As digital technologies have become ubiquitous thanks to the Internet, new modes of feedback in L2 writing have emerged, yet what remains unclear is how feedback given through alternative modes helps improve writing quality and how new feedback tools fit in the overall context of writing instruction. Therefore, the purpose of this embedded mixed-methods study is to assess how three online feedback modes help improve student writing. Thirty-three intermediate Turkish-L1 learners of English received written, audio and screencast feedback in Google Drive to improve their writing in a multi-draft essay-writing task and an essay-revision task with three parallel essays. The results indicated that it was the audio group that made the highest number of correct revisions in the essay-writing task, while there was not a significant difference among the three feedback modes in the essay-revision task. Semi-structured interviews and screen recordings provided qualitative data about their preferences and how they worked with each mode to address both microlevel and macrolevel problems. The participants did not uniformly prefer a particular feedback mode but highlighted the potential benefits and downsides of each mode.