How to Improve CAT VARC Score: 7 Targeted Fixes for 85-95 Percentile Aspirants
A diagnostic-first guide to improving CAT VARC score from the 85 to 95 percentile band: 7 specific bottlenecks (RC passage mapping, fact-vs-inference confusion, para jumble method gap, option elimination weakness, reading speed mismatch, genre blindness, TITA-skipping), targeted fixes with expected percentile lift per fix, and a sequential 12-week drill plan. Total VARC commitment: 45-60 min/day. Realistic outcome: +10 percentile lift in 12 weeks.
How to Improve CAT VARC Score: 7 Targeted Fixes for 85-95 Percentile Aspirants
The most-searched VARC question after "what is VARC" is not "how to start VARC" but "why is my VARC score not improving." The 85 to 95 percentile band is the most common plateau in CAT VARC because aspirants in that band have done the easy work (general reading, basic RC practice) and are now stuck against specific bottlenecks that volume practice alone does not fix. The lift from 85 to 95 percentile is not a "read more newspapers" lift; it is a "diagnose your specific weakness and apply the targeted fix" lift.
This guide is the diagnostic-first approach to improving CAT VARC score: identify which of the seven specific bottlenecks applies, apply the targeted fix, and re-test in 4 to 6 weeks. Aspirants who do the diagnostic correctly typically see a 5 to 8 percentile lift by week 6 and the full 10 percentile lift by week 12.
The 85 to 95 VARC plateau is fixed by diagnosing one of 7 specific bottlenecks — RC passage mapping, fact-vs-inference confusion, para jumble method gap, option elimination weakness, reading speed mismatch, genre blindness, or TITA-skipping. Each bottleneck has a targeted fix that takes 4 to 6 weeks. Total VARC time commitment: 45 to 60 min/day. Realistic lift: 5 to 8 percentile in 6 weeks, full 10 percentile in 12 weeks. The fix is not "more practice" — it is "right practice on the diagnosed weakness."
Diagnose First: Which Bottleneck Is Yours?
The single biggest mistake aspirants make at the 85 percentile plateau is starting "more VARC practice" without diagnosing which weakness is producing the cap. The seven bottlenecks below are mutually distinguishable: each one has a specific symptom pattern in mock data. Run the diagnostic table against your last 3 to 5 mock VARC sections before choosing the targeted fix.
| Symptom in mock data | Likely bottleneck | Targeted fix |
|---|---|---|
| RC accuracy < 60%, run out of time | Passage mapping weakness | Fix 1 |
| Fact questions correct, inference wrong | Fact-vs-inference confusion | Fix 2 |
| Para jumble: 2-3 wrong out of 5 | Method gap (no anchor rule) | Fix 3 |
| Down to 2 options, pick wrong one | Option elimination weakness | Fix 4 |
| Reading speed < 220 wpm | Speed bottleneck | Fix 5 |
| Strong on biz RCs, weak on philosophy | Genre blindness | Fix 6 |
| Skipping TITA in time pressure | TITA-skipping policy | Fix 7 |
The 7 Targeted VARC Fixes
Build a 30-Second Argument Map Before Answering
Symptom: reading the passage sequentially, finishing the read, then getting confused on every question. Time taken per RC: 12 to 15 min.
After reading the passage, take 30 seconds to mentally map: main argument (1 sentence), 2 to 3 secondary arguments, and the author's tone (positive, neutral, critical, sceptical). Write nothing on the screen, but build the map in your head before clicking the first question. The map becomes the index for every fact-recall question and the anchor for every inference question.
Treat Fact and Inference Questions Differently
Symptom: 80 percent accuracy on fact-recall, 45 percent accuracy on inference. Confused by partially-correct option distractors.
Before answering any RC question, classify it: fact-recall ("according to the passage..."), inference ("the author would most likely agree...", "which is the best summary..."), or tone ("the author's stance is..."). Fact questions need exact-phrase search in the passage. Inference questions require synthesis across 2 to 3 sentences plus elimination of extreme options. Tone questions need recall of the author's adjective choices.
Anchor With a Mandatory Pair, Then Eliminate Options
Symptom: para jumble accuracy at 40 to 60 percent. No systematic method, relies on intuition.
Read all 4 to 5 sentences. Identify exactly two sentences that MUST follow each other directly because of a definite connector (pronoun referring to a noun in another sentence, a connecting word like however or therefore, a definite article referring to something just introduced). This becomes the mandatory pair. Eliminate any answer choice that does not preserve the pair. The remaining 1 to 2 options are tested against contextual flow.
Use the Extreme-Word Heuristic on Inference Options
Symptom: regularly down to 2 options and picks the wrong one. The 50-50 problem.
Extreme words (always, never, only, must, none, completely, entirely) are 70 percent likely to be wrong on inference questions. Moderate words (often, sometimes, may, likely, partially) are 70 percent likely to be right. When stuck between two options, pick the one with moderate language unless the passage explicitly supports the extreme position. The other heuristic: if one option uses passage-exact phrasing and the other paraphrases, the paraphrase is more likely correct on inference questions (the exact-phrase option is often a trap).
