Beginner’s Guide to CAT Prep

The right time could be as early as November of the previous year or as late as June and will depend on …

1. How good you are with the subject i.e. Math, English language and Reasoning. 

If you are already very good … was great at maths in class X & XII, understood it very well and hence can recollect most concepts; is fluent with English, have a voracious reading habit; puzzles always interested you, are aware of common-place puzzles like magic square, weighing puzzles of identifying counterfeit coins, grid arrangement puzzles, etc … AND …

2. How much time do you have to devote to studies per day 

… AND you have on an average 4-6 hours per day for CAT prep, then you can start even as late as June. The minimum time needed to “complete the syllabus with just adequate practice”, before you start seriously and regularly with a Test Series, is approx 450-500 hours. @5 hours per day, it will require 100 days. Assuming you should move on to a Test Series latest by September-mid, it will mean the latest you need to start is June-start. Just keep in mind, starting this late is only if you are really good academically and also that you have no ‘slack’ and will need to prep every day for 4-6 hours for next 4 months.

If you are good with subject matter, but can just devote 2 hours per day for CAT prep, you should begin around February. 

If you are average with the subject matter, the above mentioned 450-500 hours can balloon to 800 hours. If you have 4-5 hours per day, you ought to start by Feb. But if you have just 2-3 hours per day, you need to start by previous November or October.

Also see the immediate next question of “how much time do I need to devote?”

Usually students overestimate the time they will study per day. With high enthusiasm at start, they plan on studying 4+ hours each day. But with college exams or work-pressures, the plan derails, there are long gaps of CAT-prep being forgotten, or one simply does not find enough time. So it is best to be realistic in estimating how much time you can devote without over-stretching yourself.

Starting early eases the time pressure, there is slack available to take care of lapses, you can practice more, you can start Test Series earlier … thus when in doubt, start early.

There are late starters who start in July or some even later than that. And the number of such students are not less. However this will compromise the prep in some way or other … e.g. skipping few topics altogether; late start of Test Series; not adequate practice of topics done; VARC and DILR sections neglected; and not to forget the continuous chatter in your mind of having started late. Only if you are really good with the subjects, your learning speed is very high and have high mental strength, starting in July or later may work.

Just covering all concepts will be about 150-250 hours depending on the course you may follow. And you will need around twice as much time to practice questions on the concepts learnt.

Thus, to just complete the syllabus and for just adequate practice, one will need atleast 450-500 hours. All this is assuming your starting level, your learning ability and your question solving pace is above average or atleast average. If the starting level is below average or you are a slow learner, the hours needed can increase upto 800 hours in total. 

Divide this 500 hours to 800 hours with the number of days left till mid-August. Increase the resulting value by 10-15% to take care of any days missed due to illness, mood, family matters, etc. And that will be the time you need to devote daily. E.g. if you approximate your need to 600 hours and are starting in mid-Feb, then it will be (600 hrs)/(180 days) × 1.1 = 3.6 hours/day.

Why-mid-August? Ideally you need to finish “syllabus and adequate practice” by mid-August and then in last 3 months, turn your focus on Test Series, past year questions (PYQs), practicing tougher/mixed-bag questions. If the time needed daily looks impossible, then your Test Series readiness will get delayed beyond August-mid; you can push starting Test Series till mid-Sep or end-Sept, but it is not without worry. Or alternately you will not be able to complete the entire syllabus and will have to take a go at the exam having prepared only selected topics. This is very common with many aspirants. Please read the last paragraph of the answer to the immediate above question for more details.

Target completion of syllabus i.e. learning concepts and practicing questions on the ideas learnt, latest by July-end, preferably by June-end.

Since QA is very theory intensive (as compared to VARC and DILR which have very little theory and are mostly skill based), spend about 60% of your time on QA, 20% on VARC and 20% on DILR till completion of QA syllabus. 

QA needs to be done daily and you can learn/practice VARC and DILR on alternate days. E.g. on day 1, spend 2-2.5 hours on QA and 45 mins on VARC; on day 2 spend 2-2.5 hours on QA and 45 mins on DILR.

