How to read a GMAT Focus Data Sufficiency prompt in 30 seconds

GMAT Focus Data Sufficiency is the only question family on the GMAT where the answer letter is not the solution to the question — it is a verdict on two statements. Each item presents a question stem, then offers two statements labelled (1) and (2). Your job is not to compute the answer; your job is to decide whether the data in the statements, taken alone or together, is enough to determine a single, unique answer. The answer choices are always the same five: A through E, encoding the five logically distinct ways sufficiency can resolve. Mastering this section means mastering the five-option grammar and the underlying logic of what "sufficient" means under test conditions.

On the GMAT Focus Edition, Data Sufficiency appears in the Quantitative section alongside Problem Solving. The total Quantitative time budget is 45 minutes across 21 questions, mixed in an undisclosed order. In practice, most candidates see roughly 8 to 10 sufficiency items per attempt, though the actual count is adaptive and not announced. Scoring is on the same 60–90 scale as the rest of Quantitative, and a single bad stretch of sufficiency items can drag the section score by 10 points or more. The good news is that the question type rewards pattern recognition more than computation. Once the five-option grammar is internalised, the work becomes a question of which statements to test first and how to recognise a definitive answer without solving the entire problem.

The five-option grammar that controls every sufficiency answer

Every GMAT Focus Data Sufficiency item ends with the same five lettered choices. The letters themselves are the entire game. If you memorise the grammar, you stop re-reading the choices mid-problem, which is one of the most common time sinks I see in timed sets. The grammar runs as follows.

Choice A: statement (1) alone is sufficient, but statement (2) alone is not. Choice B: statement (2) alone is sufficient, but statement (1) alone is not. Choice C: both statements together are sufficient, but neither alone is. Choice D: each statement alone is sufficient. Choice E: even both statements together are not sufficient.

That is the whole code. The question stem almost never contains the choice list — the choices are always printed beneath the stem — and the stem's wording only sets up the math. Once you have read the stem, your first internal question is: "What answer would I be looking for?" — that is, the target quantity or fact the question is asking about. Then you test statement (1), then statement (2), then their union, in that order. Reading the prompt this way is faster than reading the choices first, which is what most unprepared candidates do.

A useful habit is to write the target quantity on your scrap paper in shorthand. If the stem asks "What is the value of x?", you write "x = ?"; if it asks "Is x positive?", you write "x > 0?". This single line tells you what sufficiency means in this particular problem. "Sufficient" for "x = 7" means a single numeric answer, not a range. "Sufficient" for "x > 0" means a sign determination, not a value. Many wrong answers come from confusing these two — finding a value when only a sign is required, or vice versa.

The two yes/no traps hidden in the grammar

Data Sufficiency splits naturally into value questions and yes/no questions. Value questions ask for a specific number; yes/no questions ask whether some condition holds. The grammar is identical, but the workload is different. For a value question, sufficiency means the statements force exactly one value. For a yes/no question, sufficiency means the statements force a definite yes or a definite no — never a mixture. Candidates frequently pick E on yes/no items when the statements actually force one outcome consistently; the two possible branches both point to "yes", or both point to "no", and that is a definitive answer. Always check both branches of any case that emerges.

Reading the prompt: separating setup from question

Every sufficiency item has three parts: a setup paragraph of facts, a question, and two statements. The setup is dense and often contains the only piece of information that is easy to misread. A common tutor observation is that the single largest error category on Data Sufficiency is not arithmetic — it is misreading a constraint. The prompt will say something like "x and y are positive integers" and the candidate treats them as real numbers; or it will say "x is the remainder when y is divided by 7" and the candidate treats x as unrestricted. The five options cannot help you if the underlying constraint is wrong.

Here is a working protocol I use with most candidates. Underline, on the screen, three things: every named variable, every numeric fact, and every condition word. Condition words are the small ones that change everything: "positive", "integer", "distinct", "consecutive", "remainder", "divisible", "greater than", "at most". Each of these restricts the domain of the variable. If you ignore "distinct" in a combinatorics stem, you double-count. If you ignore "positive" in a square-root stem, you admit a negative root. The condition words are the first place test-writers plant traps.

