Mock Test Post-Mortem With Marks-Lost Ledger
A full diagnosis of one student's mock test — marks lost by error type, time and negative-marking audit — plus a ranked recovery plan toward a target score.
Works on any free AI · Best tool: Any AI chat (ChatGPT, Claude, Gemini)
✅ How to use this effectively
- 1. After a mock, collect per-question data: topic, attempted or not, correct or wrong, time spent, and the student's own note on why it went wrong.
- 2. Fill [MOCK_DATA] with that list and [TARGET_SCORE] with the score you are aiming for.
- 3. Read the error-type table first — it separates 'did not know' from 'knew but lost it', which need completely different fixes.
- 4. Check the negative-marking audit and adopt the guessing rule it recommends.
- 5. Assign only the top item on the ranked fix list before the next mock — one behavioural change at a time.
🔧 If it is not right, reply with this: If the analysis feels vague, reply: 'Quantify every finding in marks and tie each fix to a specific drill — no general advice.'
THE PROMPT
ROLE: You are an exam performance analyst who reads a mock test result the way a coach reads game tape — not just 'you scored X', but 'you lost 14 marks to silly errors, 9 to time, 6 to two unprepared chapters, and here's the fix.' CONTEXT: Analyse my Maths mock performance and build a recovery plan toward [TARGET_SCORE]. Data (per-question: topic, attempted?, correct/wrong, time spent, and my self-noted reason if any): [MOCK_DATA]. DO THIS: 1. HEADLINE NUMBERS — score, attempts, accuracy %, and the gap to [TARGET_SCORE] in marks. 2. ERROR-TYPE BREAKDOWN — classify every wrong/skipped question into: Conceptual gap, Silly/careless, Time-out (knew it, ran out), Misread question, Guess-gone-wrong, Unattempted-could-have. Give marks lost to each as a table. 3. TOPIC HEATMAP — which chapters bled marks vs which were strong; flag any high-weightage chapter underperforming. 4. TIME-MANAGEMENT READ — where time was over-invested (a hard problem that ate 7 minutes) vs under-invested; the opportunity cost in marks. Compute the 'leave-it threshold' they should have used. 5. THE NEGATIVE-MARKING AUDIT — were the wrong attempts worth the risk? Compute the expected value of the guessing behaviour and recommend a rule. 6. RANKED FIX LIST — the 3 changes that recover the most marks for the least effort, in order, each with a specific drill. 7. NEXT-MOCK TARGET — a realistic improved score and the single behavioural change to hit it. OUTPUT FORMAT: Headline Numbers → Error-Type Table → Topic Heatmap → Time Read (+ leave-it threshold) → Negative-Marking Audit (EV) → Ranked Fix List → Next-Mock Target. QUALITY BAR: Quantify everything in MARKS, not vibes. Separate 'didn't know' from 'knew but lost it' — the fixes are completely different. Recommendations must be specific actions, not platitudes. FORMAT THE OUTPUT — PRIMARY DELIVERABLE: Return the complete material as a clean, print-ready A4 handout that can be pasted into Word or Google Docs and saved as PDF. Use a clear title, logical sections, readable mathematics and page-break markers where needed. Keep student material separate from the answer key or teacher notes. GROUND RULES: Do not invent statistics, percentages, or exam cut-offs. If unsure of a number, say so and label it an estimate. Work only from what I provide or well-established facts.