DO NOT NEGLECT VARC OR DILR SECTIONS.
It is not a great idea to finish QA entirely and then turn your focus on DILR and VARC. In fact, we strongly suggest not having any extended period of time, not even a week, where you dont spend atleast 25-30% time on DILR and VARC. Even if the theory lessons of VARC and DILR are done with, you can start practicing sets of RC and DILR.

Ideally dont study QA for long hours at a stretch. So if you are studying QA for say 2 hours per day, divide it into 2 slots of 60 mins. You can cover some lessons of DILR or VARC in-between the slots.

After completion of syllabus, from July/August to November, spend almost equitable hours on each of the three sections, or say 40% on QA, 30% onVARC and 30% on DILR. During this time period, you need to ….

1. Solve Past Year Questions (PYQs) if not already done while completing the syllabus
2. Test Series – Starting with section tests, zero down on a test taking strategy that works best for you, and build up the frequency till 3 full length mocks per week. And not to forget analyse each test in details. One full length mock will mean 2 hours of test taking and another 4 hours of post-test-analysis-and-solving. 
3. Strengthen the weak areas as identified from the mocks.
4. Practice tougher questions i.e. those which go beyond just testing direct concepts and require higher degree of problem-solving techniques, or are multi-concept based. Ideally practice a mixed-topic-question-collection.

There is no prescribed order in which topics need to be covered and one can also follow one’s interest or instinct. If you dont have any particular choice and …..

… if starting your prep before or in March, cover the topics in the order: Number Systems → Arithmetic → Algebra → Geometry → Rest of QA or Modern Maths

… if starting in April or later, cover the topics in the order: Arithmetic → Algebra → Geometry → Number Systems → Rest of QA or Modern Maths

The logic behind this is that weightage is maximum for Arithmetic, followed by Algebra and then followed by Geometry.

Number System has less weightage, but most text-books start with this topic, so if you are starting early, you may as well do this at start. And if starting late, focus on heavy weightage topics first.

Modern Maths (P&C, Probability, Set Theory) has the least weightage.

Within Arithmetic, the order of lessons will be: Ratio Proportion → Variation → Percentages → Profit Loss Discount → Simple & Compound Interest → Averages → Weighted Average & Alligation → Mixtures → Time Speed Distance → Time and Work. 

Some minor variations are fine e.g. TSD and Work can be switched.

Arithmetic has the highest weightage – around 8 to 9 questions out of 22. But Arithmetic is a collection of atleast 4 groups – Ratio Proportion Variation; Percentages and its application in Profit Loss Discount and Simple & Compound Interest; Averages, Weighted Averages/Alligation and its application in Mixtures; Time Speed Distance & Work; Individually these groups will have just upto 2/3 questions per group.

Algebra has the next highest weightage – around 6 to 7 questions out of 22. Algebra, again is very vast, if the sub-topics are considered – Equations (Linear – Simultaneous and Diophantine, Quadratic, etc), Inequalities, Modulus, Maximum-Minimum, Functions, Series.

Geometry follows next, with 3-4 questions. Topics of Geometry are Triangles, Quadrilaterals, Polygons, Circles, Solids, Mensuration, Co-ordinate, and very very rare, once in decade type, a question on basic Trigonometry.

Number Systems has around 2-3 questions. Topics of interest are Classification, Divisibility, Cyclicality of unit and last 2 digits of n^p, Factorials, number of Factors and its applications, LCM/HCF and its application, Remainders.

There could be 1 or max 2 questions on Permutation Combination. Probability is mostly not asked (but has high weightage in NMAT, SNAP and is also tested in XAT). Set Theory also has to be done – even though there aren’t questions on it every year, but every 2nd or 3rd year it is tested and it is tested regularly in DILR.

Blunt answer – students toying with this question are underestimating the competitiveness of CAT. Entertaining this question is like accepting that you are aiming only for the 95 -96 %tile range and not for the 99+ %tile range. 

If you are aiming for the old IIMS or the likes of FMS, SP Jain, MDI, IIFT, IIT B/D, then you CANNOT skip any topic.

