Common DILR mistakes in CAT, with worked examples of each

When you review a DILR section honestly, very few wrong answers turn out to be questions you could not have solved. Most come from a small set of process errors that repeat mock after mock. The good news is that process errors are fixable in weeks, because they do not need new knowledge. They need a checklist and the discipline to use it.

This guide groups the common mistakes by where they happen, then works through a small puzzle that shows the two most expensive ones in action.

The mistakes at a glance

The choosing mistakes are covered in time management in DILR. The rest are below.

A small puzzle to learn from

Five friends A, B, C, D and E sit in a row of five seats, numbered 1 to 5 from left to right.

1. B's seat number is exactly one more than A's.
2. C sits at one of the ends.
3. E sits next to C.
4. D does not sit at an end.

Solve it. C is in seat 1 or seat 5.

• C in seat 5. E must be in seat 4. That leaves seats 1, 2 and 3 for A, B and D. The A–B pair needs two adjacent seats: 1–2 or 2–3. If A–B take 2–3, D is in seat 1, an end, which breaks condition 4. So A–B take 1–2 and D takes seat 3. Row: A B D E C.
• C in seat 1. E must be in seat 2. Seats 3, 4 and 5 remain. If A–B take 3–4, D is in seat 5, an end, which is not allowed. So A–B take 4–5 and D takes seat 3. Row: C E D A B.

Both rows satisfy all four conditions, and an exhaustive check of all 120 orders confirms there are exactly two. Now look at three questions.

Q1. Who sits in seat 3? D, in both rows. That is a must-be-true answer: D.

Q2. Which of these must be true? (a) A sits at an end (b) E sits next to D (c) B sits in seat 2 (d) C sits in seat 5.

• (a) is true in the first row and false in the second.
• (b) E is in seat 4 or seat 2, and D is in seat 3 both times, so E is always next to D.
• (c) and (d) hold only in the first row.

Answer: (b).

Mistake 1: treating one case as the only case

Suppose an aspirant finds A B D E C first and stops there. On Q2 they see that A is in seat 1 and pick (a). It is wrong, because the second row exists.

The fix is a habit, not a talent. Before any "must be true" or "which of the following is definitely" question, ask whether your arrangement was forced at every step. If you made a choice anywhere ("let me put C on the right"), you must go back and test the other branch.

Mistake 2: missing a condition

Now suppose a different aspirant overlooks condition 4, "D does not sit at an end". The puzzle then has four valid rows:

On Q1 this aspirant concludes that seat 3 "cannot be determined", which is wrong. One skipped line changed a clean answer into an apparent dead end, and they may even abandon a set that was nearly finished.

The fix: number the conditions and tick each one on your rough sheet as you use it. Before answering the first question, check that every condition has a tick. Conditions stated in the negative ("does not", "is not next to") are the ones most often missed, because they do not place anyone.

Carrying a question's condition forward

Many CAT sets add information inside a question: "If E sits in seat 4, who sits in seat 1?" In our puzzle that condition picks the first row, so the answer is A.

The trap comes in the next question. If the next one asks "who sits in seat 1?" with no extra condition, the answer is "A or C", not A. Aspirants who keep the first row in their head from the previous question answer A.

Fix: when a question adds a condition, solve it on a separate small copy of the grid or draw a clear line under it on your rough sheet. The main diagram should only ever hold what is true for the whole set.

Taking a percentage on the wrong base

In DI sets, percentage change is always measured from the earlier value. If sales go from 80 to 100 units, that is a rise of 20/80 = 25%. If they then fall back from 100 to 80, that is a fall of 20/100 = 20%. Same 20 units, two different percentages.

Questions like "by what percentage was the 2024 figure lower than the 2025 figure?" deliberately flip the base. Read "lower than X" or "higher than X" and put X in the denominator.

Over-calculating

Suppose a question asks what share of a total of 1,219 one category of 347 represents, and the options are 18%, 28%, 38% and 48%. There is no need to divide to two decimals. A quarter of 1,219 is about 305 and a third is about 406, so the share is between 25% and 33%. Only 28% fits. (The exact value is about 28.5%.)

The habit: read the options before you calculate. If they are far apart, estimate. If they are close, only then compute carefully. The DILR calculation speed guide has the fraction table and shortcuts that make estimation fast.

Messy diagrams

When the grid changes shape halfway through a set, the same deduction gets made twice or not at all. Decide your layout for each common set type in advance:

• Linear arrangement: seats as columns, one row per attribute.
• Scheduling: days as columns, slots or people as rows.
• Tournaments: a results matrix with teams on both axes, plus a points column.
• Binary logic: a truth table with one column per person's statement.

Using the same layout every time means your eyes know where to look, which saves more time than any trick.

Run a mistake audit

After each mock, tally every wrong or wasted answer by category. A made-up tally after three mocks might look like this:

In this example, "one case taken as only case" is the most persistent error, so that aspirant's fix for the next week is the "is this forced?" check on every must-be-true question.

What to do next

• Solve the five-seat puzzle above yourself, without looking, and confirm both rows.
• Number and tick conditions in every set you solve this week.
• Start the mistake audit with your next mock, as part of your DILR mock analysis.
• Practise two arrangement sets from arrangements in CAT DILR and, for each must-be-true question, write down whether the answer holds in every case.