I'm Pranshul Verma. I've been coaching CAT since 2020 — before that I was the General Manager at Career Launcher, one of the biggest CAT coaching chains in India. I left a stable corporate role to build Percentilers from scratch, and honestly? Best decision I ever made.
In the last 5 years, I've worked with 500+ students. Some came in scoring 60 percentile and walked out with IIM calls. Some came in already at 90 percentile and plateaued for months. I've seen it all. And the one thing I can tell you with full confidence is this — the way the internet tells you to prepare for CAT is mostly wrong.
This post is the real gyaan. No fluff, no fear-mongering, no "join a coaching centre now or you'll fail". Just what actually works.
The Lies the Internet Tells You About CAT Prep
Let's start here because ngl, some of this stuff is genuinely harmful.
Lie #1: "You need 6 hours of study every day." Nope. I've seen students clear CAT with 2–3 focused hours a day. I've also seen students grind 8 hours a day and score 70 percentile. CAT doesn't reward volume — it rewards quality and strategy.
Lie #2: "Start from January or you're already behind." Also not true. Some of the best results I've seen at Percentilers came from students who started in June and hit 99 percentile in November. The timeline matters less than the method.
Lie #3: "Buy every mock series available." This one hits different because it's so expensive and so counterproductive. I've seen students take 80+ mocks and not improve. Analysis beats volume every single time.
Lie #4: "VARC is about reading speed." Lowkey the most damaging myth for engineering students. Speed is maybe 20% of VARC. The rest is comprehension, inference, and understanding author intent. More on this later.
Data point: In a survey of 200 Percentilers students, the ones who scored 95+ percentile averaged 14 high-quality mocks with deep analysis — not 40+ mocks with surface-level review.
Truth #1 — CAT Is Not About Working Hard. It's About Working Differently.
Here's the vibe shift you need to make: CAT is not your 12th board exam. It's not about memorising formulas or finishing the syllabus. It's a test of how you think under pressure.
The paper is designed so that no one can attempt every question. The average 99 percentiler attempts maybe 45–55 out of 66 questions — and gets most of those right. That's it. That's the secret.
So the game isn't "can I solve this problem?" It's "should I attempt this problem right now, or skip it?"
That's a completely different skill. And most students never train for it because they're too busy doing question after question without building the metacognitive awareness to make those calls in real time.
At Percentilers, we spend a significant chunk of time on attempt strategy — which questions to pick, in what order, and when to cut your losses. This is one of the main reasons 15% of our students make it to top B-schools, compared to the industry average of around 3.5%.
The 3 Phases Every Serious CAT Aspirant Must Go Through
Phase 1 — Concept Building (April to June): This is where you lay the foundation. VARC reading habits, DILR pattern recognition, Quant fundamentals. No full mocks yet. Sectional tests only. The goal is understanding, not speed.
Phase 2 — Speed + Accuracy Building (July to August): Now you start adding the clock. Timed sectionals, topic-wise sprints, building your "picking sense" — learning which questions are your questions and which ones to skip. Mock frequency: 1 per week.
Phase 3 — Full Mock Mode (September to November): Minimum 2 full mocks per week, with thorough post-mock analysis. This is also when you identify your "exam day strategy" — the exact sequence and time allocation you'll use on the actual paper.
Most students mess up by skipping Phase 1 and jumping straight to mocks. Then they wonder why their score isn't moving. The foundation isn't boring — it's literally the thing that makes everything else work.
VARC — The Section Most Engineers Completely Misread
Real talk: if you're an engineer, VARC is probably your biggest threat. Not because you're bad at it — but because you're approaching it like a Quant problem.
Engineers want to find "the answer". In RC, there often isn't one clean answer — there are answers that are more right than others, based on what the author is actually saying, not what logically follows from the passage.
Here's the thing: CAT RC questions are almost never about inference from external knowledge. They're about understanding the author's tone, purpose, and argument. That requires a completely different reading mode than what most engineers are used to.
What works:
- Read 1–2 quality articles daily (The Atlantic, Aeon, The Hindu Opinion, Livemint) — not for vocabulary, but for understanding argument structure
- For Para Jumbles and Para Summary: forget logic, follow the author's rhetorical flow
- In RCs, identify the author's main claim first — then all questions become easier
- Stop highlighting everything — train yourself to hold the argument in your head
In 2025, VARC had 24 questions — 16 RC and 8 VA. The students who cleared 98%ile in VARC at Percentilers averaged 20+ correct out of 24. That requires consistent reading habits, not last-minute cramming.
DILR — Where 99 Percentile Is Won or Lost
DILR is the most unpredictable section of CAT. It's also the one where the gap between 70 percentile and 99 percentile is the narrowest — because it's mostly about set selection, not raw ability.