And if you are aiming for just an average performance, then the must-do topics are Arithmetic, most of Algebra and most of Geometry. Rest you may skip at your own risk. The more you skip, the lesser will be the chances of doing well in exam. And even if you are skipping some topics, we would strongly advise you to not skip the entire topic, and atleast cover the basics of ALL topics.  

Short answer – There are no books that suffice as a comprehensive prep for CAT QA. You will need a complete course including test series (i.e. courses of IMS, CL, TIME or the smaller & newer ones e.g. Cracku, Elitesgrid, 2IIM, ourselves – Takshzila, etc) to be able to compete at the level of competition that CAT has.

No – Arun Sharma book is not enough, in my opinion. In fact, again in my opinion, it is not a good book to start with. Nothing personal against the book, mentioning it here just because it is widely popular and many students will be having a question “is Arun Sharma book enough?”

Essentially you will need to cover all the concepts in multiple rounds and not try to learn a topic from basics right to the most advanced level in one single flow. So, in round 1, you will cover only the basics, but of all topics – Numbers, Arithmetic, Algebra, Geometry, Rest of Quant – every topic but only the basics. The basics here will refer to those concepts that are taught at the start of any of the topics; and the questions to be practiced are those which directly follow the theory learnt or are formulaic in approach. It is unfortunate that no CAT material has categorised the topics into such rounds-friendly-learning pedagogy. So it’s best to rely on text-books and post that on easy QA books like R S Agarwal. NCERT text-books (class VII to X) are good for building basics. But these need to be done super-fast e.g. one entire class in 1 week, finish all of VII to X within 1 month. Next, 1 month finish R S Agarwal QA or any book on QA for competitive exams of banking or govt exams.

Round 2 will be to start with a CAT course but cover only those ideas/concepts in the course that you find comfortable. If there is a difficult advanced concept (can be identified by its complexity and will usually be later in the flow of ideas), you can skip this but do make a note of the lessons/ideas/question-types that you are skipping. Thus, round 2 will be to graduate from basics to intermediate stuff. Here you will practice questions that are not the same as those learnt and will need to get used to finding your way around unseen questions though based on the ideas learnt.   

A very important part of your learning will be clearing doubts. Every small step you are unable to understand or a question that you are unable to solve needs to be discussed with a fellow-aspirant or faculty. And dont just ask a faculty for the solution, instead discuss the problem … what did you understand from the question, what did you try, where are you stuck … seek clues/hints and not complete solution, finish completing the question yourself.

Keep in mind that understanding is the key, not memorising. So dont chase formulae, short-cuts or ‘mug’ question-types. Always understand, it’s good to use short-cuts but only when you understand what the short-cut is doing behind its form.

Finally, round 3 will be those concepts/question-types that you skipped. 

Remember there are no easy way out, there are no short-cuts. A student who is good at QA has got to be good only by having gone through these rounds over the years. You need to do it in months. So dont search for any magic pill or some secret way to study, instead start studying the way this answer proffers.

When learning a topic, one needs to solve around 50-100 questions depending on the vastness of the topic. Topic here means a closely bound theory e.g. Ratio Proportion Variation is one topic, Linear Equations is another topic which will include Simultaneous Linear Equations and also Linear Diophantine equations, Quadratic is another topic; Triangle will be one vast topic, Quadrilaterals/Polygons or Circles, each, is a very short topic. With 20+ topics in total, this will be around 1000+ questions.

Having learnt the topics, one will have to practice a further 200 questions in each of Numbers, Arithmetic, Algebra, Geometry and Rest of QA. These 1000 questions will be a little more difficult that direct theory based and will require application and problem-solving skills. 

All the above 1000+1000=2000 questions need to be done by mid-Aug, before Test Series. 

And in the last 3 months, one needs to practice another 1000+ questions, bringing the total to 3000.

Keep in mind that Quantity will never compensate for Quality. So, solving these 3000 questions very well, spending quality time on them is much better than solving 5000 questions in a rush only to feel like one does not remember a lot in the end. 

