Re-open the guide and re-read the criteria, every script.
Hunt each requirement through templates, tables and appendices.
Compose every comment from a blank box.
Watch your own standard drift as the night wears on.
Open a script with the ticks already sitting where the evidence sits.
Edit comments that already name the criterion and quote the student.
Overrule anything you disagree with. It is your call, every time.
Sign, hand back annotated documents, and the batch clears.
Timing measured on one trainer’s batch of 24 assessments at 16¼ minutes by hand. Real evidence, not yet a cohort study. How we measured it.
On average, I can now complete Knowledge Questions in around 5 minutes instead of 15, and Task 3 assessments in around 15 minutes instead of 40+min, including marking, providing feedback in Moodle and uploading the marked assessment.”
Every criterion pre-matched to where its evidence sits in the script, ticked or flagged before you open it.
Each one names the guide’s requirement, quotes what the student wrote and says what to add. You edit instead of compose.
The 30th script is read as carefully as the 1st. No drift because you’re tired, rushed or three coffees deep.
Word templates, PDFs, portfolios, scans. Nothing gets retyped into quiz fields first.
The batch is compared for copying, against classmates and the sample answer, on your machine.
Nothing is final until you make it final. Confirm, edit or overrule every single mark.
Across business, management, accounting, sciences and maths, and written communication units. Whatever your students actually submit.
Arts assessment. Judging creative work is a human reading of intent. It should stay that way.
Apollo never returns a mark. It returns a prepared case: evidence found, criterion cited, comment drafted. Then it asks the person whose name goes on the outcome to decide.