GMAT Verbal study plan for low-B2 English candidates

Most candidates who sit the GMAT Focus with English reading skills sitting around the B1 to low-B2 band on the CEFR scale lose points on Verbal not because they cannot reason, but because the section punishes a specific pair of weaknesses at once: slow decoding of academic prose, and a vocabulary that is too thin to anchor Critical Reasoning conclusions. This article lays out a preparation plan built for that reader profile. It assumes the candidate can hold a conversation in English and read a newspaper article with a dictionary, but has never read an academic journal abstract without help. The plan below respects the GMAT Focus format — three Verbal item families, 23 questions, roughly 45 minutes — while quietly rebuilding the underlying English engine. It is engineered in three blocks of four weeks, with a hard reassessment gate at the end of each block so the candidate never wastes a month grinding material that is not the bottleneck.

Why a low-B2 candidate needs a different GMAT Verbal plan

Most generic GMAT Verbal study plans assume a reader who decodes academic English at near-native speed and who only needs to learn the section's argument logic. That assumption collapses below B2. At low-B2, the candidate's first problem is not the argument; it is the sentence. Long noun phrases, embedded clauses, and conditional connectives (whereas, insofar as, provided that) add three to five seconds of decoding time per sentence, and on a 45-minute section that compounds to roughly seven minutes of pure friction — almost a sixth of the total clock. The candidate who feels "I understood the passage but ran out of time" is almost always paying this tax.

The second problem is lexical anchoring. Critical Reasoning stems on the GMAT Focus repeatedly test the difference between words that look synonymous but are not: mitigate versus eliminate, imply versus suggest, compelling versus persuasive. A B2 reader often picks the answer that is plausible in conversational English and loses the question because GMAT Verbal rewards the more precise word. The study plan must therefore interleave two tracks: a reading-and-decoding track, and a precision-vocabulary track, with the latter starting in week one and never stopping.

The third problem is confidence calibration. Low-B2 candidates tend to misjudge their own comprehension. They read a sentence, translate it mentally into their first language, and assume they understood it. The translation is often slightly off, and the slight off-ness is exactly where GMAT Verbal wrong answers live. For that reason, the plan below uses a "read, summarise aloud, then re-read" loop that catches the gap between the impression of understanding and the reality of it.

What this plan is not. It is not a quick fix, and it is not a vocabulary-list memorisation scheme. The candidate who spends 30 days on a 2,000-word Anki deck and zero days on argument logic will arrive at test day with a fuller vocabulary and the same V40. The plan is also not a substitute for professional English lessons; it sits on top of them. The candidate who can pair this plan with two hours a week of structured English instruction with a teacher will move faster than a candidate working alone.

The 12-week architecture: three blocks, three gates

The plan runs 12 weeks and is divided into three four-week blocks. Each block ends with a low-stakes diagnostic: a 23-question Verbal section from the official GMAT Focus practice catalogue, taken under timed conditions, and scored against the candidate's previous gate. The candidate only progresses to the next block if the gate shows movement; if it does not, block two or three is repeated rather than abandoned.

Block 1: English engine rebuild (weeks 1–4)

Block 1 is the unsexy block. It does not contain a single Critical Reasoning drill, and the candidate will not see a full Verbal section for the first ten days. The goal is to move the candidate from "I sort of understand" to "I understand the sentence the first time I read it." Three sub-tracks run in parallel:

Daily reading, 40 minutes. Two academic English sources only. The Economist's science section, the Aeon magazine long reads, or the New Yorker's longer reportage pieces. No fiction, no news headlines, no Twitter. The candidate reads a single 1,200-word article per day, twice, with a 15-minute gap, and writes a five-sentence summary in English. The summary is the test: if the candidate can produce it, the comprehension was real.
Sentence deconstruction, 20 minutes. The candidate takes ten long sentences from GMAT-style Reading Comprehension passages, identifies the subject, the verb, and the object of each clause, and rewrites the sentence in plain English. The point is to train the eye to find the grammatical spine before the eye gets lost in modifiers. For most low-B2 candidates, this drill is the single highest-leverage activity in the first month.
Precision vocabulary, 25 minutes. Twenty new academic-English words per day, drawn from a curated list of roughly 600 words that overlap heavily with GMAT Verbal stems. The list is grouped by semantic family (causation, contrast, qualification, evaluation), not alphabetically, so the candidate learns the differences inside a family rather than the words in isolation.

Block 1 also installs the question bank. The candidate should buy or obtain one of the official GMAT Focus question packs and use it as the source for every later drill. Random online Verbal questions have a different stem register, and training on them at this stage builds habits that will have to be unlearned.