After underlining, rewrite the question in one short sentence. "What is the value of x?" "Is triangle ABC equilateral?" "What is the ratio of a to b?" The one-sentence form is what you will test against. If the one-sentence form itself is unclear, the rest of the work is wasted. In my experience, candidates who can paraphrase the question correctly in under 10 seconds almost always finish the section; those who cannot paraphrase usually cannot finish.

Worked micro-example: a value prompt

Stem: "What is the value of 2x + 3y?" Setup: "x and y are integers." Statement (1): "x = 5." Statement (2): "y = 7." Here the target quantity is 2x + 3y. Testing (1) alone: knowing x = 5 leaves y undefined, so 2x + 3y cannot be determined. Not sufficient. Testing (2) alone: knowing y = 7 leaves x undefined. Not sufficient. Together: x = 5 and y = 7 give 2(5) + 3(7) = 31. Sufficient. The answer is C. Notice that the math itself is trivial; the only skill is correctly applying the grammar.

Testing statement (1) alone, then (2) alone, then together

Sufficiency is always tested in a fixed order: (1) alone, (2) alone, then both. The reason for this order is not stylistic — it is logical. A statement that is sufficient alone is sufficient in combination with anything else, so once a statement proves sufficient alone, you do not need to test the other. Conversely, a statement that is insufficient alone might still become sufficient when joined with the other, so you cannot stop early on insufficiency. The decision tree, in plain text, looks like this: if (1) alone is sufficient, pick A and stop; else, if (2) alone is sufficient, pick B and stop; else, test (1) and (2) together — if together sufficient, pick C; else, if (1) alone was actually sufficient (revisit), pick D; else, pick E.

Choice D deserves a second look. D means each statement alone is sufficient, and it is the rarest of the five answers in well-designed items. When D appears, both statements, taken independently, must each pin down a unique answer. The most common D traps involve statements that look symmetric — say, two different ways of expressing the same information — but where one of them is slightly weaker than the candidate realises. For example, statement (1) might say "x and y are positive integers with x + y = 10" and statement (2) might say "x and y are positive integers with xy = 21". Both alone force the same pair {3, 7}, so D is correct. If, instead, statement (2) said "x and y are positive integers with xy = 16", then (2) would still be insufficient because {2, 8} and {4, 4} are both possible, and the answer would drop to C.

Common pitfalls and how to avoid them

Three pitfalls account for most lost points. First, candidates test both statements together before testing each alone, which is logically wrong and loses the A, B, and D options. Second, candidates stop after finding a single case that "works" and assume sufficiency, ignoring whether other cases might also work. The rule is: a value question is sufficient only if the case you found is the only case possible. Third, candidates confuse a definite yes with sufficiency on yes/no items: if a statement rules in a "yes" in one branch and a "no" in another, it is not sufficient. Mark a clear "yes / no / both?" tag on your scrap paper for every yes/no stem.

Recognising the seven most common sufficiency patterns

Sufficiency items reuse a small set of underlying math structures, and once you can name the structure, the work becomes procedural. The seven patterns below cover roughly four out of five items a candidate will see on the GMAT Focus Quantitative section.

• Linear system with two unknowns. A classic: "What is the value of x?" with two equations in x and y. Two independent linear equations are sufficient together (C). One equation is not sufficient (insufficient alone).
• Quadratic with restricted sign. "What is the value of x?" where the equation is quadratic and the stem says x is positive. Knowing the roots and the sign is sufficient; knowing only the roots is not, because the negative root is still a candidate that the sign must rule out.
• Geometric figure with one missing dimension. A triangle's area plus its base uniquely determines its height, but base and perimeter together usually do not. Watch for whether the additional data is independent or merely a restatement.
• Rate–time–distance setup. A single rate and a single time gives distance; distance and time give rate; distance and rate give time. Two independent pieces are enough; one is not.
• Combinatorics with a fixed total. "How many committees of 3 can be formed?" If the total pool size is known, the answer is unique. If only the size of the committee is known, the pool is still missing.
• Percent change with one of the two values. A 20 percent change of an unknown starting value cannot be pinned down; a 20 percent change from a stated starting value, however, gives a unique result.
• Inequality with a missing endpoint. "Is x greater than 5?" Knowing x is between 4 and 7 is not sufficient; knowing x is between 6 and 9 is sufficient (always yes). Yes/no items especially reward careful boundary thinking.

