p = 0.275). The main effect of content type on writing performance was not significant (F(1,30)=0.001, p 0.973), but the main effect of proficiency level on writing performance is significant such that the students who have a higher level of writing proficiency have better performance than the students who have a lower level of writing proficiency (F (2,30) = 5.653, p 0.008). The researchers conclude that both types of authentic materials are equally effective to improve the students' essay writing performance.