Block 2: argument logic and item-family fluency (weeks 5–8)

Block 2 introduces the three GMAT Verbal item families: Reading Comprehension, Critical Reasoning, and the standalone Sentence Correction–style items that test grammar and concision. Roughly 60% of block 2 is Reading Comprehension, because for a low-B2 candidate, RC is the family where reading speed and lexical depth actually convert into points. Critical Reasoning gets 30%, and the grammar items get 10%.

Each week in block 2 has the same shape: five timed RC passages (about 9–10 questions total), three untimed CR drills focused on a single stem type (strengthen, weaken, evaluate, boldface, inference), and one full 23-question Verbal section on day six as a mock-mini. Day seven is rest and a 30-minute vocabulary review. The candidate should not score the mock-mini obsessively in week five; the first mock is a baseline, not a verdict.

Block 3: pacing, endurance, and test-day simulation (weeks 9–12)

Block 3 is where the candidate learns to use the English engine under pressure. The first two blocks should have eliminated most of the per-sentence decoding tax. Block 3's job is to push the candidate's Verbal time per question from "panicked" to a stable two minutes or less, with a four-question buffer in case a passage is unusually dense.

Week 9 is a hard reset: the candidate takes a full official GMAT Focus mock under realistic conditions, scores it, and writes a one-page reflection identifying the two item families that lost the most points and the two that lost the most time. Weeks 10 and 11 are content refinement, drilled against the families that lost points in week 9. Week 12 is taper: two more full mocks, but only one per week, with a review day after each.

Reading-comprehension training when English is the bottleneck

Reading Comprehension is where the low-B2 candidate will gain or lose the most Verbal points, and it is also the family where the gap between conversational English and academic English hurts the most. GMAT Focus RC passages run 200 to 350 words and are written in a register closer to a long-form magazine article than to a textbook chapter. They use long sentences, embedded conditionals, and vocabulary that rewards precision over breadth.

The single most useful training habit is a three-pass read of every passage, even on questions that look like detail questions. Pass one is a 30-second skim for topic and structure. Pass two is a real read, sentence by sentence, with the candidate pausing at the end of each paragraph to write a one-line gist on a piece of paper. Pass three is the question-by-question read, where the candidate returns to the text only to confirm what pass two already established. Low-B2 candidates who skip pass two and try to answer detail questions directly spend an extra 30 to 60 seconds per question and still pick wrong answers, because the detail was anchored to a sentence they never fully understood.

The second habit is the "two-stem test." When a candidate is stuck between two RC answer choices, the test is to find the word or phrase in the passage that supports the chosen answer. If the candidate cannot find it within ten seconds, the answer is wrong. This is a useful guard against the conversational-English trap, where the answer that sounds most reasonable is the one that has no textual support.

The third habit is paragraph-mapping. Before reading the questions, the candidate labels each paragraph with a single word or short phrase: claim, example, counter, data, implication. The map takes 30 seconds to build and saves minutes later, because the candidate can locate the question's target paragraph in two seconds instead of rescanning the whole passage. For low-B2 readers, paragraph-mapping is also a comprehension check: if the candidate cannot summarise a paragraph in one word, that paragraph is the one that needs a slower second read.

Critical Reasoning as argument logic, not as a vocabulary test

Many low-B2 candidates approach Critical Reasoning as a vocabulary test, and that is the first habit to break. CR on the GMAT Focus is, at its core, four move types — claim, evidence, assumption, counter — and a small library of stem shapes. The vocabulary inside the argument is rarely the obstacle; the obstacle is that the candidate treats the argument as a paragraph to be understood rather than as a structure to be mapped.

The protocol below is the one I would walk a low-B2 candidate through at the whiteboard. It is the same protocol for every CR question, regardless of stem.

Step 1 — Read the question stem first. The stem tells the candidate what role they are playing. "Which of the following most strengthens the argument?" means the candidate is hunting for the missing link. "Which of the following most seriously weakens the argument?" means the candidate is hunting for the counterexample that breaks the chain. Reading the stem first prevents the candidate from re-reading the argument through the wrong lens.
Step 2 — In one sentence, state the conclusion. The candidate writes down, in their own words, what the argument is trying to prove. If the candidate cannot state the conclusion in a single sentence, the argument is too dense and the candidate should re-read. This step is also where the B2 ceiling shows up: the candidate thinks they know the conclusion but is actually summarising the topic.
Step 3 — Identify the evidence. What fact, statistic, or observation is the conclusion built on? Often there is only one piece of evidence, and identifying it sharply reveals the assumption.
Step 4 — State the assumption. The assumption is the unstated link between evidence and conclusion. For a low-B2 candidate, this step is the highest-leverage. A candidate who can name the assumption can answer 70% of strengthen, weaken, and evaluate questions without even reading the choices.
Step 5 — Only now read the answer choices. Many candidates reverse this order. Reading the choices before mapping the argument forces the candidate to evaluate each choice against an undefined target, which is exactly the situation in which the conversational-English trap fires.