Build to 250-300 WPM on RC Passages
Symptom: cannot finish all 4 RC passages within the section time. Reading speed below 220 wpm.
Daily 20-minute reading practice on long-form journalism (Aeon, The Atlantic, The Caravan, The New York Times Magazine). Read at slightly uncomfortable pace, then test comprehension by writing a 3-sentence summary. After 4 weeks, baseline reading speed lifts by 30 to 50 wpm. The goal is not maximum speed; it is the 250 to 300 wpm sweet spot where comprehension stays above 80 percent on inference-heavy passages.
Different Genres, Different Reading Pace
Symptom: strong on business and current-affairs RCs, weak on philosophy, sociology, and literary criticism RCs.
Each genre has a different reading pace and inference structure. Philosophy RCs need slower reading and explicit tracking of "A argues X, B counters with Y, the author concludes Z." Sociology RCs need attention to qualifier words ("most", "some", "in particular contexts"). Literary criticism RCs need awareness of unmarked author opinion buried in descriptive language. Rotate your daily RC practice across all 5 to 7 CAT genres rather than sticking to comfortable business and science passages.
Never Skip the No-Negative-Marking Questions
Symptom: running out of time on the last 3 to 5 questions, which are usually TITA. Leaving 6 to 8 free marks on the table.
TITA non-MCQ questions in VARC (typically para summary and para jumble) carry +3 for correct and zero penalty for incorrect. Mathematically, every TITA attempt is profitable. Move TITA into the first or second pass of your section pacing. Optimal VARC pacing: 3 min scan, 12 min on the first 12 questions including all TITA in their natural position, 25 min on the remaining RC. The CAT 2026 marking scheme guide covers the +3/−1/0 math in detail.
Aspirants try to apply all seven fixes simultaneously and end up doing none of them well. Pick the single highest-yield fix from your mock diagnostic, drill it for 4 to 6 weeks, then move to the next. Sequential focus produces faster percentile lift than parallel scatter.
The 12-Week VARC Score Lift Plan
A realistic 12-week plan structures the seven fixes into a sequential drill cadence. The plan assumes 45 to 60 minutes daily on VARC across the full 12 weeks.
- Week 1: Diagnostic. Take 2 free VARC sectional mocks (see the CAT 2026 free mock tests guide), tag every wrong answer by bottleneck category, identify the top 2 bottlenecks.
- Weeks 2 to 6: Drill the highest-yield bottleneck. 3 RC passages per day plus alternate-day para jumble and odd-one-out. Weekly sectional mock to track lift.
- Weeks 7 to 10: Drill the second highest-yield bottleneck. Maintain the first bottleneck through ongoing RC practice. Add full-length mocks every 5 days.
- Weeks 11 to 12: Integration. Full mocks with strict section pacing, focus on TITA-first policy and option elimination discipline in mock conditions. Mock analysis includes per-question time tracking.
Want a personalised VARC improvement plan calibrated to your specific bottleneck and current percentile band?
Build My VARC PlanHow VARC Connects to Overall CAT Strategy
The 10 percentile VARC lift from 85 to 95 typically translates to a 3 to 5 percentile lift in overall CAT score, depending on QA and DILR baselines. For aspirants whose Quant is also stuck below 95, the parallel how to improve CAT Quant score guide covers the analogous diagnostic approach for the QA section. Aspirants targeting overall 99 percentile typically need both VARC and QA at 95+ along with DILR at 90+.
For aspirants whose VARC plateau is genuinely below 85 percentile, the diagnostic approach is the same but the bottleneck list is different (reading speed and vocabulary depth become the dominant constraints). The fixes in this guide are calibrated to the 85 to 95 band specifically.
The single highest-leverage habit for VARC improvement is keeping a daily error log. Every wrong answer in mock or practice gets tagged by bottleneck (fact-vs-inference, option elimination, etc.). After 4 weeks, patterns emerge: one or two bottlenecks dominate the error count, and the targeted fix becomes obvious. The error log replaces guesswork with diagnosis.
- Diagnose before drilling. Run mock data through the 7-bottleneck table first.
- Build a 30-second argument map before answering RC questions.
- Classify each RC question (fact / inference / tone) before answering.
- Use the mandatory-pair method on every para jumble.
- Apply the extreme-word heuristic to coin-flip-zone options.
- Target 250 to 300 wpm reading speed, no faster.
- TITA first or in natural-order pass. Never skip TITA in VARC.
VARC does not improve from more reading. It improves from the right kind of reading applied to the right diagnosed weakness, for 12 disciplined weeks.
Get a Personalised VARC Improvement Plan
The Optima Learn CAT 2026 waitlist builds a personalised VARC plan calibrated to your specific bottleneck, current percentile, and weekly time commitment.
Build My VARC Plan
Optima Learn Editorial Team
CAT preparation specialists publishing structured guides on the CAT exam, IIM admissions, and MBA entrance prep. We track CAT VARC bottleneck patterns, percentile-band lift outcomes, and sectional fix-to-mock conversion ratios across cycles.
Practice VARC on real CAT passages
Tagged RC, para-summary, and sentence-completion drills from CAT 2010 to 2024.