Reading Comprehension for Placements: Strategies and Patterns

Reading comprehension is the verbal sub-section that decides whether the average engineering student clears the TCS NQT verbal cut-off or loses 8 marks they cannot afford to lose. Most students under-prepare for it. It does not feel like aptitude in the same way that quants or reasoning does.

The pattern is consistent across the major Indian IT campus drives. A passage of 200 to 400 words, followed by 4 to 8 questions covering factual recall, inference, vocabulary in context, and main idea or tone. The questions look easy in isolation; the time pressure is what makes them hard. This article covers why RC is tested, the passage types you will encounter, the strategies that move scores, how to practise efficiently, and what is changing in 2026 as AI-graded verbal assessments become standard.

Why reading comprehension is in every placement test

The verbal section of every major IT mass-hiring test includes reading comprehension as a core component. The reason is not academic taste; it is that the actual job of a fresher software engineer at TCS, Infosys, or Wipro begins with reading a 6-page requirements document, a Confluence page on internal tooling, or a client email thread, and extracting the actionable items. Companies test RC because they need the skill on day one of training.

The structural facts for the most common drives a 2026 final-year engineering student will sit:

Test	Verbal section	RC weight	Passage style
TCS NQT Foundation	24 questions, 25 minutes	5 to 8 RC questions, usually 1 to 2 passages	Short, factual, contemporary topics
Infosys InfyTQ	English section ~22 questions	6 to 10 RC questions, 1 longer passage	Medium length, mixed inferential and factual
Wipro Elite NTH	Verbal ~22 questions, 18 minutes	4 to 6 RC questions, 1 passage	Short, dense, business-style
Capgemini	Verbal aptitude ~25 questions	4 to 6 RC questions, 1 to 2 passages	Short, vocabulary-heavy
Cognizant GenC	Communication section	5 to 8 RC questions, 1 passage	Short, contemporary topics

The exact question count shifts cycle to cycle. The shape does not. If you can solve RC reliably on a short passage in under 5 minutes, the verbal section becomes a manageable timing exercise rather than a cliff. The TCS NQT aptitude questions with worked solutions cover the verbal sub-section format in detail, including the kind of passage that appears in the Foundation section.

The four RC passage types you will see

Recognising the question type before you read the passage is half the work. There are four patterns that account for almost every RC question on a campus placement test.

Factual / direct-recall questions

The answer is somewhere in the passage as a direct statement. Question stems look like “According to the passage…” or “The author states that…”. These are the fastest to answer because the work is locating, not interpreting. Skim the passage to know which paragraph holds which topic, then go directly to the relevant lines for each factual question. Time per question: 30 to 45 seconds.

Inferential questions

The answer follows from the passage but is not stated directly. Stems use phrases like “It can be inferred…”, “The passage suggests…”, “Which of the following would the author most likely agree with…”. Inferential questions reward careful elimination more than fast skimming. The wrong options are usually plausible-sounding restatements that go one step further than the passage actually supports. Time per question: 45 to 70 seconds.

Vocabulary-in-context questions

The stem gives a word from the passage and asks for its meaning as used in the passage, not its dictionary meaning in general. The trap is that the most common dictionary meaning is often offered as a wrong option. The right approach is to substitute each option back into the original sentence and pick the one that preserves the sentence’s meaning. Time per question: 30 seconds.

Main-idea, title, or tone questions

The stem asks what the passage is centrally about, what title fits best, or what tone the author uses (objective, critical, sympathetic, persuasive, neutral). These reward knowing the passage’s structure rather than its details. The right answer is usually the option that covers the whole passage, not just one paragraph. Time per question: 40 to 50 seconds.

A practical heuristic: if you see a 5-question passage, expect roughly 2 factual, 1 to 2 inferential, 1 vocabulary, and 1 main-idea or tone question. The mix shifts but rarely all in one direction.

Five strategies that actually move RC scores

These are the strategies that show up consistently in the practice debriefs of students who roughly double their RC accuracy across a four to six week preparation cycle.

1. Skim first, read on demand

Spend 60 to 90 seconds skimming the passage for structure: what is each paragraph about, who is the author addressing, what is the conclusion. Do not try to absorb every detail. Then read each question stem and go back to the passage for the specific answer. Reading the passage word-by-word in full first costs 2 to 3 minutes you do not have, and you will re-read most of it anyway when answering the questions.

2. Anchor on keywords from the question stem

Every question stem contains keywords (names, numbers, technical terms, dates, place names) that appear in exactly one part of the passage. Locate the keyword in the passage first, then read the surrounding two or three sentences. This collapses a 4-paragraph passage into a 3-sentence answer-source for each factual question.