For low-B2 candidates, the most common failure mode on CR is not picking the obviously wrong answer. It is picking the answer that defends a different conclusion than the one in the stem — usually a conclusion the candidate inferred from the topic rather than from the argument. The four-step protocol above closes that gap, because step 2 forces the candidate to commit to a specific conclusion in writing before the answer choices appear.

Vocabulary and grammar: building the precision layer

Vocabulary and grammar work sits underneath every other block and never stops. The list below is the routine I would set for a low-B2 candidate across all 12 weeks. The total time per day is roughly 50 minutes, and it can be split across two sessions.

20 new words per day, with example sentences. The candidate learns the word, the part of speech, two synonyms, two antonyms, and one GMAT-style sentence using the word. The example sentence is the test: if the candidate cannot construct a plausible GMAT sentence, the word is not yet owned.
15 minutes on connectives and qualifiers. Words like nonetheless, conversely, albeit, insofar as, to the extent that, with the exception of. These are the words that change the meaning of an entire sentence by one degree, and they appear in nearly every Reading Comprehension passage. A candidate who can read "the data are compelling, albeit limited" correctly understands that the author is hedging, not endorsing.
15 minutes on grammar patterns tested in Verbal. Subject-verb agreement with collective nouns, pronoun-antecedent agreement across clauses, modifier placement, parallelism. These patterns appear in the standalone sentence items on the GMAT Focus and they also appear inside RC passages, where a misplaced modifier can flip the meaning of a paragraph.

The candidate should not use a general English vocabulary list. The list should be one of the curated GMAT Verbal word lists, and the candidate should cross-reference it with the words that appear in their wrong-answer log. If the candidate is consistently missing Critical Reasoning questions because they confuse imply and infer, those two words go to the top of the daily review, regardless of what the original list said.

Pacing, endurance, and the four-question buffer

The GMAT Focus Verbal section is 23 questions in roughly 45 minutes, which works out to about 117 seconds per question. For a low-B2 candidate, that budget is too tight, and the only way to make it work is to build a four-question buffer. The buffer is the candidate's permission to spend two minutes and forty seconds on a difficult CR question without panicking, because the easy RC detail question that follows can be answered in 45 seconds.

The way to build the buffer is in the second half of block 2, after the candidate has internalised the four-step CR protocol and the three-pass RC read. The training is mechanical: a daily 30-minute drill where the candidate answers as many questions as possible from a fresh pack, tracks time per question, and looks for any family that is consistently running over two minutes. The family that runs over is the family that needs another week of work, not more raw practice.

Endurance is its own variable. A candidate who can sustain 90-second-per-question reading for 15 minutes often collapses at minute 35, when attention drifts and re-reading starts. The fix is a 5-question warm-up before every timed Verbal set, even in the early weeks. The warm-up is not scored, and the candidate's first five answers of the day are not timed. By block 3, the candidate should be doing a 10-question warm-up followed by a 23-question set, with a 10-minute rest in the middle if the candidate's endurance is still shaky.

Test-day simulation is the last layer. In weeks 10 to 12, the candidate takes two full-length GMAT Focus mocks, including the Quant and Data Insights sections, under realistic conditions. The purpose is not to chase a high score; the purpose is to confirm that the Verbal section can be paced from question 1 to question 23 without a late-section collapse. If the candidate's score on questions 18 to 23 is consistently lower than on questions 1 to 5, the plan needs another week of endurance work, not another week of content.

Common pitfalls and how to avoid them

The pitfalls below are the ones I would flag for a low-B2 candidate before week one, because each one has cost candidates I have worked with weeks of wasted time.

Pitfall 1: starting CR drills in week one. The candidate is bored with the reading work and wants to "do real GMAT questions." The problem is that without the English engine rebuilt, every CR drill is paying a 30-second-per-question tax for the reading, on top of the argument logic. The plan delays CR drills to week 5 on purpose.
Pitfall 2: memorising word lists without example sentences. A candidate who can recognise mitigate in a multiple-choice list but cannot use it in a sentence is not ready for a CR question that uses mitigate as the key verb. The example sentence is the test of ownership.
Pitfall 3: skipping the second read. Many low-B2 candidates read an RC passage once, feel they understood it, and move to the questions. The second read is where the comprehension becomes reliable. Skipping it costs the candidate 3 to 5 points per Verbal section.
Pitfall 4: scoring mocks daily. A daily mock produces a daily score that bounces by 5 to 7 points, and the candidate panics. The plan calls for a scored mock only at the end of each four-week block, with mid-block work measured by sub-skill drills.
Pitfall 5: chasing a target score without checking the underlying level. A candidate who needs V80 to be competitive at a target programme, but whose real level is V60 with weak English, will not reach V80 by doing more CR drills. The English engine has to move first. The gates exist to surface this.

