CAT Score vs Percentile: What the Marks Actually Buy You
There is no official CAT score-to-percentile table. The IIMs have never published one and show no sign of starting. Every table you have ever seen on this subject — including the one further down this page — was reconstructed after the fact from scorecards that candidates chose to share.
That is not a caveat to skim. It is the most useful thing on the page, because it explains something otherwise baffling: the marks needed for 99 percentile moved by several points between the last two cycles, and nothing unusual happened.
The reported bands
Figures below come from recent cycles. Each carries a working band. The band is the honest part — it is the width of the disagreement between the sources this was assembled from, rather than a hedge added afterwards to look careful.
| Percentile | Reported overall scaled score | Working band | What it typically opens |
|---|---|---|---|
| 99.9+ | ~110 (2025 cycle) | 105–115 | Comfortable across the older IIMs |
| 99 | ~96 (2024) · ~100 (2025) | 95–105 | Calls from most older IIMs, subject to profile |
| 95 | ~70 (2024) | 66–74 | Newer IIMs and strong non-IIM programmes |
| 90 | ~57 (2024) | 53–61 | A wide set of good non-IIM B-schools |
| 85 | Around the high 40s | 45–52 | Many respected regional programmes |
| 80 | Around the low 40s | 38–45 | Entry point for a broad tier of colleges |
Plan against the third column. The second is what one source reported for one paper in one year, and a candidate who targets exactly 96 because a table once said 96 has anchored their entire preparation to a number that has already moved. Notice too that the bottom rows are wider. Far fewer people at 80 percentile publish their scorecards than at 99, so there is simply less to reconstruct from down there.
One figure surprises almost every first-timer: roughly half the available marks has been enough for 99 percentile in recent cycles. The paper is not built to be finished.
Section-wise, which is what institutes actually screen on
Almost every institute applies a sectional minimum alongside the overall percentile. So the sectional picture matters more than candidates assume — a 99 overall carrying a 70 in one section gets rejected by shortlists that a balanced 97 clears comfortably.
Here is what roughly 99 percentile required in each section in the 2024 cycle.
QA needs the fewest marks. Not because QA is easier — because more candidates score badly in it, which drags the whole distribution downward and lifts any given score's rank with it. Run that logic the other way and you get VARC: a great many candidates read reasonably well, so the same rank costs more marks to reach.
There is a direct planning consequence, and it is worth acting on. At the top of the range a mark gained in QA buys more percentile than a mark gained in VARC. A candidate deciding where to spend a final month usually gets more movement from quant.
CAT Mock Test — see where your score actually lands
A raw score means nothing on its own. Take one full paper and read your overall and sectional percentile against everyone else who attempted it.
Four things that move the curve
They do not always push the same way. That is why the mapping wanders.
Paper difficulty
A harder paper means fewer marks are needed for the same percentile. Everyone's score falls together, so rank holds at a lower number.
Number of candidates
More candidates means the same rank represents a higher percentile. It matters most in the thin air above 99.5.
Slot scaling
Raw scores are normalised across the three sessions before ranking. For most candidates this is a nudge. At a cut-off it is not.
Section mix
One brutal section drags the overall band down on its own, which is why the sectional tables move before the overall one does.
The scaling step deserves its own sentence, because it is where most confusion begins. The exam runs in multiple slots with different papers, so raw marks are not directly comparable across them. A statistical normalisation converts everything to a common scale, and the percentile is computed on the scaled score, not the raw one. That is why your scorecard shows two different numbers and why they refuse to match.
Estimating from a mock score
You cannot do this precisely, and anyone claiming otherwise is selling something. You can do it well enough to plan. Four steps.
- Compute the raw score properly Correct × 3, minus wrong MCQs × 1. Typed-answer questions carry no penalty, so wrong ones cost nothing. Most candidates overestimate here by quietly forgetting the negatives.
- Adjust for the mock's difficulty Easier than a real paper? Shade down a few marks. Harder? Shade up. Everyone skips this step and it is worth more than the other three combined.
- Read the band, not the row Land in 66–74 and call it "around 95". Not 95.2. The precision you are reaching for does not exist in the underlying data.
- Check the sectionals separately An overall estimate hides the one weak section that will actually cost you shortlists.