3. Eliminate aggressively on inferential questions

Inferential questions almost always have one or two answer choices that are textbook traps:

• An option that restates a passage detail correctly, but does not actually answer the question asked.
• An option that goes beyond what the passage supports — true in general life, not supported by the specific passage.
• An option using extreme words like “always”, “never”, “all”, “none” that the passage does not actually say.

Eliminate these first; the remaining option is usually correct even if it does not feel definitive. The right inferential answer often sounds slightly weaker than the trap option, because it sticks closely to what the passage permits.

4. Time-box every passage

If you have not finished a passage in your allotted time (4 to 5 minutes for short passages, 6 to 8 for longer ones), mark the remaining questions for educated guesses and move on. The marginal value of one more careful inferential answer is lower than the marginal value of attempting 4 to 5 fast questions in the rest of the verbal section. With no negative marking on TCS NQT, every blank is a guaranteed zero, so an educated guess always beats no answer.

5. Trust the passage, not common sense

The most common reason a strong student gets an RC question wrong is choosing the answer that is true in real life but not supported by the specific passage. The test rewards reading what is on the page, not what you already know about the topic. If a question is about the author’s view on remote work, the right answer is what this author says, even if you personally disagree.

How to practise RC and self-assess

Volume is not the variable that fixes RC scores. Error analysis is. A student doing 30 untimed RC questions a day with no debrief usually plateaus around the same accuracy they started with. A student doing 4 to 6 timed RC questions a day with a structured debrief usually improves measurably over the same period, because the debrief is what surfaces which strategy is failing.

A workable daily practice routine:

• Pick one passage at the difficulty of your target test. Set a timer for 5 to 6 minutes for a short passage or 8 minutes for a longer one.
• Attempt every question in that window. Do not pause the timer to look something up.
• After the timer ends, mark the answers honestly without changing them.
• Spend 15 minutes on the debrief: for every wrong answer, identify which type of question it was (factual, inferential, vocabulary, main-idea), why the wrong option attracted you, and what would have eliminated it.
• Track the question-type breakdown of your wrong answers across two weeks. The pattern reveals which strategy above you need to drill.

For sourcing passages at the right difficulty, the best sites to prepare for placements round-up covers free practice platforms with verbal sections at TCS NQT and Infosys difficulty. Rotate sources every two weeks. The passage style on a single platform is consistent enough to memorise patterns rather than test general comprehension.

The TCS Ninja test pattern covers the full Foundation section structure, including the verbal sub-section breakdown. That helps when calibrating practice difficulty against the actual test.

Two reasonable targets for test day:

• High accuracy on RC questions, with three out of every four attempted answers correct on a representative practice set.
• Average pace of around 90 seconds per question across a passage, including the time spent skimming.

If you can hit both consistently in practice, the verbal section in any of the five major drives above is no longer the section that decides your overall score.

The AI layer: how verbal assessment is changing in 2026

Two things shifted in Indian IT campus hiring in 2026 that change how you should think about verbal preparation.

First, the verbal sections in many adaptive tests now run on AI scoring. SHL, Mercer Mettl, and HackerEarth all offer AI-graded essay-writing and structured-response questions in their 2026 product lines. The skills being measured (reading speed, inferential accuracy, tone recognition) are the same. The format is different: an adaptive test serves harder passages if you answer the easier ones correctly, and AI-graded essay questions evaluate not just spelling and grammar but argument structure and logical coherence.

Second, the composition of fresher hiring at the major IT services companies has shifted. TCS CHRO Sudeep Kunnumal stated in March 2026 that 60% of FY26 TCS fresher hires are AI-skilled, up from 10 to 15% three years ago. Separately, Financial Express reported that TCS cut its FY27 fresher intake to around 25,000 from 44,000 in FY26. Fewer total seats, with a heavier share going to AI-skilled candidates, raises the bar across the board, and verbal sections are no exception, because the candidates who clear the AI-skill bar tend to clear the verbal cut-off comfortably as well.

What this means for an RC preparation plan today: the verbal section format in 2026 hiring cycles is recognisably the same as five years ago, but the cohort competing for each seat reads faster and more accurately on average. The five strategies above are still the strategies. The accuracy threshold needed to be competitive is just a few percentage points higher than it used to be.

If you are interested in the AI tools that now grade verbal sections, and what it would take to build one yourself, TinkerLLM is the entry point for hands-on work with the LLM APIs that sit behind these new adaptive verbal scoring systems. The focus is on building, not on placement prep. Writing a small grader yourself changes how you think about answering one.