Logical Reasoning and Data Interpretation (DILR) for CAT: The Complete Guide

Read the setup of all 4-5 sets before committing to any of them. You're not solving yet — you're judging: how much data is there, how many constraints, does the question language feel familiar from your practice, and does the set "click" on a first read or feel murky.

Don't pick sets based on liking DI more than LR, or vice versa — pick based on whether the specific set in front of you starts forming a clear structure in your head within the first 30-45 seconds of reading it. A DI set that doesn't click is worse than an LR set that does, regardless of your general topic preference.

Divide your remaining ~35 minutes across your chosen sets (roughly 10-12 minutes each), and treat that as a hard boundary. If you're not seeing progress by your cap, that's your signal to reassess — not necessarily abandon, but check whether you're actually stuck or just being thorough.

Most sets have 4-5 linked questions of varying difficulty. It's common to be able to answer 2-3 of them confidently even without fully cracking every constraint — bank those before spending remaining time pushing for the hardest question in the set.

Since TITA questions carry zero penalty for a wrong answer, any TITA question where you've narrowed the range through partial logic is worth attempting, even without full certainty — this is free expected value that MCQ negative marking doesn't allow elsewhere in the section.

Setup: A table shows the annual revenue (in ₹ crore) of five companies (P, Q, R, S, T) across six years. Questions ask for: (a) the company with the highest percentage growth over the period, (b) the average revenue growth rate, and (c) the company with the most stable (least volatile) revenue trend.

Approach:

1. Scan the table for 15-20 seconds to spot obvious outliers — a company with a clear steady climb or an erratic zigzag pattern often answers the "most stable" or "least stable" question without full calculation.
2. For percentage growth, approximate rather than calculate precisely — round revenue figures to the nearest 50 or 100 before computing growth, since exact precision rarely changes which company wins a "highest growth" comparison.
3. Compute only what each specific question needs — don't calculate every company's exact growth rate if the question only asks you to identify the single highest performer.
4. Cross-check your outlier-based intuition from Step 1 against your Step 3 calculation before finalizing.

Setup: Seven friends (A through G) sit in a row facing north. A sits third from the left. B sits immediately to the right of A. Two people sit between B and C. D does not sit at either end. E sits immediately left of F. G sits at one end.

Approach:

1. Draw seven blank positions immediately — don't attempt to hold the arrangement mentally.
2. Apply the most specific, position-fixing clue first: "A sits third from the left" fixes a definite position immediately — always start here rather than with relative clues like "immediately left of."
3. Layer in directly dependent clues next: B's position follows immediately from A's fixed position.
4. Use elimination for the remaining, less-fixed clues (D not at either end; G at one end) to narrow the remaining open seats.
5. Place E and F last, since "immediately left of" clues are easiest to slot into whatever open, adjacent positions remain once the more restrictive clues are locked in.

The general principle: always sequence your clue application from most restrictive (fixes an exact position) to least restrictive (relative/relational) — solving in the wrong order is the single biggest time-waster in arrangement puzzles.

Setup: Four teams (W, X, Y, Z) play each other exactly once in a round-robin tournament. A win earns 3 points, a draw 1 point each, and a loss 0. After all matches, W has 7 points, X has 5 points, Y has 3 points, and Z has 1 point. Questions ask you to determine the exact result of each individual match.

Approach:

1. List all matches first: with 4 teams playing each other once, there are exactly 6 matches. Write them out as pairs before touching the points.
2. Work from the most constrained team's score. A score like 7 (from 3 matches) can only come from specific win/draw/loss combinations — list the possible combinations (e.g., 2 wins + 1 draw = 7) before assigning them to specific opponents.
3. Use total points as a cross-check: the sum of all teams' points should match the total points distributed across 6 matches (accounting for 3 points per decisive match, 2 total points per draw) — this catches arithmetic errors early.
4. Resolve the most constrained team's exact results first, then use elimination on the remaining matches, since each resolved match removes possibilities for the two teams involved.

Games & Tournaments sets reward writing out every match explicitly rather than trying to reason about scores in the abstract — the visual match list is what makes the constraints tractable.

Setup: A paragraph describes a retail store's monthly sales across three product categories over a quarter, embedding figures in sentences rather than a table (e.g., "Category A's sales grew by 15% in February over January, while Category B's sales in March were double that of Category A in January...").

Approach:

1. Before attempting any question, extract the paragraph's numbers into a small table of your own — this single step is what separates fast caselet solvers from slow ones, since re-reading the paragraph for every question wastes enormous time.
2. Assign clear variable names or short labels for any figure not given as an absolute number (e.g., "A_Jan = x" if only relative figures are given), and build simple relationships between them as you read.
3. Solve for any one absolute value if given, then use the paragraph's relative statements to derive the rest — caselets are often solvable as a small system of linear relationships once extracted.
4. Answer questions directly from your extracted table rather than returning to the original paragraph each time.

Caselets often look intimidating purely because of their format — once extracted into a table, they frequently turn out to be simpler than an equivalent chart-based DI set.

20 questions (typically 14 MCQ and 6 TITA) for 60 marks, out of CAT's overall 204 total marks.

Yes, for MCQs: +3 for a correct answer, −1 for an incorrect one. TITA questions carry no negative marking — +3 for correct, 0 for incorrect or unattempted.

Roughly 11-13 questions at 80%+ accuracy, translating to around 30-34 marks out of 60 — this is directional and varies with each year's overall difficulty. Solving 2 sets thoroughly and accurately is a viable strategy, since DILR typically has lower average accuracy across the candidate pool than VARC or QA.

Typically 4-5 sets, each containing roughly 4-5 linked questions, though the exact split can vary by year and slot.

No. CAT enforces a strict 40-minute sectional lock in a fixed order (VARC, then DILR, then QA) — you cannot move to another section early, and unused time in one section doesn't carry over.

Focus on fewer sets deeply. With only 40 minutes for 4-5 sets, most high scorers attempt 2-3 sets thoroughly rather than spreading thin attempts across every set on the paper.

Committing to a set based on topic preference rather than how quickly it "clicks" on first read — and staying with a set too long out of sunk-cost thinking rather than reassessing against a time cap.