Two parts to this answer … one is about the content to be used, this is simple and is replied to in the next FAQ; more important is the expectation or the mind-set, which is answered here.

LR is not like QA where you need to learn some theory/concepts and the questions that follow will be based dependent on the confines of what is learnt. LR has practically no theory to learn. Infact, LR has no ‘topics’. The names used to categorise sets into topics is only for our communication, but two “arrangement” sets can be very different in the reasoning involved – and may not have anything to do with what is commonly understood as “arrangement” … it may be based on mathematical reasoning, or on true-false reasoning, or on visual/spatial reasoning. Further many sets cannot even be categorised into a ‘type’ easily.

So as a start to preparing for LR, accept that there is no theory and each set is solved based on the conditions specific to the set. You cannot ‘prepare’ i.e. learn some conditions/situations and paste them in every set. Just like you cannot pre-learn specifics that a RC passage is about. You comprehend the given information then and there.

Next, is accepting that irrespective of however many sets you practice, the sets asked in further exercises/mocks/CAT will be ‘new’ … a new situation, new conditions, new possibilities. So practicing with an aim to be familiar of what is asked and to reproduce the steps practiced, will not help.

Also, in most sets you try, you will hit a “wall” i.e. reach a stage where you are clueless what to do next. This is what makes a puzzle a puzzle, if it would not confound you, then there is no ‘reasoning’ being tested.

With your expectation and mindset set right, next is to focus on the practicing/solving the set in-hand. Start with knowing that the ‘solution’ is hidden in the clues given. No outside knowledge other than comprehending the clues is needed.

Ponder on how best to represent the given information. If you start with one representation but get stuck and can think of another representation, dont hesitate to start all over again … solving a set is like finding your way through a maze, you may need many trials.

When stuck, go over all the clues again, over and over if needed, analysing the implication of each word, phrase. Words like ‘only’, ‘exact’, ‘atleast’, ‘distinct’, etc all are relevant, you cannot miss any. Try using the word/clue that is still not used in its full extent. With practice you will identify the clue that is underutilised or is the key to the impasse.

It may help if cases are made by assuming some aspect. Assuming takes the solution one step further and it may help in eliminating or confirming the case, but there needs to be a framework to assume such that all cases are exhausted.

 If no head-way is made, take help of explanation, but only to proceed ahead, to find what link you are missing, and then complete the set on your own.

Finally, having completed a set, dont forget to cement your learning .. what was unique about the set, what new situation/condition did you get familiar with, how was the “wall” overcome, what were you missing, how did the clues interact, spend time taking an overview of the set and the process to solve it.

Essentially keep trying and practicing even when solving sets seems elusive. The more you practice, the deeper and wider will your reasoning prowess improve. Your concentration skill, ability to hold on to complex and long chain of reasoning (if-then-if-then links), even using creative ways to find the solution, etc, all will improve.

Ideally you should have access to one DILR course. The usual choices are the national level traditional players – IMS, CL, TIME; and the newer ones e.g. Cracku, ElitesGrid, 2IIM, ourselves (Takshzila) and there are many many more. A course from any of these will help you get acquainted with all the common varieties and will also have a smoother learning curve, starting with basic sets and moving on to more complex sets. Try winding any one course first and latest by June.

Having done this, you can next practice from Past Year CAT papers. From 2017 to 2025, there are a total of 126 sets to practice. And you can also get another 24+ sets from CAT papers prior to 2017, those sets that are relevant even today. So these are 150 sets, giving you a real taste of CAT flavour. Cover these sets in July-August.

From Sep to Nov, along with taking mocks and practicing DILR from them, you may also solve sets from assorted free sources on the internet (given below). Most of these are chosen sets from past mock papers of IMS, CL, TIME or modelled on them. Being chosen ones, they are the tougher ones … who will want to put solving an easy set on social media, right? So dont be too disappointed on not getting them, instead use them to learn.

