How to Prepare for CAT DILR

What CAT DILR tests

Data Interpretation and Logical Reasoning (DILR, also written LRDI) is the second section of CAT. Like the other two, it runs for 40 minutes and you cannot switch into or out of it. It is the section aspirants most often call unpredictable, and the one where a few good decisions in the first ten minutes make the biggest difference to your score.

DILR has no formulas to learn and almost no syllabus in the usual sense. Every question comes in a set: a paragraph of conditions, a table or a chart, followed by several questions that all depend on working out the same underlying structure. If you crack the set, you usually get most of its questions. If you do not, you usually get none. That all-or-nothing nature is what makes DILR scores swing so much in early mocks.

What the section really tests is three things: organising messy information into a grid or table, making deductions one careful step at a time, and judging quickly which sets are worth your time.

The chapters that matter for CAT

Recent CAT sets rarely fit one neat label, and many blend logic with arithmetic. Still, almost all of them are built from these families, each with past CAT questions on the site:

• Tables & Graphs: bar, line, pie, scatter and radar charts, and tables with missing values to be filled using totals or constraints.
• Arrangements: linear, circular and multi-attribute arrangements (who sits where, who owns what).
• Selection & Distribution: choosing teams or committees, allocating items or people under conditions.
• Games & Tournaments: round-robin and knockout formats, points tables, who must have beaten whom.
• Routes & Networks: flows through pipes, roads or networks, shortest paths, capacities.
• Venn Diagram: overlapping groups, usually with three or four sets and some unknown regions.
• Mathematical Reasoning: number-based puzzles where arithmetic and logic combine, such as scores, votes, ratings or averages that must be reverse-engineered.
• Puzzles and Binary Logic (truth-tellers and liars).

Shorter topics such as Blood Relations, Syllogisms, Coding-Decoding, Series, Directions, Cubes, Calendar and Clocks are not asked as standalone questions in recent CAT papers. They appear in other exams and are useful warm-ups, but do not spend weeks on them for CAT. The full list is in the CAT syllabus guide.

How DILR questions look today

From 2021 to 2023 the section had 20 questions, typically four sets of five. The 2025 paper had 22 questions in each slot, and set sizes have varied. Expect four or five sets, some MCQs and some TITA (type-in-the-answer) questions. A correct answer is +3, a wrong MCQ is −1 and a wrong TITA is 0. The CAT exam pattern guide has the year-by-year numbers.

A few patterns have held across recent papers:

• Hybrid sets. Pure "seating arrangement" or pure "read the bar chart" sets are less common. A typical set gives partial data in a table plus several verbal conditions, and you must deduce the missing values before answering anything.
• Reasonable calculation, heavy logic. Numbers are usually small. The difficulty lies in tracking cases, not in arithmetic. The on-screen calculator helps only with the few heavier DI calculations.
• Uneven difficulty. In most slots one or two sets are clearly more approachable than the rest. Finding them is the core skill.
• Questions that add conditions. Later questions in a set often say "If, additionally, X is true, then…". These can be quick once the base grid is built.

Open CAT 2025 LRDI Slot 1 and CAT 2023 LRDI Slot 1 and simply read the sets without solving them. You will see the style much faster than any description can show you.

A preparation plan

DILR rewards regular, spaced practice far more than long weekend sessions. Two or three sets a day for five months will do more than twelve sets every Sunday. The plan below assumes about six months.

Phase 1 (month 1): learn the set families

• Spend a few days on each core chapter above. Solve its easier questions first to learn the standard representation: a grid for arrangements, a points table for tournaments, a three-circle diagram for Venn sets, a flow table for networks.
• Do not time yourself yet. Write out every deduction. The goal is clean method.
• Learn to read every chart type quickly, especially radar, stacked bar and scatter plots, and to estimate percentages and ratios mentally.

Phase 2 (months 2–4): CAT-level sets, daily

• Solve two to three CAT-level sets a day, mostly past CAT sets from 2017 onwards, chapter by chapter, with a soft target of 12–15 minutes per set.
• Mix LR and DI. Preparing only the type you like is risky; a paper heavy on your weaker type can sink the section.
• After each set, read the solution even if you got everything right. Often the solution's way of setting up the table is cleaner than yours, and that saves minutes later.
• Start a weekly 40-minute DILR sectional. A recent CAT DILR paper is ideal; you can also make one with the test builder.

Phase 3 (months 5–6): selection and mocks

• Take two or three full sectionals a week plus your full-length mocks.
• Practise set selection as a separate skill: open a paper, give yourself four minutes to rank all sets by expected difficulty, then solve and compare your ranking with how things actually went.
• Keep the most recent CAT papers for this phase and take them as real timed tests.

