A mock test is not a performance. It is a diagnostic tool.
The score you get tells you very little. The data behind the score — which questions you attempted, in what order, how long you spent on each, where you were right, where you were wrong, and why — tells you everything.
Most students never look at that data. This is why most students plateau.
The Problem With How Most Students Use Mocks
The typical mock cycle looks like this: take the mock → check the score → feel good if it was high, feel bad if it was low → identify two or three questions you got wrong → revise those topics → take the next mock.
This is not analysis. This is score-checking with extra steps.
The questions you got wrong are not the primary problem. The questions you got right for the wrong reasons, the questions you skipped that you should have attempted, the questions you spent too long on — these are where the real losses are.
A mock is a 120-minute dataset. Most students use 5 minutes of it.
The Four Layers of Mock Analysis
Layer 1: Time Distribution
Before looking at individual questions, look at your time log.
- How much time did you spend on each section?
- Within each section, how was time distributed across questions?
- Were there questions where you spent more than 4 minutes?
- Were there questions where you spent less than 30 seconds?
Questions you spent 4+ minutes on are almost never worth it. In DILR, a set that costs you 12 minutes and yields 2 correct answers is a poor trade compared to a set that costs 8 minutes and yields 4. In QA, a question that costs 5 minutes is a loss regardless of whether you got it right.
Layer 2: Attempt vs Accuracy
For each section, calculate:
- Questions attempted / Questions in section
- Correct / Attempted (accuracy)
- Net score from attempted questions
A student who attempts 18 QA questions and gets 14 right (77% accuracy) will outperform a student who attempts 22 and gets 15 right (68% accuracy) — despite attempting fewer questions. Higher accuracy from fewer attempts is often the better strategy.
Most students attempt too many questions and dilute their accuracy. Finding your optimal attempt range — the number of questions where your accuracy remains above 75% — is one of the most valuable things mock analysis can reveal.
Layer 3: Question Classification
Every question you encountered in the mock falls into one of four categories:
Should have got right — got right. These are fine.
Should have got right — got wrong. These are the most important. These represent either a concept gap or a careless error. Find out which. If it is a concept gap, revise that topic. If it is careless error, identify the type (misread, calculation slip, wrong option marked) and track frequency.
Should not have attempted — got wrong. These are negative marks you chose to take. Ask why you attempted this question. Was it overconfidence? Poor time management? A misleading first impression?
Should not have attempted — got right. These are lucky guesses. They feel good but mask risk. If you had been wrong (which you might be next time), you would have lost a mark. Track these — they represent a false sense of accuracy.
Layer 4: Sectional Pattern
After 5-6 mocks, patterns emerge:
- Which question types do you consistently get wrong?
- Which topics are your strongest — and are you spending proportionate time on them?
- Is there a time in the section (first 10 minutes vs last 10 minutes) where your accuracy drops?
- Do you perform differently under time pressure?
These patterns tell you where your preparation should go. Not "revise QA" — but "revise circular seating arrangements" or "stop attempting probability questions" or "spend 2 fewer minutes on each RC passage."
How Many Mocks Should You Take
This is the wrong question. The right question is: how many mocks can you fully analyse?
If you take 30 mocks and analyse none of them, your score will not improve. If you take 10 mocks and analyse each one thoroughly, your score will improve significantly.
The optimal cadence for most students preparing for 3-4 months:
- First month: One mock every 10 days. You do not have enough content prepared to benefit from more.
- Second month: One mock per week. Content is mostly done, building speed and familiarity.
- Third month: Two mocks per week. Full exam simulation, fine-tuning strategy.
- Final 2 weeks: One mock every 3-4 days. Maintain sharpness without burning out.
Analysis should take at least as long as the mock itself. A 2-hour mock deserves 2 hours of analysis.
The One Behavioural Pattern That Predicts CAT Success
The students who improve the most between their first mock and their last mock are not the most intelligent students. They are the students who are honest with themselves about their data.
If your DILR analysis shows that you consistently waste time on seating arrangements but perform well on scheduling, the response is to deprioritize seating arrangements and maximize scheduling attempts. This requires admitting that seating arrangements are a weakness — and many students resist that admission.
The data does not care about your self-image. It tells you exactly where your marks are going. The students who act on that data — specifically and without ego — are the ones who sit at 99 percentile.
Mocks are the most valuable tool in CAT preparation. Most students treat them like practice tests. The students who treat them like performance reviews are the ones who make the jump.