One problem that teachers generally face is to choose prompts for EFL academic writing assessments or classroom practices. Studies show that the performance of students in writing is influenced by the choice of the prompts. This study investigated whether a prompt can cause a difference in overall quality and cohesion when it attempts to activate appropriate background knowledge of students, to provide students with enough guidance to the topic and also tries to make students think more on the topic.. f he main purpose of this study was to investigate the relationships among types of writing prompts, cohesion and overall quality of EFL students’ writings, in particular, the study aimed at investigating the possible effect of two different prompts, traditional and thought provoking, on the overall quality of a written product and the frequency of cohesive devices used in that written product. The traditional prompt is a prompt which broadly states the topic with a vague focus. The thought provoking prompt, on the other hand, is a prompt which attempts to make students think more on the topic without losing the focus of it. The secondary purpose was to see whether there was a relationship between the number of the cohesive devices used and the overall quality of the written product. Data were collected through two classroom applications at the Foreign Languages Education Department at Middle East Technical University. Thirty-eight advanced level students were asked to write about two different prompts in a timed writing test. The writing prompts described the same topic but differed in length, information load and wording. Data were analyzed by employing descriptive and inferential statistics. The descriptive statistics included the calculation of means and frequencies. The inferential statistics was conducted to investigate the effect of the two prompts on the overall quality and cohesion of the essays. The inferential statistics was also used to investigate the relationship between cohesion and the overall writing quality of the essays. The results of the study showed that a prompt which tries to activate appropriate background knowledge, to provide enough guidance to the topic and to provoke thinking about the topic resulted in a better performance in overall quality. The results also indicated that the difference between the prompts caused no difference in the use of cohesive devices, and cohesion did not contribute to overall quality. In the light of the findings, the research suggests that the lexical formation and the importance of the information about the topic presented through a prompt should be taken into account in EFL writing and testing.