Core approach: build an explicit attempt-defer-skip rule for each of the three sections, because time saved in one section cannot be transferred to another. In English, name the grammar rule before fixing a sentence and separate inference from vocabulary-in-context in RC. In Quantitative Aptitude, check the options before calculating and use approximation and problem-family recognition. In Reasoning, time the puzzle setup separately from solving and set an abandon checkpoint. Track readiness with a rubric: rule-naming accuracy in grammar, approximation used in most DI items, and at least one disciplined puzzle abandonment per drill. Self-check scores are learning milestones only, not predictions of your result.
Why the prelims needs per-section triage, not one global strategy
Because each section is timed separately, your total score is the sum of three independent decisions about what to attempt, defer, or leave. Build a written triage rule for each section and rehearse it inside timed drills.
A single 'solve fast' plan fails here because speed is not transferable between sections. Being quick through English passages does not buy you extra minutes for a seating arrangement, so each section must be practised as a closed unit with its own internal checkpoints. In practice, that means setting a rough mid-section marker (for example, halfway through your question budget) and checking whether you are on the kinds of questions you planned to reach.
Triage is a skill you rehearse, not a decision you improvise on exam day. Sort every question type you meet into three categories: attempt first (reliable within your self-set budget), defer (attempt if the checkpoint allows), and skip without guilt. Write the list down, use it in every drill, and revise it as your accuracy data changes. The table below is a starting template you should adapt to your own drill results.
| Question type | Attempt first when... | Defer when... | Abandon cue |
|---|---|---|---|
| RC vocabulary-in-context | You can substitute your own word before looking at options | The passage paragraph is dense and you have not read it | You need a dictionary-style definition, not context |
| RC inference/detail | You can point to a supporting line for each option | The question asks about the whole passage and you skimmed it | Two options both look defensible after two checks |
| DI set (tables, charts) | The option gaps are wide, so approximation is safe | Options are close together and require exact ratios | One question needs a long chain of exact calculations |
| Standard arithmetic item | You recognise the problem family immediately | The setup is wordy and the family is unclear | Two different setups both seem valid |
| Puzzle or seating set | The frame locks quickly from a fixed-position clue | Most clues are comparative ('somewhere left of') | The frame has not locked after two reading passes |
| Syllogism / inequality | You can apply the standard diagram or sign rules directly | The conclusions need a drawn possibility case | You are guessing between 'either-or' readings |
Reading Comprehension: separating inference from vocabulary-in-context
RC items test two distinct skills: inferring what the passage supports, and identifying what a word means in that specific sentence. Train them separately, because each has a different failure mode.
Inference questions demand an answer traceable to the passage text. The plausible traps are options that overstate the author's claim, import general knowledge the passage never states, or extend a limited point into a universal one. Build the habit of asking, for every option, 'which line proves this?' If you cannot point to a line, the option is a trap even if it sounds reasonable. Practise this on short passages first, where you can verify the line reference quickly, then move to longer ones.
Vocabulary-in-context items invert dictionary habits: the correct meaning is the one the sentence supports, not the most common one. The reliable method is to cover the options, substitute your own plain word, and only then match the choices. For example, if a passage on a market slump says the 'muted' response of investors confused analysts, your own substitute might be 'weak' or 'quiet' rather than a sense of silencing someone. Options that fit the word's other meanings exist precisely to punish dictionary-first reading.
Grammar error-spotting: name the rule before you fix the sentence
The error items you practise can usually be diagnosed with a small set of named rules: subject-verb agreement, tense consistency, pronoun reference, misplaced modifiers, and parallelism. Diagnosing the rule first prevents plausible-sounding wrong edits.
Subject-verb agreement items in practice sets are usually built with distance between the subject and the verb: 'The list of approved items is enclosed' is correct because 'list', not 'items', is the subject. Recurring variants include collective nouns treated as one unit, amounts of money or time taken as single quantities, and 'each' or 'either' as singular subjects. Practise by striking out prepositional phrases between subject and verb until you can see the agreement choice instantly.
Modifier and parallelism items require a different check: read the sentence for what each phrase attaches to. A misplaced 'only' changes meaning without any grammar flag, and an unbalanced 'not only... but also' pairs a noun with a clause. A useful drill is to take ten error items and, before answering, write the rule name you believe each one tests. Compare with the explanations afterwards; the goal is not the answer but correct diagnosis, because a wrong diagnosis produces a confident wrong edit.
- Agreement traps: prepositional phrase between subject and verb; collective nouns; quantities treated as one unit.
- Tense traps: a past-tense main clause with an unprompted present-tense subordinate clause.
- Modifier traps: opening phrases that do not describe the subject that follows; floating 'only'.
- Parallelism traps: lists mixing nouns with verb phrases; 'not only X but also Y' where X and Y differ in form.
Data Interpretation: approximate first, calculate only when forced
DI sets reward ratio sense and approximation over exact arithmetic. Before any calculation, read the options: wide gaps mean estimation is safe, while narrow gaps signal that one exact step may be necessary.