The honest shortcut is to attempt a real past paper under real timing and score it against that same year's reported table. This removes step two altogether, because the paper and the table came from the same cycle. It is the closest thing to a calibrated reading available before results day.
Four ways this table gets read wrong
Skip the estimation entirely
Every step above exists because you are guessing at a number. Attempt a full paper and read the overall and sectional percentile directly instead.
- Overall percentile
- All three sectionals
- Accuracy breakdown
Questions candidates actually ask
How many marks are needed for 99 percentile in CAT?
Recent cycles put it in the region of 95 to 105 scaled marks out of 198. Around 96 was reported for the 2024 cycle and roughly 100 for 2025, which was widely described as a harder paper in DILR and QA. Plan against the band rather than either figure — the requirement moves by several marks each year with paper difficulty, and no official table is ever published to fix it.
Is there an official CAT score vs percentile table?
No. The IIMs release individual scorecards showing your scaled score and percentile, but never a public mapping between the two. Every table you find online is reconstructed after results from scorecards that candidates shared, which is why different sources disagree by a few marks at every level. The reconstruction is reasonably reliable near 99 percentile, where many scorecards circulate, and noticeably noisier at 80, where far fewer do.
Why does the same percentile need different marks each year?
Because percentile measures your rank among candidates, not your marks. A harder paper pushes everyone's scores down together, so a lower mark still lands at the same rank — and the marks needed for 99 percentile fall. Three other factors move it too: the number of candidates who appear, the statistical scaling applied across slots, and whether one section ran unusually brutal, which drags the overall band down on its own.
What is the difference between raw score and scaled score in CAT?
The raw score is what the marking scheme gives you directly: three marks for each correct answer, minus one for each wrong MCQ, with no penalty on TITA questions. Because the exam runs in multiple slots with different papers, raw marks from different slots are not comparable, so a statistical normalisation converts them to a common scale. Your percentile is computed on that scaled score. This is why the scorecard shows two numbers that do not match.
How many marks give 95 percentile in CAT?
Roughly 66 to 74 scaled marks, with about 70 reported for the 2024 cycle. That is a little over a third of the total 198 marks, which surprises candidates who assume a high percentile needs a high proportion of marks. It does not — CAT is designed so that finishing everything is not realistic, and the distribution of scores sits far lower than most first-time candidates expect.
Which section needs the fewest marks for 99 percentile?
Quantitative Aptitude, in most recent cycles — around 33 scaled marks against roughly 38 for DILR and 41 for VARC in the 2024 data. This is not because QA is easier. It is because more candidates score poorly in QA, which pushes the distribution down and makes any given score rank higher. The practical consequence is that a mark gained in QA is often worth more percentile than the same mark gained in VARC.
Can I estimate my CAT percentile from a mock test score?
Approximately, not precisely. Compute the raw score correctly including negatives, adjust for whether the mock ran easier or harder than a real paper, read the resulting figure as a band rather than a decimal, and check each section separately. The adjustment step is the one everyone skips and it matters more than the rest. The cleaner method is to attempt an actual past CAT paper under real timing and score it against that same year's reported table, which removes the difficulty adjustment altogether.
Does a 99 percentile guarantee an IIM call?
No. Percentile enters a composite calculation that also weighs Class 10 and 12 marks, graduation performance, work experience, and academic and gender diversity, with weights that differ from one institute to another and are published in their admission policies. Two candidates with identical percentiles routinely receive different call lists. A strong percentile makes you eligible for the calculation rather than finishing it.
What percentile do I need for a good MBA college?
It depends entirely on which colleges you mean, and the honest range is wide. The older IIMs generally sit in 99-plus territory once profile weights are applied; newer IIMs and strong non-IIM programmes commonly work in the mid-90s; and a broad tier of respected regional programmes accepts scores in the 80s. Build a target list first and work backwards from the institutes you would genuinely attend, rather than chasing a percentile number in the abstract.
Like what you read? Share with a friend
Senior Content Writer
Co-Founder at PrepGrind, working on education content, product development, and student-focused learning resources. I also write articles and guides for MBA/MBS entrance exams, covering preparation strategies, exam updates, and useful resources for students.