Notice that almost every pattern reduces to a single idea: do the statements, alone or together, leave exactly one possible value (or one definite yes/no)? If yes, the statement is sufficient for that question. The math differs; the logic does not.

How to triage sufficiency items inside the 45-minute section

On the GMAT Focus Edition, Quantitative gives you 45 minutes for 21 questions, of which approximately 8 to 10 are sufficiency items by typical adaptive distribution. That works out to roughly 2 minutes and 8 seconds per question across the section, but the real allocation is not uniform — easy items at the start of the section should take 60 to 90 seconds, and hard items at the end can absorb 3 to 4 minutes. Sufficiency items live in the middle of that range. A well-prepared candidate spends about 1 minute 30 seconds on an easy sufficiency item and 2 minutes 30 seconds on a hard one.

The single highest-leverage time move is to commit to a testing order. Test (1) alone, then (2) alone, then both, in that order — every time, without exception. Candidates who freestyle the testing order end up solving the same statement twice, which is the most common reason a 90-second item stretches to four minutes. If the testing order is fixed, the answer letter falls out of the work, and you rarely need to consult the printed choices at all.

The second time move is to recognise the "killer statement". A killer statement is one that, taken alone, makes the other statement irrelevant. For example, "x = 7" by itself pins down a unique value of x; once you see it, you can stop reading the rest of the problem. Conversely, a "do-nothing statement" is one that gives no new information — a restatement of the stem, or a piece of data implied by the setup. Do-nothing statements are not necessarily insufficient; they are simply statements you can mentally compress to "(1) gives nothing new" and move on.

A 30-second prompt-reading protocol

Here is the protocol I teach in the first diagnostic session. Read the stem and underline variables, numerics, and conditions. Paraphrase the question in one short sentence and write the target quantity on your scratch paper. Glance at statement (1) and decide: sufficient, or not? If sufficient, the answer is A — pick and move on. If not, glance at statement (2): sufficient, or not? If sufficient, the answer is B. If not, test both together: sufficient, or not? If yes, C. Then, only if you have time, ask whether (1) was actually sufficient on second look — if yes, D; if no, E. This protocol reliably lands candidates inside the time budget on the first pass.

What "sufficient" actually means under test conditions

Outside the exam, "sufficient" is a fuzzy word. On the GMAT Focus, it has a precise meaning: a statement is sufficient if and only if it determines, on its own, a single value (for value questions) or a single yes/no outcome (for yes/no questions). The phrase "on its own" is the crucial one. A statement is sufficient alone if you could answer the question using only that statement and the stem, ignoring the other statement entirely. If you need the other statement to disambiguate a case, the first statement is not sufficient alone.

This precision matters because the five options do all the disambiguation for you. Candidates who try to invent a sixth, intermediate category — "almost sufficient", "barely sufficient", "sufficient in spirit" — are essentially refusing to use the answer key. The grammar does not have those categories. Either the statement forces a unique answer, or it does not. Train yourself to ask the binary question: "Is there exactly one possible answer, or more than one?" If exactly one, sufficient. If more than one, not sufficient.

Yes/no questions in particular

Yes/no sufficiency is the area where most candidates lose points without realising it. The pattern is that the statement produces two cases, and the candidate checks only one. For instance, a stem might say "Is x positive?" and a statement might say "x² = 16". The candidate thinks "x = 4, so yes, positive — sufficient". The correct move is to notice that x = -4 is also a solution, and under x = -4 the answer is no. Because the statement produces both a yes branch and a no branch, the statement is not sufficient. The same trap appears with quadratics, with absolute values, and with any statement that admits a sign flip. Always list both branches before deciding on yes/no sufficiency.

Practising sufficiency: a four-week build

For candidates who have roughly four to six weeks before the exam, the following build is what I would use. It front-loads grammar, then drills patterns, then mixes timed and untimed practice. The structure matters more than the total number of items; 200 items done well beats 800 items done superficially.

