DILR Set Selection: Which Sets to Attempt, Which to Skip in 40 Minutes

Why DILR Set Selection Decides Your CAT Score More Than DILR Skill Does

Most CAT aspirants preparing for CAT 2026 treat DILR (Data Interpretation and Logical Reasoning) as a pure skill problem: solve more sets, get faster, improve. That is necessary but not sufficient. DILR is a 40-minute, fixed-timer section with 5 sets and no ability to borrow time from VARC or QA. Inside that block, the single biggest score driver is not raw solving speed. It is set selection: correctly judging, within the first few minutes, which sets are worth your time and which will trap you.

A candidate who solves 3 sets cleanly with high accuracy will consistently outscore a candidate who attempts all 5 sets but gets stuck on 2 of them and finishes with partial, error-filled work across the board. This is the core premise of every effective CAT DILR strategy, and it is what this guide builds around: a structured scan-and-rank method, clear difficulty markers you can spot in under a minute, and a worked example showing the decision in practice. If you have been collecting LRDI CAT tips from forums and mock analyses, this is the one that ties most of them together 

The 5-Minute Scan Rule

Before solving anything, spend the first 5 minutes of the DILR section reading all 5 sets, question stems included, without attempting calculations. This is not wasted time. It is the single highest-leverage 5 minutes of the section, because it converts DILR set selection from a gut reaction into a ranked decision.

During the scan, ask three questions of each set:

• Can I see the structure immediately? (A table that fills in directly from the given data, a seating arrangement with unambiguous constraints, or a graph with a clear trend are good signs.)
• Does the data given actually determine a unique answer, or does the set require multiple cases and branching logic?
• How many of the individual questions look answerable even if I cannot fully crack the set (partial credit potential)?

Rank the 5 sets from most workable to least, based on this scan. Then attempt strictly in that order. This single habit, practiced consistently as part of DILR preparation in the weeks before CAT 2026, is what separates aspirants who finish DILR with 3 confident sets from those who finish with 5 half-attempted ones.

Set Difficulty Markers: What to Look For in Under a Minute

Not every DILR set announces its difficulty clearly, but most give away signals within the first 60 to 90 seconds. Use these markers during your scan:

Signal	Likely Workable	Likely Skip
Data structure	Direct table or single graph with clear axes	Multiple interlinked tables or graphs requiring cross-referencing
Constraints (for LR/arrangement sets)	3 to 5 clear, non-contradictory conditions	6+ conditions, several conditional (if-then) clauses
Question phrasing	Direct lookups, single-step calculations	Questions requiring you to first fully solve the set before any question is answerable
Case branching	One clear arrangement or scenario	Multiple valid cases that must each be checked
First-question difficulty	You can answer Q1 within 90 seconds of finishing the read	Q1 itself requires solving most of the set first

None of these markers are absolute. A set with several constraints can still be fast if the constraints are non-contradictory and lock in quickly. The point of DILR set selection is not a rigid checklist; it is building the pattern recognition to make this call in under a minute, which only comes from timed practice, not from reading about it.

Worked Example: Ranking 5 Sets in Real Time

Here is how the scan-and-rank process plays out in practice, using a representative CAT-style DILR paper as illustration:

Set	Type	60-Second Read Verdict	Decision
Set 1	DI: Single bar graph, 4 questions	Direct data lookups, one calculation step per question	Attempt first
Set 2	LR: Seating arrangement, 5 conditions, no if-then clauses	Constraints look linear and non-contradictory	Attempt second
Set 3	DI: Two linked tables, cross-referencing required	Q1 needs both tables fully mapped before answering	Attempt third, time-box to 10 minutes
Set 4	LR: Conditional puzzle with 3 if-then branches	Multiple valid cases likely; first question depends on full resolution	Skip unless time remains
Set 5	DI: Data sufficiency-style set with ambiguous units	Unclear whether data given is sufficient without deeper reading	Skip

In this example, a candidate following the ranking would spend roughly 10 minutes each on Sets 1 and 2 (secured, high-accuracy attempts), 10 minutes on Set 3 with a hard stop even if incomplete, and use the remaining 10 minutes to attempt 1 or 2 individually answerable questions from Set 4 or 5 rather than starting either set cold. This yields 2 fully solved sets, 1 partially solved set, and a few opportunistic points elsewhere, a stronger outcome than an unranked attempt at all 5.

Time-Boxing: The Discipline That Makes Set Selection Work

Set selection only works if paired with hard time limits per set. Without a time-box, even a correctly ranked “easy” set can quietly consume 15 to 18 minutes if you chase a full solution instead of securing partial marks and moving on.

