Three slots, one exam, and a normalisation step in between. Why your raw marks are not the number that decides anything.
Your percentile is not computed from your marks. It is computed from a scaled score, which is your raw score after it has been adjusted for how difficult your particular slot was. CAT runs in three slots on the same day, each with a different paper, and normalisation exists so candidates across slots can be compared fairly. The practical consequence: a lower raw score in a harder slot can produce the same scaled score as a higher raw score in an easier one.
Straightforward arithmetic from the marking scheme.
That last line is worth more than it looks. A TITA question you can reason your way towards is a free option: there is no downside to attempting it. Candidates routinely leave TITA questions blank out of habit carried over from the MCQ sections, which costs marks for no reason.
Three slots, three different papers, one ranking. If the afternoon paper happened to be harder, candidates in that slot would be penalised for nothing they did. Normalisation corrects that.
The method is a statistical adjustment based on the distribution within your own slot — broadly, the slot's average performance, how spread out the scores were, and how the strongest candidates in that slot performed. Those statistics calibrate a transformation that converts your raw score into a scaled score comparable with every other slot. Scaling is applied to each section as well as the total.
You cannot work out your percentile from your raw score and last year's score-versus-percentile table. The raw score needed for a given percentile shifts with slot difficulty — a harder slot's threshold sits higher than an easier slot's for the same percentile. Published "score vs percentile" charts are rough historical guides, not predictions.
Percentile is a rank position, not a mark. A 95th percentile means you performed better than 95% of candidates who took the exam that year. It is relative to that year's pool, which is why the same scaled score can produce slightly different percentiles across years.
You receive percentiles for each section and overall. Both matter: most institutes apply a sectional minimum, and a strong overall percentile with one weak section can fail that filter before anything else is considered.
The percentile gets you considered. It does not get you a call on its own, and it is a smaller part of the final decision than most candidates assume.
Shortlisting uses a composite score that also weighs Class 10 and 12 marks, graduation, work experience and diversity — how shortlisting actually works, and you can see the arithmetic on your own profile with the composite score calculator.
And at the final stage, after interviews, the CAT score typically carries less weight than the WAT-PI rounds do. That surprises almost everyone — WAT-PI explained.
Nothing about normalisation is actionable, and that is the useful conclusion: stop trying to game the slot, and stop estimating percentiles from mock raw scores. The two things that are still in your control are sectional balance — because sectional minimums are a hard gate — and attempting every TITA question you can reason about.
The rest of the preparation plan is in the compressed plan and the mock review method.
In three steps. Your raw score is computed from the marking scheme. That raw score is then normalised against the statistics of your own slot to produce a scaled score, because CAT runs in three slots of differing difficulty. The percentile is then computed from the scaled score - not the raw score - both section-wise and overall.
An adjustment that makes scores comparable across the three slots, since each slot's paper can differ in difficulty. It uses the distribution within each slot - typically the slot mean, the spread, and the performance of the top candidates - to convert raw marks into a scaled score. A lower raw score in a harder slot can map to the same scaled score as a higher raw score in an easier one.
Three marks for a correct MCQ answer, minus one for an incorrect MCQ answer, and zero for a question left unattempted. TITA questions, where you type the answer rather than select it, carry no negative marking - which makes attempting every TITA question you can reason about a free option.
Slot difficulty varies year to year and no slot is consistently easier. That is precisely what normalisation exists to correct for, so choosing a slot in the hope of an easier paper is not a strategy worth building around.
Not directly, and not by any published formula. Your marks give a raw score, the raw score is normalised against the statistics of your own slot to produce a scaled score, and the percentile is computed from that scaled score. So the same marks in two different slots can produce different percentiles, and no marks-to-percentile table can be accurate.
No. The normalisation step uses the distribution of scores within each slot, which is not released, so the conversion cannot be reproduced from outside. Tables offering raw-score-to-percentile conversions are built from previous cycles, which had different papers and different candidate pools.
Not sure how to approach an alumnus? How to contact MBA alumni for honest guidance — what to ask, and why a verified Real One answers differently.
Same score, very different outcomes depending on the rest of the profile. Verified alumni who have been through it.
Find a Real One