Aptitude Jab YouTube Channel – 450+ sets
Elites Grid YouTube channel – 150+ sets
Rodha YouTube channel – 60+ sets
LRDI Masters Facebook Group – 50+ sets
Topmytest YouTube Channel – 100+ sets
Anastasis – Shankar YouTube channel – 250+ sets (This is a collection of tough sets and is to be done only when you are well practiced and want to try tougher ones)

All the above mentioned sets – one course, past year papers, free internet resources – add up to more than 1000 sets. You will not be able to complete them. So stop searching for more material and instead finish these.

DI has some QA in it e.g. growth rates as an application of percentages, successive percentage changes, market share – value or volume wise and its inter-linkages, measures of productivity or utilisation and other measures of ratios, usage of weighted average in some rare situations – but the theory is very minuscule and can be easily adapted once it is learnt in QA. So start DI only after having done Ratios, Percentages, Weighted Average from Arithmetic. 

After learning the theory, it is just a matter of practicing sets, figuring out that whether it is tables or graphs like line, bar, scatter or pie charts, essentially they are just data points and the question types are also very restricted, so it just boils down to understating the data, the linkages between them, comprehending the question, drawing information from the data and calculation. Try to pay attention to these above points when you practice sets. Dont categorise the sets as Line or Bar or Column or Pie … any Line graph can also be converted to a Bar or a Column, even a Pie-chart data can be given as Bar/Column. Instead focus on the data given e.g. is it a time-series data, what are the relation between the terms used, distinguish between volume & value, etc. If two graphs are given, find the common term between them to move from one to the other. Focus on what sort of questions are asked for each data type. Very soon you should realise that there are a limited types, just they may appear different visually.

The theory, solved examples and practice set in any DILR course should suffice. Solving about 50 odd sets is good enough practice to start your Test Series. Thereafter it is just keeping the momentum intact by practicing sets regularly.

In getting familiar with the common-types of sets asked i.e. in completing a DILR course, one would solve about 150 LR sets and about 50 DI sets. As advised elsewhere, first finish any one course and latest by June-end.

In August, solve the 150 DILR sets from Past Year CATs (PYQs).

And from September to November, solve another around 250 sets from current year mocks, past year mocks, assorted free resources on the net. 

This will total to about 600 sets, which will count as a good prep.

If you have the time, then absolutely YES.

Limiting yourself to CAT related material will restrict your practicing to just CAT-alike-sets. But reasoning skills can develop with more flourish when faced with unstructured puzzles. It is just like … in addition to regular working out in a gym, if one also adds yoga practice or calisthenics, it will help in an overall well-being related to health itself. Similarly, reasoning skills developed while solving puzzles will surely help in tackling CAT-like sets as well.

Sudoku is just one specific kind of a puzzle. Yes, there are grid-arrangement sets which have conditions very similar to those in Sudoku e.g. each column has exactly one tick and each row has exactly one tick.

However these sets will be counted as easy ones and you will also learn solving such a set in any course. Further, it is just one of the situations from amongst an almost infinite possibilities of situations that can be asked. So solve Sudoku, but also quickly go beyond it, and dont just stay there thinking this is very important. It is a very small part of your prep.

You don’t have to do a Hard or Expert level of Sudoku, the techniques needed are too technical, too specific and limited to just Sudoku. Once the basic ideas are learnt, move ahead, rather than stick to just this one type, solve more varied puzzles. 

Yes, there are. But they are general purpose reasoning books for recreation, which does help in building reasoning prowess. Just that they dont have sets exactly like CAT, hence one may think one is wasting time. However, as explained in FAQ “Should I solve puzzles”, reasoning skills build in areas beyond CAT-prep will also definitely help indirectly in CAT. Some popular books are:

The Great Book of Puzzles & Teasers – George Summers. The author has 15+ books on puzzles and teasers.

What is the name of this book – Raymond Smullyan. The author also has far too many books to list here.

The Mensa Genius Quiz-A-Day Book

Books by author Martin Gardener

Again dont go overboard, keep your time availability in mind. Hardly anyone who has cleared CAT would have gone through them. Read them if you are interested in puzzles, have lots of time and not just for direct CAT prep.