How to Improve Your CAT Quant Score, Step By Step Now

QUANT

The jump people ask about most is the one from a middling quant section to a strong one, and it is genuinely the most achievable move in CAT preparation. It is also the one most often attempted with the wrong method, because the thing that produced a middling section is not a lack of effort.

Before anything else, one correction. Percentile is your position relative to everyone who sat the paper, and there is no fixed mapping from marks to percentile, so nobody can tell you the marks that separate those two bands. What can be described precisely is what changes between an aspirant who is scoring mid-table and one who is not, and that is a list of about four things.

What Produces a Middling Quant Section

The profile is remarkably consistent, and recognising it is most of the diagnosis.

The typical aspirant in this band has covered a great deal of the syllabus and mastered very little of it. They can start almost any question and finish rather fewer, which under a clock produces a lot of half-attempts and a lot of time spent without marks attached.

They also have a large volume of solved questions and almost no record of why the wrong ones went wrong. Without that record the method never changes, so the next hundred questions get solved exactly as the last hundred were.

And they attempt too many questions, because attempting feels like scoring. Negative marking converts that instinct into a steady drain that they cannot see, since nobody adds up what their wrong MCQs cost.

The Four Changes That Actually Move It

Each of these is narrow, and together they account for most of the movement in this band.

Depth Is What Converts Recognition Into Speed

An aspirant who has done four hundred arithmetic questions recognises a ratio problem as a ratio problem in seconds. One who has sampled six chapters lightly spends thirty seconds working out what they are looking at, and that thirty seconds is repeated across the whole section. Working a chapter such as percentage questions to a stable accuracy buys speed everywhere, because percentage thinking sits inside so many other questions.

The Error Log Is the Slowest Habit to Start

It takes about a minute per error and nearly nobody does it, because reviewing feels less productive than solving. Yet without it there is no mechanism by which your method improves. Grouping a fortnight of errors by cause usually reveals that two causes produced most of the mistakes, and fixing two specific things is a far smaller job than getting generally better.

Selection Is Where the Marks Are Hiding

Run the arithmetic on a real section. Take a paper of 66 questions worth 198 marks. A candidate attempting 63 of them at 41 percent accuracy gets roughly 26 right, worth 78 marks gross, and nets 54. The 37 wrong attempts cost 24 marks against simply leaving those questions blank. That is not an argument for a specific attempt count, because the right count depends on you and the paper. It is an argument that an attempt only pays when your odds on that question are good.

Behaviour	Middling section	Strong section
Chapter coverage	Wide, shallow	Deep in recurring areas first
Error review	Reads the solution	Names a cause per error
Attempts	Chases a count	Chases good odds
TITA questions	Treated like MCQs	Attempted whenever reasoned

A Twelve Week Shape

The sequence matters as much as the content, so here is a workable structure.

Weeks one to four are depth weeks. One chapter at a time from the recurring end of the syllabus, four sessions in each, attempting before any solution is opened, error log updated every session. The target is stable accuracy in the chapter across two sittings, not coverage.

Weeks five to eight add the clock. Same depth work, plus one timed sectional a week where the review takes longer than the test. This is where selection discipline gets built, because it only exists under time pressure.

Weeks nine to twelve are consolidation. Chapters chosen by what the error log says rather than by the generic order, mixed revision from chapters closed earlier, and two timed sections a week.

What to Measure While You Wait for the Score

Sectional totals are noisy, so track the inputs instead and you will know months earlier whether this is working.

Accuracy on committed questions is the first number. It should rise as depth accumulates, and it rises before the total does because early on it comes with fewer attempts.

Marks lost to wrong MCQs is the second. Multiply your wrong MCQ count by one. That figure should be falling as selection improves, and it usually falls faster than anything else on this list.

The third is the share of your errors that you can name a cause for. If most errors are still unclassified after a month, the error log exists in name only, and the method is not being corrected by anything.

Why the Score Lags the Behaviour

A sectional total is your behaviour plus the paper you happened to get. Across one mock the paper dominates; across five, the behaviour does. That is why two or three flat mocks after a real change in method is the expected pattern rather than a warning sign.

The Chapters Worth Protecting

Once a chapter reaches stable accuracy it starts decaying unless it is used. Reserve one session a week for questions from chapters closed more than a month ago, chosen at random. Our guide on building speed and accuracy together covers how to keep that from crowding out new work.

How to Choose the Next Chapter

By the second month the generic sequence should stop driving your plan, and something more specific should take over.

List the chapters that produced errors in your last three timed sections, and count the errors per chapter. Then cross that against how often each area has historically recurred. A chapter producing many errors in a heavily recurring area is the obvious next target; one producing many errors in a rarely recurring area can wait, though it should not be deleted.

The judgment people get wrong here is treating a chapter they dislike as a chapter that is not worth the time. Those are different claims, and the first is usually a report about an early bad experience rather than about return per hour. Geometry gets postponed for exactly this reason, and by October the postponement has become permanent.

One more filter is worth applying. If two chapters look equally worth doing, take the one that appears inside other chapters. Percentage and ratio fluency quietly raises your speed in several places at once, while a self contained area raises it in one.

What Not to Do in This Band

Three responses are common at this stage and all three slow the move down.

The first is adding hours. An aspirant already studying five hours a day who moves to seven usually adds two hours of the same unreviewed practice, which is the activity that produced the plateau. Fewer, denser hours with a review habit attached beat more hours without one.

The second is switching sources. New books and new question banks feel like progress and change nothing about method. The questions were never the constraint.

The third is chasing the hardest questions. Difficult questions feel like the route to a higher band and they are not where the marks are. A section is decided far more by reliable execution on the questions you can handle than by occasional success on the ones you cannot.

If your quant has been flat for a while despite steady work, the cause is usually one of the four changes being absent rather than effort. A CAT preparation strategy review will name which, and a personalised CAT preparation plan can hold the depth weeks in place.