Three named techniques handle a large share of typical DI practice sets: fraction-percentage equivalents (knowing that 4/11 is roughly 36%, 5/8 is 62.5%, and so on), relative comparison of two ratios by cross-multiplication instead of division, and the successive percentage change formula for two-step growth. Checking the options first is part of the method, because an option spread of ten points makes estimation reliable while a spread of one point does not. Practise stating your estimate before you compute anything.
Worked scenario: profits rise from 3.6 lakh to 4.5 lakh and the question asks the percentage increase. A common mistake computes 0.9 divided by 4.5, giving 20% — the change relative to the new value. The better decision is to divide the change by the original value: 0.9 over 3.6 equals 25%. It matters because 20% appears among the options as a distractor, so the error is not a slip but a systematic confusion of the base. In every percentage-change drill, write down the base before computing.
Arithmetic: recognise the problem family before reaching for formulas
Arithmetic questions map onto recognisable families: time and work, mixtures and alligation, profit and loss, simple versus compound interest, and speed-time-distance. Identification is the primary skill; the formula follows from it.
Time-and-work items are best handled with a total-units method: set total work as a common multiple of the given days, convert each agent into units per day, and add rates for combined work. Mixtures and alligation problems reduce to weighted averages on a straight line between the two component strengths. For interest items, keep the distinction crisp: simple interest grows by a fixed amount each period on the original principal, while compound interest recalculates on the growing balance, which is why the two diverge over time.
Worked scenario: a pipe fills a tank in 6 hours and a leak empties a full tank in 12 hours; how long to fill with both open? A plausible mistake adds the rates, 1/6 plus 1/12, giving 4 hours. The better decision treats the leak as a negative rate: 1/6 minus 1/12 equals 1/12, so the tank fills in 12 hours. It matters because the sign of the leak is the entire concept being tested; adding rates means the item's premise was missed. In your drill log, tag any wrong sign as a 'direction error', not a calculation error.
Puzzles and seating arrangements: commit or abandon at a fixed checkpoint
A puzzle set is a fixed-cost investment: in a typical practice set, extracting the setup consumes a large share of the time, and the payoff only arrives if the arrangement locks. Decide in advance at what checkpoint you will continue or abandon the set.
Process matters more than speed here. Read every clue before placing anyone, draw a grid or line, and start with the most restrictive clue — a fixed position such as 'R sits at the extreme left end' — because it anchors everything else. Handle comparative clues ('P sits somewhere to the right of S') by making small branches rather than one rigid diagram. Logical deduction items such as syllogisms and inequalities follow their own standard routines: possibility cases for syllogisms and sign-chain comparison for inequalities, so keep them in a separate fast-attempt bucket from full puzzles.
Worked scenario: a seven-person linear arrangement gives two fixed-position clues and four comparative clues. A plausible mistake is to start placing people after reading only the first two clues, then rebuild the diagram when a later clue contradicts it — doubling the time spent. The better decision is a complete reading pass first, starting the frame from the fixed positions, and a pre-declared checkpoint: if the frame has not locked after two passes, mark the set and move on. It matters because sunk-cost pressure in a puzzle is a time decision, not an intelligence one, and it is trainable.
A weekly drill sequence with a self-check rubric and readiness checks
Rotate one section per day, keep a mistake log that records the rule or decision violated, and run one full timed sectional per week. Measure yourself against a rubric of budgets and behaviours, not raw question counts.
A workable rotation for one week: Day 1, two RC passages plus fifteen to twenty grammar items, logging the rule name for each error; Day 2, two DI sets with estimation forced by covering the calculator habit; Day 3, two puzzle sets with the setup phase timed separately from solving; Day 4, one arithmetic family (for example, alligation) across ten items; Day 5, a mixed sectional run using your triage table; Day 6, a full timed run of all three sections back to back; Day 7, review the mistake log and revise the triage table. Repeat the rotation, adjusting the mix toward whichever section's rubric scores are lowest.
Practical exercise with expected observations: run one timed English drill, one timed DI set, and one timed puzzle set this week and score them against this rubric. Grammar: you named the tested rule before answering in at least eight of ten error items. DI: you stated an estimate before computing in at least three of five questions, and no percentage-change error used the wrong base. Puzzles: you read all clues before placing anyone in both sets, and you made at least one deliberate abandonment within your budget. If a behaviour is missing rather than inaccurate, that is the finding to act on — process gaps and knowledge gaps have different fixes.
- Readiness check 1: you can state your attempt-defer-skip rule for each section from memory, without looking at the table.
- Readiness check 2: in grammar items, rule-naming is faster than answer-selecting.
- Readiness check 3: you wrote down the base (original value) before every percentage-change computation this week.
- Readiness check 4: at least one puzzle set per drill was abandoned on schedule, with no second attempt.
- Remember: these rubric scores are learning milestones, not predictions of your exam result.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.