What are the best CAT preparation strategies and score improvement tips from an IIM A alumnus?

An IIM A alumnus with 10 years of CAT mentoring experience has shared some of the most grounded, specific guidance we've come across — cutting through the noise on everything from VARC inconsistency to admissions cutoffs. Here's what that experience distills to.

On VARC inconsistency

Fluctuating scores — swinging from 40+ down to below 20 — are rarely a skill problem. In our experience, this pattern almost always points to wavering focus and test anxiety rather than any real gap in language ability. The fix is deliberate: practice under suboptimal conditions. Take mocks at odd hours, simulate pressure, and build the habit of refocusing mid-test when distraction sets in. For a student who scored 97.32 percentile in CAT 2023, the advice holds — don't panic into changing your approach. Consistent, focused post-mock analysis is what moves the needle, not reactive strategy shifts.

On mock analysis

The instinct to spend hours dissecting every mock is understandable but counterproductive. What we've seen work better: spend only 10 minutes on a preliminary section review yourself, then have a mentor spend 30 minutes per section with you — specifically on question selection strategy and option elimination techniques. The goal isn't to understand every question you got wrong. It's to sharpen the judgment of which questions to attempt in the first place.

On DILR improvement

A perspective we've tracked consistently from serious mentors: all DILR sets are doable. The differentiator isn't raw practice volume — it's finding the optimal approach for each set quickly. This has an important implication for where you source your practice: it matters less which material you use and more that it maintains genuine CAT-level difficulty. Don't drift toward easier sets because they feel productive.

On Quant time management

The specific benchmark worth internalizing: solve 5 easy questions in 6 to 8 minutes. That front-loaded efficiency is what frees time for harder problems later in the section. As a pacing check, if you're attempting at least 20 of 22 questions, you're moving at roughly the right speed. Under that threshold, time management — not concept gaps — is likely what's holding the score back.

On coaching centre mocks vs. actual CAT

This is one of the more practically important insights we've seen from experienced mentors, and it's worth stating plainly: coaching centre mocks are deliberately harder than the actual CAT, in part to justify the fee structure. The direct consequence — if you're consistently hitting 95 percentile on coaching mocks, you very likely have the capability for 99+ on the actual exam. Don't let mock percentiles deflate your confidence or cause you to over-correct your preparation.

On profiles and admissions cutoffs

The numbers here are specific and worth knowing. For IIM Ahmedabad, 99.8+ percentile is the realistic threshold for candidates with strong but standard profiles. That said, candidates with exceptional backgrounds — CAs, or graduates from top IITs or NITs with a 9/9/9 grade profile — tend to face lower effective percentile requirements. On undergraduate college quality: from what we've tracked across admissions cycles, UG college matters less than many aspirants assume. Even graduates from tier-3 colleges can and do break through — if they maximize their CAT percentile. The percentile is the lever; pull it as hard as you can.

On exam anxiety

The recommendation is blunt and quantified: complete 50 to 60 mocks before exam day. Volume builds a kind of muscle memory that arguments and affirmations simply don't. The anxiety before CAT mirrors what students felt before board exams and JEE — and in hindsight, that stress almost always looks disproportionate. The perspective shift comes after the exam; the preparation work has to happen before it.

The overall picture that emerges from a decade of mentoring: focus on fundamentals, analyze mocks strategically rather than exhaustively, simulate realistic conditions, and track consistent percentile performance rather than obsessing over absolute scores in any single attempt.

---