Sample week-by-week allocation for the 12-week plan

The table below summarises the daily and weekly rhythm of the plan. The candidate can adjust the time blocks by ±15 minutes, but the shape — reading and vocabulary every day, item-family work three to four days a week, one mock per block — should stay constant. A candidate who skips the daily reading and tries to compensate with extra CR drills will arrive at week 12 with sharper argument logic and a slower reader. The two tracks have to advance together.

Block	Daily reading and vocabulary	Item-family drills	Mock / diagnostic	End-of-block gate
1 (weeks 1–4)	40 min reading, 45 min vocabulary/grammar, daily	None in week 1; sentence deconstruction in weeks 2–4	Mini-mock of 10 RC questions at end of week 4	RC accuracy ≥ 60% on the mini-mock
2 (weeks 5–8)	30 min reading, 40 min vocabulary/grammar, daily	5 RC passages + 3 CR drills, 4 days/week	Full 23-question Verbal at end of week 8	Verbal scaled score ≥ 5 points above the baseline mock
3 (weeks 9–12)	20 min reading, 30 min vocabulary/grammar, daily	Targeted drills on weak families identified in week 9	Two full GMAT Focus mocks, weeks 10 and 12	Verbal scaled score at or above the candidate's target band, with stable pacing on questions 18–23

Most candidates who follow this plan move from a low-B2 starting point to a Verbal band in the mid-V70s by the end of week 12, with the strongest gains in Reading Comprehension. Critical Reasoning moves more slowly, because argument logic is its own skill, and a candidate who was already reasoning well in their first language will see CR gains earlier than a candidate who was not. The plan is also deliberately conservative on its target. The candidate who can sustain a V76 on a mock is, in my experience, ready for test day. The candidate who is still oscillating between V72 and V80 should defer the test by two weeks and run block 3 again, because the oscillation is usually a sign of an unstable pacing plan, not a weak ceiling.

Finally, a piece of advice that I would give any low-B2 candidate sitting down to plan: the English engine is the multiplier. Every minute spent rebuilding it in weeks 1 to 4 pays back two or three minutes in weeks 5 to 12. The candidate who resists the slow start and jumps to "real" GMAT work will spend the same total hours and arrive at a lower score. The plan above is built around that trade-off, and the gates are the mechanism that keeps the candidate honest about whether the English engine is actually moving.

TestPrep Europe's diagnostic Verbal assessment is a natural starting point for low-B2 candidates who need a clearer picture of where their reading speed, lexical precision, and argument-mapping skills actually sit before the 12-week plan begins.

Frequently asked questions

How long does a low-B2 candidate realistically need to prepare for GMAT Verbal?

In my experience, a candidate starting at low-B2 needs 12 to 16 weeks of structured preparation to reach a stable mid-V70s score, assuming roughly 15 to 18 hours of study per week. The English engine rebuild in the first four weeks is non-negotiable; candidates who try to compress it into two weeks almost always stall at V65 and lose another month to plateaus.

Should a low-B2 candidate start with Critical Reasoning or Reading Comprehension on the GMAT Focus?

Reading Comprehension first. CR is built on argument logic, but on the GMAT Focus the logic is wrapped in academic English that a low-B2 reader will mis-decode. RC drills force the candidate to slow down, summarise, and re-read — habits that transfer directly to CR in weeks 5 to 8. A candidate who starts with CR tends to develop the habit of skimming, which is the single most damaging habit a low-B2 reader can build.

What Verbal score should a low-B2 candidate realistically target on the GMAT Focus?

V76 to V84 is a realistic 12-week target for a low-B2 candidate who follows the plan above. V85 and above is reachable but usually requires another 4 to 6 weeks, and the marginal effort is best spent on reading speed and lexical precision rather than on extra question drills. A candidate whose target programme requires V85+ should plan for 18 to 20 weeks, not 12.

Is it worth paying for an English course alongside GMAT Verbal preparation?

Yes, for most low-B2 candidates. Two hours a week of structured English instruction — ideally with a teacher who has worked with non-native test takers — accelerates the reading-and-decoding track more efficiently than self-study. The teacher does not need to know the GMAT; the candidate's job is to apply the teacher's reading skills to GMAT-style passages during the week.

How can a low-B2 candidate tell whether they are plateauing on Verbal versus improving slowly?

Look at the sub-skill data, not the section score. If RC accuracy is climbing and CR accuracy is flat, the candidate is improving on the family that the English engine first unblocks, and the CR family will follow once the underlying reading speed stabilises. If both families are flat across two consecutive gates, the candidate has hit a real ceiling and the plan needs a structural change — usually a heavier reading-and-vocabulary block, not more drills.