Students work independently. They read and annotate texts, then complete digital writing activities, building text-dependent responses that demonstrate specific knowledge. They iterate on their writing based on multiple rounds of AI feedback—working through up to five revision attempts per prompt to build precision, evidence use, and clarity. In Reading for Evidence specifically, students use "because, but, so" sentence stems to cite, restate, and build claims rather than answer multiple-choice questions.

The AI is trained to look specifically at the knowledge and ideas in a student's response, rather than simply giving feedback on writing as a skill. It provides instant, sentence-level feedback that pushes students back to the text, rephrases evidence in their own words, and corrects grammatical errors—using Socratic-style coaching designed not to reveal the answer. The AI also provides a bank of exemplary responses for comparison. Quill Diagnostic identifies sentence-level skill needs across grammar, syntax, and sentence construction, then routes students to targeted practice activities matched to those needs.

The teacher assigns Quill activities and uses the dashboard, which surfaces completion, skill mastery, and growth across the class, to see where students are struggling. From there, teachers can read texts in advance to provide targeted guidance, circulate to give individual support during writing, and lead whole-class discussions to synthesize learning after the activity.

Leaders make adoption decisions, set the scope of Quill use across schools, and evaluate alignment with the core curriculum. Once adopted, they can use Quill's administrator dashboard to monitor teacher and student usage (sentences written, learning hours, completion); premium-tier leaders can also access school- and school system-level reporting, teacher PD on writing pedagogy, and tech integrations to make Quill accessible system-wide.