Moving answer-script evaluation from valuation camps to a screen touches a lot of people: exam office staff, scanning teams, evaluators, chief examiners, the results committee and eventually students who ask for revaluation. Piloting in one department is the safest way to find out how on-screen marking fits your rules before it carries a whole semester.
Step 1: Pick the right department and paper
Look for a department with:
- a head of department and chief examiner who want to try it, rather than ones being told to;
- one or two papers of moderate size, enough scripts to test allocation and second valuation, few enough to fall back on paper if needed;
- a mix of question types, so you see how long answers, diagrams and numerical working look on screen;
- a results date with some slack in it.
Avoid a final-year course for the pilot paper, since any delay there affects degree award.
Step 2: Write down your valuation rules first
On-screen marking makes your rules explicit, which is useful and occasionally uncomfortable. Before setup, agree in writing:
- How many valuation passes each script gets.
- The variance rule: how far apart two marks can be before a further look is triggered.
- Who resolves a variance and how that decision is recorded.
- Whether AI mark suggestions are shown to evaluators for this paper. If they are, the evaluator always has the final say on every mark.
These rules go into Proctorly Examination as configuration, and every mark is then traceable to them.
Step 3: Plan digitization and chain of custody
Scripts are digitized with chain of custody, so you can show where each one has been from the hall to the evaluator. For the pilot, decide where scanning happens, who handles answer books between collection and scanning, and how a missing or damaged script is reported. The scanning setup and equipment for your centers is agreed during onboarding. Answer-book tracking in hall operations helps here, because a script that was issued, written and collected is accounted for before marking starts.
Step 4: Prepare evaluators
Evaluators need less training than people expect, but they do need some. Cover:
- how anonymous marking works, since candidate identity is hidden;
- how to annotate and enter marks per question;
- the evaluator face check that runs while marking, and why it is there;
- what happens when they disagree with an AI suggestion (they simply enter their own mark).
Give each evaluator a short practice batch before live scripts are released.
Step 5: Run marking and watch the exceptions
During the pilot, the exam office should watch three things closely: scripts waiting for allocation, scripts that triggered a variance rule, and evaluators who are falling behind. A full rollout will feel the pressure in exactly these places.
Step 6: Carry marks into results
Marks go into results on the same record, with grading rules, SGPA and CGPA, the exception center and committee approval applying before controlled publication. For the pilot, check that the results committee is comfortable approving from the exception center rather than from a printed sheet.
What to measure
We do not publish a figure for how much faster results will be, because that depends on your scripts, panel and rules. Measure it yourself against the same paper’s previous cycle:
| Measure | Why it matters |
|---|---|
| Days from last exam to results approval | The headline question every registrar will ask |
| Share of scripts that triggered a variance rule | Shows how consistent the panel is |
| Totaling or entry corrections needed | Should fall, since the system adds totals |
| Revaluation requests and time to close them | Tests the record behind each mark |
| Evaluator feedback | Adoption depends on it |
After the pilot
Write a short report for the academic council with the measures above and what you would change. Departments that want to go further often connect the bank side next: questions built in the outcome-mapped question paper generator keep their mapping to outcomes and Bloom’s levels, so marks feed accreditation analytics without extra work. The full cycle is described on our exam management system page.
Frequently asked questions
Do we need the whole Examination module to pilot on-screen marking?
On-screen marking is a stage of Proctorly Examination. Talk to us about scoping a pilot around the stages you need.
Can online answers and hall scripts be marked together?
Typed answers from online sittings and digitized hall scripts can be marked in the same evaluator workflow, and handwritten working photographed during online exams is attached to the question. Confirm the setup for your papers with us.
Can AI decide a mark?
No. AI can suggest a mark beside the answer. The evaluator reads the answer and decides every mark.
Bring a sample paper and your valuation rules, and we will show how your scripts would move from hall to result. Book a demo.