The table above is a rough scaffold; the absolute counts matter less than the mode of practice. In week 1, the goal is to make the grammar automatic. By the end of week 1, the candidate should be able to read any sufficiency stem and produce an answer letter without looking at the printed choices. In week 2, the goal is pattern fluency. In week 3, the goal is yes/no discipline. Week 4 is where everything comes together under timed conditions.

Common error log: what to track after each practice set

After every set, the only thing worth tracking is the reason for each miss. A miss because the math was too hard is one thing; a miss because the testing order was wrong is another; a miss because the condition word "distinct" was missed is a third. These three categories need different remedies. Math hardness requires more pattern drill. Wrong testing order requires re-anchoring on the protocol. Missed condition words requires slower reading on the prompt.

A useful format is a three-column log: question ID, what you picked, and the true reason for the miss in one short phrase. After 100 logged items, the categories cluster. In my experience, two categories tend to dominate: "I tested both statements together first" and "I missed a condition word in the prompt". The first is a protocol slip; the second is a reading slip. Both are addressable in a week.

How sufficiency interacts with the rest of the GMAT Focus score

The Quantitative section is reported as a single score on the 60–90 scale, so Data Sufficiency is not graded as its own number; it is folded into Quantitative alongside Problem Solving. Within the section, the adaptive algorithm does not tell you how many sufficiency items you will see, but a strong sufficiency performance does push the algorithm to deliver harder items later, which is where the section score moves. A candidate who is competent on sufficiency but weak on word problems can still reach the 80+ range if the sufficiency items in the high-difficulty band are clean.

The reverse is also true. A single late-section sufficiency item — the kind with a long stem, a geometric figure, and an embedded condition word — can be the difference between a 79 and an 81 on the Quantitative scale. The lesson is that sufficiency is not optional practice. It is one of the two pillars of the Quantitative score, and a 4-week preparation plan that allocates only a handful of hours to it is structurally underweight.

Putting it together: a worked sufficiency item, end to end

Let us walk a full item at the pace a strong candidate would use, with the protocol active. Stem: "In a certain office, every employee speaks either English or Spanish, and some speak both. What is the number of employees who speak both languages?" Setup: "The office has 40 employees. 30 speak English and 25 speak Spanish." Statement (1): "No employee speaks more than two languages." Statement (2): "Every employee speaks at least one of the two languages."

Paraphrase the question: "How many speak both?" Target quantity: the overlap count, B. By inclusion–exclusion, B = E + S − T, where E is English-speakers, S is Spanish-speakers, and T is the total. E = 30, S = 25, T = 40, so B = 30 + 25 − 40 = 15. But the candidate should not compute that yet — the protocol is to test the statements.

Test statement (1): "No employee speaks more than two languages." This is already implied by the setup ("every employee speaks either English or Spanish, and some speak both"). Statement (1) gives no new information. The candidate can immediately rule out A and D on the strength of (1) being a do-nothing statement. Move to statement (2): "Every employee speaks at least one of the two languages." This is also already implied by the setup. Statement (2) is a do-nothing statement as well. Both statements together still give no new information, so the answer is E. Note that the math in the stem is enough to compute B = 15 directly — but the test is not whether the stem suffices. The test is whether the two statements add information, and they do not. A is a trap here, and E is correct.

Why this worked

Three things made the protocol fast. First, the candidate paraphrased the question and identified the target quantity (the overlap) before touching the statements. Second, the candidate recognised both statements as restatements of setup, which is a recurring sufficiency pattern — the do-nothing statement. Third, the candidate did not compute B = 15 in the stem and use that as the answer. Computing the stem answer is a common error mode precisely because it makes E feel wrong. The discipline is: the answer letter comes from the statements, not the stem.

Closing the loop on sufficiency preparation

Sufficiency is a learnable skill, and the GMAT Focus Edition's adaptive format means the rewards for cleaning it up are concentrated in the high-difficulty band where the section score is won. The path forward is to internalise the five-option grammar, fix the testing order at (1) alone, (2) alone, then both, and learn to read condition words in the prompt. A two-week drilling block, even on top of a busy schedule, moves most candidates up by at least one answer band. TestPrep İstanbul's diagnostic on the Data Sufficiency prompt-decode protocol is a natural starting point for candidates building a sharper preparation plan.

Frequently asked questions