In any given CAT paper, there are roughly 4–5 DILR sets. Some will be solvable in 8 minutes. Some will destroy you for 20 minutes and give you nothing. The skill is knowing which is which — within the first 60–90 seconds of reading the set.
This is what I call the "First 90 Seconds Rule": scan the set, identify the constraints, estimate solve time. If it looks like a nightmare, mark it and move on. No attachment. No ego. Move on.
What to practise:
- All CAT papers from 2015–2024 (the actual papers, not third-party recreations)
- Timed sectionals with 4 sets in 40 minutes — train the selection reflex
- Types to master: Grids, Scheduling, Network Flow, Rankings, Venn Diagrams, Games & Tournaments
Data point: In CAT 2024, just solving 3 DILR sets completely and accurately was enough for 95+ percentile in that section. You don't need to do everything. You need to do the right things.
Quant — Stop Doing Every Topic. Do These Instead.
CAT Quant is not GATE. It's not JEE. It's not even your college semester exam. The level of math is roughly Class 10–11, but the application is sneaky and the options are designed to trap overconfident engineers.
The biggest mistake I see: students spending 3 weeks on Permutation & Combination when 2–3 P&C questions appear in the actual exam. Meanwhile, Arithmetic (which is ~35% of Quant) gets barely 2 weeks of attention.
Here's the honest priority list for Quant:
- Non-negotiable (do these first): Arithmetic (percentages, ratios, time-work, speed-time), Algebra (equations, inequalities, functions), Number Systems
- High ROI: Geometry (triangles, circles, coordinate geometry), Modern Maths (P&C, probability — but only the standard types)
- Low ROI (only if you're targeting 99.9%ile): Advanced P&C, complex number theory, advanced coordinate geometry
Also — learn approximation. CAT is a multiple choice exam. You don't need the exact answer, you need the closest option. Most engineers waste 3–4 minutes calculating precisely when a 30-second approximation would give the same answer.
The Mock Test Trap (And How to Avoid It)
This is my personal soapbox and I will not stop talking about it.
Mock tests are only useful if you analyse them. Taking mock after mock without spending equal time on analysis is like going to the gym and never tracking your lifts — you're just sweating without getting stronger.
The minimum analysis protocol after every mock:
- For every wrong answer: why was it wrong? Concept gap, silly error, or misread question?
- For every skipped question: was it the right call? Could you have gotten it in 2 minutes?
- Section-wise: where did you lose the most time? Where did you attempt questions you shouldn't have?
- Overall: what's your accuracy rate on attempted questions? (Target: 80%+)
You should spend at least as much time on analysis as you spent taking the mock. So if the mock was 2 hours, spend 2 hours on the review. Non-negotiable.
Start full mocks only in September. Before that — sectional mocks and topic-wise tests. Full mocks taken too early just demoralise you without giving actionable data.
What Percentilers Students Do Differently
I'm not going to tell you Percentilers is the best coaching in India. That's not a thing that matters, and frankly, it depends on the student. What I can tell you is what we do differently — and why it works for the students it works for.
We don't have 5,000 students per batch. Our batches are small enough that I actually know my students' names, their weak areas, their test-taking habits. When your score isn't moving, I'm looking at your mock data — not just telling you to "practise more".
We're obsessed with strategy over content. Most coachings dump 400 hours of video content on you. We focus on how to think, not just what to study.
And the results speak. 15% of our enrolled students convert to top B-schools — IIMs, XLRI, FMS, IIFT. The industry average is 3.5%. That's not a marketing claim — that's 5 years of tracking outcomes.
If you're a self-motivated student who wants strategy-focused, personalised coaching — Percentilers is probably a good fit. Check us out at percentilers.in.
FAQ
When should I start CAT 2026 prep?
If you're reading this in April, you're already at the right time. April to June is ideal for concept building. Starting in June is still fine. Starting in August is tough but doable if you're disciplined. Starting in October? Very risky unless you have prior prep.
How many hours a day do I need to study for CAT?
2–3 focused hours on weekdays, 4–5 on weekends, is enough for most people. The key word is focused — phone away, no multitasking, real practice. Don't let anyone guilt you into thinking you need 8 hours a day. Quality beats quantity every time.
Can I crack CAT without coaching?
Yes, 100%. Plenty of self-studiers crack CAT every year. The risk is spending months going in the wrong direction without feedback. Coaching helps if it provides structured strategy and real feedback — not just content dumps. If you're self-studying, be extra disciplined about mock analysis and strategy.
What's the difference between CAT and other MBA entrance exams like SNAP or XAT?
CAT is the hardest of the three in terms of DILR complexity. SNAP is faster-paced with lower difficulty. XAT has an additional Decision Making section. If you crack CAT, SNAP and XAT come relatively easily — they share 80% of the same syllabus. But each needs its own last-month specific prep.
Ready to start your CAT 2026 prep the right way? Head to percentilers.in to explore our courses — or just stalk our free content first. Either way, we've got you.