Quick answer: spend the first four to five minutes reading only the setup of every set, give each one a quick doability score, and start with the highest. If a set has no clear entry point after about six to eight minutes, leave it. Two fully solved sets usually beat four half-solved ones.
Why set selection decides your DILR score
A DILR section in recent CAT papers has had around 22 questions spread across sets of four to six questions, in 40 minutes. That is roughly ten minutes per set if you tried to attempt everything, which few people can do. The section is designed so that some sets are much friendlier than others, and the difficulty is rarely obvious from the first line.
This makes DILR different from QA or VARC. In QA, a bad question costs you two or three minutes. In DILR, a bad set can cost you twelve minutes and leave you with nothing, because the questions only open up once the whole arrangement is solved. The score you get depends heavily on which two or three sets you commit to.
So the goal is not to read every set carefully. It is to find, quickly and reliably, the sets you can finish.
The five-minute scan
Before solving anything, do one pass through all the sets. For each set, read the setup and glance at the questions; do not start drawing tables yet. Allow roughly 45–60 seconds per set.
1Read the first paragraph and the constraint count
Note what is being arranged (people, days, scores, routes) and roughly how many conditions there are. Many short, crisp conditions are usually friendlier than a few long, interlinked ones.
2Look for a fixed starting point
A doable set almost always has an anchor: one clue that pins a value directly (“P sits in seat 3”, “the total in March was 120”). If you cannot see where you would begin, that is a strong warning sign.
3Check the questions
Are the questions about the core arrangement, or do they add new conditions (“if X were also true…”)? Are some TITA? Sets where most questions follow directly from one solved table give you the most marks per minute.
4Score it and move on
Give each set a quick 1–3: 3 = I can see the table and the start, 2 = probably doable with some cases, 1 = no clear entry. Write the score on your rough sheet. Then start with the highest-scoring set.
Signals of a doable set
| Signal | Usually friendly | Usually costly |
|---|---|---|
| Representation | Obvious table or grid | No clear way to draw it |
| Starting clue | One value fixed directly | Every clue is relative |
| Cases | One or two cases | Many branches |
| Data | Complete numbers | Missing values to back-solve |
| Questions | About the solved table | Each adds a new condition |
| Length | Short, crisp clues | Long, wordy conditions |
None of these is a rule on its own. A long set with a fixed anchor and an obvious table can be the easiest one in the section. Use the signals together, and trust the anchor most.
Also, do not judge by topic alone. “I am good at games and tournaments” is helpful, but a hard tournament set is still hard. Choose sets, not topics.
Exit rules: when to leave a set
The most expensive habit in DILR is sunk cost: “I have already spent eight minutes, I cannot leave now.” Decide your exit rules before the mock, so the decision is not made under pressure.
- No entry by about six to eight minutes: if you still have no partial table that you trust, leave the set and go to your next-ranked one.
- Too many cases: if you are on your third branch and none has closed, step back. Either find a missed constraint or leave.
- Partial solve: if the table is half done, answer every question that the half already answers, then decide whether to push on.
- Last ten minutes: do not start a fresh, complex set. Go back to a partly-solved one, or pick up direct questions from a set you have read.
- 0–5 min: scan and score every set.
- 5–20 min: your best set, to completion.
- 20–33 min: your second set.
- 33–40 min: a third set only if it is clearly doable; otherwise, recheck answers and pick up partials.
A worked example
Imagine you scan four sets:
Scan notes
- Set A, seating (6 people): one person fixed, short clues, all questions about the arrangement. Score 3.
- Set B, bar chart of sales: clean data but two questions need a “percentage change of a ratio”. Score 2.
- Set C, tournament: no results given directly, every clue relative. Score 1.
- Set D, scheduling over a week: long text, but two days are fixed. Score 2.
Order: A, then B or D. Leave C unless you finish with plenty of time. If A turns out harder than it looked and you have no table by minute twelve, move to B instead of forcing it. In your post-mock review, check whether your scores were right: did the set you rated 3 really solve cleanly? Over a few mocks, this tells you how accurate your scan is.
Drills to get better at choosing
- Scan-only drill: take a past DILR section, give yourself five minutes, and only rank the sets. Then solve all of them untimed and see if your ranking was right.
- Exit drill: solve sets with a hard ten-minute stop. Note where you were when the timer rang.
- Post-mock set log: for every set in a mock, record your scan score, time spent, questions correct, and “should I have picked it?”. This is the single most useful DILR log you can keep.
Our mock-analysis guide covers how to fit this set review into a complete post-mock routine.
Frequently asked questions
How many DILR sets should I attempt in CAT?
It depends on the paper, but many strong scorers fully solve two to three sets and pick up a few questions from a fourth. Accuracy on the sets you choose matters more than the number you touch.
Should I attempt DILR sets in order?
No. The order in the paper says nothing about difficulty. Scan every set first and start with the one that looks most doable to you.
Is it a waste of time to scan all the sets?
A five-minute scan feels slow, but it usually saves far more time than it costs by keeping you out of the one set that would have eaten fifteen minutes.