A practical time-boxing structure for the 40-minute DILR section, built around 3 target sets:

• Minutes 0 to 5: Scan and rank all 5 sets (no solving).
• Minutes 5 to 15: Attempt your top-ranked set fully.
• Minutes 15 to 25: Attempt your second-ranked set fully.
• Minutes 25 to 35: Attempt your third-ranked set, with a hard stop at minute 35 regardless of completion.
• Minutes 35 to 40: Revisit any skipped, individually answerable questions from lower-ranked sets; do not start a fresh unattempted set.

This structure is a starting template, not a fixed rule. As DILR preparation progresses, most aspirants adjust the time-box based on their own solving speed on DI-heavy versus LR-heavy sets. What should not change is the discipline of a hard stop: walking away from a set that is not yielding progress, even mid-solution, protects your overall attempt count far more than finishing what you started.

Common DILR Set Selection Mistakes

Most LRDI CAT tips focus on solving faster. The mistakes below are about choosing badly, and they cost more marks than slow solving does. 

• Starting with the first set on the page instead of scanning all 5 first, simply because it appears first.
• Sunk-cost solving: continuing on a set past the time-box because “I’m already 80% through it,” when that 80% has taken 20 minutes.
• Judging difficulty by set type alone (assuming all LR sets are harder than all DI sets, or vice versa) instead of by the specific set’s structure.
• Skipping the scan entirely under exam pressure, which is precisely when a structured process matters most.
• Not practicing set selection as its own skill separately from solving speed during DILR preparation, and only realizing the gap on exam day.

A note on CAT 2026 timing:

With CAT 2026 scheduled for 29 November 2026 and registration having closed 22 September 2026, DILR set selection is a skill best built now, through regular timed practice, rather than left to the final weeks. Practicing the 5-minute scan and time-boxing structure under real section-timer conditions, not just untimed set-solving, is what converts this framework into an exam-day habit.

Confirm CAT 2026 exam-day details and section timing directly at the official IIM CAT site: https://iimcat.ac.in

COMPARISON TABLE: DILR SET TYPES AT A GLANCE

Set Type	Typical Time Cost (if workable)	Typical Time Cost (if trap)	Best Indicator
Direct DI (single table/graph)	6 to 9 min	N/A, rarely a trap	Data maps to questions with 1-step calculations
Linked/multi-source DI	8 to 12 min	15+ min	Whether Q1 needs full cross-referencing first
Linear arrangement LR	7 to 10 min	14+ min	Number and clarity of constraints
Conditional/branching LR	9 to 12 min	16+ min	Presence of if-then clauses creating multiple cases
Data sufficiency-style DI	6 to 8 min	12+ min	Ambiguity in whether given data is actually sufficient

FAQs

How many sets are in CAT DILR?
CAT’s DILR section typically has 5 sets (a mix of DI and LR sets) with roughly 4 questions each, totaling around 20 questions, to be solved in 40 minutes. The exact split between DI-heavy and LR-heavy sets can vary slightly year to year, so treat this as a structural expectation, not a guarantee, and confirm the final pattern on exam day instructions.

What is a good DILR attempt count for a 99 percentile?
Based on patterns reported by aspirants and coaching analyses (not an official IIM figure), attempting 3 full sets with high accuracy (12 to 16 correct questions) has historically placed candidates in the 99+ percentile range in DILR. This is a directional benchmark from third-party sources, not a confirmed cutoff, since IIMs do not publish an official attempts-to-percentile table.

Should I attempt the easiest set first or read all sets before starting?
Use the 5-minute scan described in this guide: skim all 5 sets first, rank them by clarity of the underlying logic, then attempt in that ranked order starting with the clearest set. Reading everything first costs 3 to 5 minutes but prevents you from getting trapped in a hard set with no fallback, which typically costs far more time than the scan itself.

Is it okay to leave a set incomplete in DILR?
Yes. A half-solved set with 2 of 4 questions correct is a better outcome than a fully-attempted set with 1 of 4 correct after burning 18 minutes on it. Partial attempts on a workable set, backed by process-of-elimination on the remaining questions, are a standard and accepted strategy, not a compromise.

How is DILR different from LR (Logical Reasoning) in other exams?
In CAT, DI (Data Interpretation) and LR (Logical Reasoning) are combined into a single 40-minute section with integrated sets, unlike some exams that separate them into distinct sections with independent timing. This means a CAT aspirant must be equally comfortable judging both types quickly within the same time block, which is the core skill this guide addresses.