This qualitative study investigates how writing instructors compose feedback in multimodal digital environments, focusing on the rhetorical and relational dimensions of their design choices. Drawing on social semiotics and multimodal composition theory, the study analyzes feedback artifacts, instructor interviews, and student surveys from six first-year writing courses. Findings reveal that instructors engage in complex feedback design work across communication modes, often without formal training or shared frameworks. Instructors tended to default to text-based habits shaped by genre memory but adapted their strategies in response to communicative breakdowns and student needs. The study identifies three core themes: reliance on print-era conventions, rhetorical problem-solving through modal layering, and ambiguity in feedback interpretation. Despite these challenges, instructors demonstrated creativity and care in their attempts to communicate clearly and relationally. The article calls for a rhetorical framework to support multimodal feedback design, emphasizing the need for pedagogical reflection, professional development, and student co-interpretation. As genAI and platform automation continue to evolve, the findings underscore the importance of feedback as a site of human judgment and presence. The article concludes with recommendations for instructors, writing programs, and institutions to better support feedback as intentional, relational work.

Writing instructors spend considerable time responding to student writing with the expectation that students will use that feedback to improve their writing. However, a number of studies have questioned the extent to which students apply instructor feedback to improve their writing or transfer it to new writing situations. Timing of feedback and students’ interest in feedback are frequently discussed in the literature on response as two factors that impact students’ ability to apply and transfer response. In this article I consider the relationship between the two factors and whether students’ behavior as they access feedback is related to when in the writing process feedback is provided. I report the results of a study using site statistics collected by a learning management system that compares students’ rates of opening instructor feedback on preliminary drafts and final papers. I also examine whether students’ rates of accessing feedback on preliminary drafts changed over the course of the semester from the first assignment to the final assignment. This study illustrates that the timing of instructor feedback significantly impacts students’ behavior as they access feedback and suggests that instructors prioritize feedback on preliminary drafts to encourage students to apply and transfer feedback.