Practice
CAT — Using mock tests as a measurement
A CAT mock is not a knowledge test with a score attached. It is a rehearsal of a set of decisions: which passages to read, which sets to attempt, when to abandon a question, how to spend the last five minutes of a section. Those decisions improve measurably with practice and analysis, and they are what separate candidates with similar ability.
How to analyse a CAT mock properly
- Spend longer on analysis than on the attempt. Three hours of review after a two-hour mock is a reasonable ratio, and almost nobody does it.
- For every question, classify: solved, could have solved, could not have solved. The middle category is where your score improvement lives.
- Review your selection decisions specifically. Was the set you skipped actually harder, or just unfamiliar?
- Track section-wise percentiles, not just the overall. Sectional cutoffs mean your weakest section constrains your calls.
- Look at the last ten minutes of each section. Panic decisions there account for a surprising share of lost marks.
On volume: taking many mocks without analysing them produces very little. A candidate who takes fifteen mocks and analyses each thoroughly will outperform one who takes forty and reviews the score. The mock is the raw material; the analysis is the preparation.
- The three-way classification, and what each pile means
- After every mock, sort each question into solved, could have solved, and could not have solved. The first pile is your current level. The second is your immediate opportunity and is almost always the largest — these are marks lost to selection, pacing or carelessness rather than to knowledge. The third is the only pile that calls for more study.
- Why the second pile matters most
- For most candidates, the gap between their score and their potential score is mostly in questions they could have done and did not reach, misread, or abandoned wrongly. That gap closes through better decisions rather than through more content, which is why a candidate who only studies after a mock rarely improves.
- Why section-wise percentiles beat the overall figure
- Calls depend on sectional performance as well as the aggregate. An overall percentile that has risen while one section has stagnated is a worse position than the number suggests, and only section-wise tracking makes that visible.
Look specifically at the last ten minutes of each section in your analysis. A section that ends in a scramble — several questions attempted in the final minutes, accuracy collapsing — is almost never a knowledge problem. It usually means the first ten minutes were spent on a set or a passage that should have been abandoned, and the whole section was played from behind as a result. That diagnosis is invisible in a score and obvious in a timestamped review.
Fewer mocks, properly analysed
Fifteen mocks each followed by two hours of analysis will produce a better candidate than forty taken and scored. The mock is raw material; the analysis is the preparation. Candidates who take mocks daily are usually substituting the measurable comfort of activity for the harder work of confronting what the last one showed.
Reading a CAT mock, where selection decisions matter more than the score.| What you see | What it means | What to change |
|---|
| Time sunk in a set you did not finish | Selection, the single largest lever in this paper. | Rank every set in the first minutes and solve in that order; review whether the ranking was right. |
|---|
| VARC accuracy low on inference questions | Answering from knowledge rather than from the passage. | For each wrong answer, find the line that supports the correct option. |
|---|
| Quantitative section unfinished | Exact computation where estimation would settle it. | Decide the precision each question needs before calculating. |
|---|
| Sectional scores uneven | Sectional timing means a weak section cannot be rescued. | Study time follows the weakest section. |
|---|
| Score volatile between mocks | Normal, given set difficulty varies. | Track selection decisions and accuracy rather than the raw percentile. |
|---|
