JeddAI generates feedback by reading your success criteria, rubric and comment banks, then drafting comments that map each observation back to a specific criterion. You review, edit or accept every draft, so the feedback reflects your standards rather than a generic model's judgement.
What does JeddAI read before it writes any feedback?
Before drafting a single comment, JeddAI reads the marking materials you already use: your success criteria, your rubric or marking guide, any comment banks you have built, the task instructions, and the student’s submitted work. It treats these as the source of truth rather than relying on a generic idea of what good work looks like.
This matters because the same piece of writing can meet very different expectations depending on year level, subject and task. A Year 9 persuasive essay and a senior analytical response share surface features but are judged against separate criteria. By grounding itself in your materials first, JeddAI aims to comment on what you actually asked students to demonstrate.
You stay in control of what goes in. If your success criteria are vague, the draft feedback tends to be vague too, so the clearer and more specific your criteria, the more targeted the drafts become.
How does JeddAI map a comment to a specific success criterion?
JeddAI works criterion by criterion, checking the student’s work against each success criterion and drafting a comment that names the criterion it relates to. Instead of one overall impression, you get feedback organised around the things you told students would count.
For each criterion, a draft typically notes where the student has met the expectation, where they have fallen short, and what a sensible next step might be. Because the comment is anchored to a criterion, it is easier for you to check whether you agree, and easier for the student to see which part of the task the feedback is about.
This structure also surfaces gaps. If a criterion has no supporting evidence in the submission, that absence shows up in the draft rather than being quietly overlooked, which is useful when you are marking a large stack under time pressure.
Where do your rubric and comment banks fit in?
Your rubric and comment banks shape both the standard JeddAI applies and the wording it uses. The rubric supplies the performance descriptors that separate one level from the next, while your comment banks give it your phrasing and recurring teaching points, so drafts sound closer to how you would put it.
This is what keeps the feedback yours rather than generic. Two teachers marking the same essay against the same curriculum can still prioritise different things, and your banks and rubric encode those priorities. JeddAI leans on them so the drafts reflect your emphasis, not a default template.
It also means the tool improves as your materials do. A well-built comment bank that covers common misconceptions, sentence-level fixes and next-step prompts gives JeddAI more to draw on, so the drafts arrive closer to finished. If you have refined a rubric over several years of moderation, that work carries straight into the feedback students receive.
- Success criteria: the specific, task-level expectations students were given.
- Rubric or marking guide: the performance levels and descriptors that define quality.
- Comment banks: your saved phrasing and the feedback points you use most often.
- Task instructions: what the question or assignment actually asked students to do.
What does the teacher's review step look like?
Every draft comment is a starting point you review, not a final grade JeddAI issues. You read each suggestion, keep what is accurate, rewrite anything that misses the mark, and delete what does not apply before the feedback reaches a student.
This review step is deliberate. AI can misread a subtle argument, miss context you hold in your head about a particular student, or apply a criterion too generously. Keeping the teacher as the final decision-maker means those errors get caught, and it keeps the professional judgement where it belongs.
Over time the review also gets faster. As your criteria and comment banks get sharper, fewer drafts need heavy editing, and you spend more of your marking time on the responses that genuinely need your attention.
How does criterion-aligned feedback stay consistent across a class?
Applying the same success criteria to every submission is where consistency comes from, and it is exactly the part humans find hardest at the end of a long marking session. JeddAI applies your criteria the same way to the first paper and the fortieth, so a student is not marked more harshly simply because of where they landed in the pile.
Consistency does not mean identical feedback. Because the drafts are tied to individual evidence in each submission, two students who both miss the same criterion can still receive different, specific next steps. What stays constant is the standard being applied, not the wording.
For teachers who moderate or team-teach, criterion-aligned feedback is also easier to compare. When comments are labelled against shared criteria, a colleague can see the basis for a judgement instead of guessing at it.
How do you get started with success-criteria-aligned feedback?
Start by making your success criteria explicit and specific, because everything downstream depends on them. Well-written criteria describe observable qualities of the work, not just ‘analyse the text’ but what a strong analysis actually contains, and they give both the student and the tool something concrete to check against.
It is worth sharing those criteria with students before they submit, too. When learners know exactly what they are being marked against, the feedback they receive afterwards lands as a next step rather than a surprise, and the aligned comments JeddAI drafts connect directly to language they have already seen.
From there, a tool like JeddAI can draft feedback aligned to those criteria and your rubric, apply them consistently across the class, and hand you drafts to review and edit so the marking still sounds like you and stays under your control. The time saving comes from not writing every comment from scratch, not from handing over judgement. You can Get started with JeddAI to see how aligned feedback fits your own criteria and marking routine.
Frequently asked questions
Does JeddAI decide the final grade?
No. JeddAI drafts feedback and suggested marking against your criteria, but you review, edit and confirm everything, so the final judgement stays with the teacher.
What happens if my success criteria are unclear?
Vague criteria produce vague drafts. The more specific and observable your criteria, the more targeted and useful the feedback JeddAI can align to them.
Can JeddAI use my existing rubric and comment banks?
Yes. It draws on your rubric, success criteria and saved comment banks so the drafts reflect your standards and phrasing rather than a generic template.
Will every student receive the same feedback?
No. Feedback is tied to the evidence in each submission, so students who miss the same criterion can still receive different, specific next steps.
How does aligning feedback to criteria save marking time?
You edit drafts instead of writing every comment from scratch, and consistent criterion checking means less re-reading, while you still control what reaches students.
Get started with Jeddle
Jeddle gives teachers and students instant, syllabus-aligned feedback powered by JeddAI.
Looking for study material? Browse Jeddle's Australian-English subject resources, or explore more articles on Using Jeddle.