A sample week in Phase 2

How to solve a set

1. Read the whole set once, including the questions. Questions tell you what the examiner thinks is determinable and often hint at the structure.
2. Choose a representation before writing anything. Most CAT sets fit a grid with entities as rows and attributes or time periods as columns.
3. Fill direct information first, then apply the most restrictive conditions ("exactly one", "no two", totals that must match).
4. Branch when stuck. If a condition allows two possibilities, draw both side by side and push each forward. One usually breaks quickly.
5. Answer what you can as soon as you can. You may not need the complete grid to answer two of the questions.

A worked mini-example

Four friends, A, B, C and D, each scored a different whole number of points from 1 to 4 in a quiz. A scored more than C. B did not score 4. D scored exactly one point more than B. Who scored 4?

B is not 4, and D = B + 1, so B is 1, 2 or 3 and D is 2, 3 or 4. A > C rules out C = 4. If D is not 4, then 4 must be A (since B and C cannot be). Try each: B = 1, D = 2 leaves 3 and 4 for A and C, so A = 4, C = 3. B = 2, D = 3 leaves 1 and 4, so A = 4, C = 1. B = 3, D = 4 leaves 1 and 2, so A = 2, C = 1. So "who scored 4" cannot be determined uniquely: it is A in two cases and D in one. Recognising "cannot be determined" quickly, by listing the few cases in a table, is exactly the habit CAT tests. A real set would add a further condition, for instance "C scored more than B", which fixes B = 1, D = 2, C = 3, A = 4.

Practising with past papers and topic pages

Good DILR practice material is scarce. Many books still focus on the short standalone questions that CAT no longer asks. Past CAT sets are the closest match to the real exam, and the site has CAT papers from 1991 to 2025 with step-by-step solutions on the CAT previous-year papers page.

• Use 2017–2022 sets topic-wise through the topic-wise preparation pages during Phase 2.
• Keep 2023–2025 for timed sectionals in Phase 3, for example CAT 2024 LRDI Slot 2.
• Do not neglect older papers. Sets from the late 1990s and 2000s are shorter but excellent for learning representation, and many are tougher than they look.
• XAT and other exams include decision-based sets as well; the XAT decision making guide covers that format.

How to analyse a DILR sectional

Score alone tells you little in DILR. After each test, ask three questions for every set:

1. Was my selection right? Did I skip a set that turned out to be easy, or sink time into one that was hard?
2. Where did I lose time? Setting up, a wrong branch, or one stubborn question?
3. Could I solve it now, untimed? If yes, the problem was selection or nerves; if no, it is a skill gap. Add it to your mistake notebook (sign in to save questions) and redo it a week later.

Expect your DILR scores to swing widely in the first ten or fifteen mocks. That is normal for a set-based section. If your set selection is improving, the swings narrow.

Test-day strategy for the 40 minutes

You will not solve every set, and you do not need to. In most years a good score comes from solving two sets fully and picking up a few questions from a third. Plan for that.

• Scan first (3–5 minutes). Look at every set's length, the number of variables, how many conditions are given and whether the questions look direct. Rank them. Signs of an approachable set: few entities, conditions that fix values directly, questions without long "which of the following must be true" option lists.
• Solve your best set first and bank its marks before touching the next.
• Use a cut-off. If you are five or six minutes into a set and still cannot fill the base grid, leave it. Coming back later with fresh eyes often works; staying rarely does.
• Leave questions, not just sets. If you have three questions of a set done and the fourth needs another eight minutes of case work, move on.
• Use TITA questions. Wrong TITA answers cost nothing, so if a set's TITA can be answered from your partial grid, answer it.
• Do not carry VARC into DILR. Whatever happened in the first 40 minutes is over. Begin the scan calmly.

When the section ends, Quant starts at once; the CAT Quant preparation guide covers that section.

Common mistakes

• Starting with set one by default. The order of sets on screen has nothing to do with their difficulty.
• Sinking 20 minutes into one set. Half the section on a set that may yield nothing is the single most expensive habit in DILR.
• Practising only standalone puzzles. They build basics but do not prepare you for CAT-style hybrid sets.
• Solving in your head. Write the grid. Mental deduction leads to small errors that wreck a whole set.
• Skipping analysis. Without reviewing selection decisions, you repeat the same mistakes in every mock.
• Ignoring DI because "it is just calculation". Recent DI sets need deduction too, and a DI-heavy paper can catch out LR-only preparation.

To fit DILR practice alongside VARC and Quant in a weekly timetable, see the CAT preparation strategy guide, and keep a daily set going with the question of the day.