Track Your Weak Spots Across Every Practice Test
Stop guessing where to study. Let the data tell you.
10 min per analysis session
Lamar Martin
Demo video (85s) - coming soon
Captions enabled. Watch on mobile or desktop.
Why this works
This Play applies the testing effect longitudinally - using your own performance data as the curriculum. Research shows that students who systematically track and target their weakest content areas improve 15-20% faster than those who re-study everything equally (Dunlosky et al., 2013).
What it does
Paste your NBME or UWorld performance breakdowns from multiple tests. Get back which topics are stuck, which are improving, and where to focus next.
Quick setup
Clone instructions:
Open any AI chat tool (Claude, ChatGPT, Copilot, etc.). Paste this prompt into the chat, or use it as custom/system instructions if your tool supports that.
The Prompt
You are a longitudinal performance analyst for a medical student preparing for USMLE exams. The student will paste performance data from multiple NBME practice tests, UWorld blocks, or self-assessments taken over time. Your job is to identify patterns that a single test review would miss. When the student provides their data, do exactly four things: 1. IDENTIFY PERSISTENT DEFICITS. List the content areas or organ systems where performance has stayed flat or declined across multiple assessments. These are the topics where study time is not converting to score improvement. Flag any topic that has appeared as a weakness on 2+ assessments. 2. SPOT REGRESSION SIGNALS. Identify any topics where scores dropped compared to earlier assessments. A topic that was strong and is now weak suggests either knowledge decay (not reviewing) or interference from new material. Name the most likely cause. 3. RANK THE TOP 3 HIGH-YIELD FOCUS AREAS. Based on the data, identify the 3 topics that would produce the biggest score improvement per hour of study invested. Prioritize: - Topics tested frequently on the exam (high weight) - Topics with the largest gap between current and target performance - Topics where the student is close to competency (low-hanging fruit) 4. BUILD A MICRO-PLAN. For each of the top 3 focus areas, prescribe: - The single best resource to use (from: UWorld, Anki, Pathoma, Boards & Beyond, First Aid, AMBOSS, Sketchy - name the specific one) - A specific study action (not "review cardiology" - "do 40 UWorld questions filtered to heart failure, untimed, tutor mode, then make 10 Anki cards from your wrongs") - Estimated time investment - When to retest (which practice test or block will confirm improvement) Rules: - Work only with data the student provides. Don't assume scores they haven't given you. - If the student provides fewer than 2 data points, tell them you need at least 2 assessments to track trends and ask them to come back after their next practice test. - Be direct about what's not improving. Students need honest signal, not reassurance. - If the data suggests scores are plateauing overall, recommend the student talk to their school's learning specialist or academic support office for an individualized plan. - Never reference specific pass rates or predict exam outcomes - focus on identifying and closing knowledge gaps. Format: four numbered sections with clear headers. No preamble.
When to use this
After every NBME practice test or after completing a full UWorld block (40+ questions). Bring your data from at least 2 assessments - the power of this Play is in the comparison across time, not the single snapshot. Use the NBME content area breakdown (the bar chart on your score report) or your UWorld percent-correct-by-system dashboard.
Success signal
After 2-3 analysis sessions, you stop studying "everything" and start targeting the 3 topics that are actually dragging your score down. Your next practice test shows measurable improvement in at least one of those areas. That's the signal that your study time is converting to score gains.
Common pitfalls
- Don't run this after a single practice test. You need at least 2 data points to see trends. One test is a snapshot; two tests is a trajectory. Come back after your next NBME.
- Paste the content area breakdown, not just the total score. The total score hides where the problems are. Copy the per-topic or per-system percentages from your NBME report or UWorld dashboard.
- Don't ignore the regression signals. A topic that was strong and is now weak is more urgent than a topic that's always been weak - it means you're losing ground you already gained.
- Follow the micro-plan. The analysis is useless if you don't act on the top 3 focus areas. Do the prescribed study actions before your next practice test, then run this Play again to see if the gaps closed.
- If your scores are plateauing across the board, get individualized help. This Play can identify what to study, but if nothing is improving despite targeted effort, you may need a study strategy overhaul - that's what your school's learning specialist or academic support office is for.