What Is the AD/BCE Method for GMAT Data Sufficiency?
AD/BCE is a two-pass elimination framework for Data Sufficiency questions: evaluate Statement (1) alone first, decide whether it splits the five answer choices into {A, D} or {B, C, E}, then evaluate Statement (2) alone and narrow further. Every Data Sufficiency question on the GMAT Focus Edition uses the same five fixed answer choices, so the method works identically on every single one:
| Choice | Meaning |
|---|---|
| A | Statement (1) ALONE is sufficient, but statement (2) alone is not sufficient. |
| B | Statement (2) ALONE is sufficient, but statement (1) alone is not sufficient. |
| C | BOTH statements TOGETHER are sufficient, but NEITHER statement ALONE is sufficient. |
| D | EACH statement ALONE is sufficient. |
| E | Statements (1) and (2) TOGETHER are NOT sufficient. |
(Wording per GMAC's own sample Data Sufficiency question on mba.com.) Because the choices are fixed, you never need to read them on test day — write "AD / BCE" on your scratch pad before you even look at the statements, then cross off whichever half doesn't survive each statement. It's the same discipline Simon Flynn has drilled into candidates across 15+ years of GMAT coaching (see his background): isolate, don't skip steps, re-check before you commit.
How Do You Actually Run AD/BCE on a Question?
Work Statement (1) in total isolation — don't let your eye drift to Statement (2) yet.
- Statement (1) sufficient → you're now choosing between A and D. Cross out B, C, E.
- Statement (1) NOT sufficient → you're now choosing between B, C, E. Cross out A, D.
Then evaluate Statement (2) alone, using only the surviving half:
- If you crossed to {A, D}: does Statement (2) alone also work? Yes → D. No → A.
- If you crossed to {B, C, E}: does Statement (2) alone work? Yes → B. No → move to the final step: do the two statements together settle it? Yes → C. No → E.
Worked example. Is ?
- (1)
- (2)
Statement (1): means or — two possible answers to "is ?" (yes and no). Not sufficient. Cross out A and D; you're now in {B, C, E}.
Statement (2): has exactly one real solution, , because cube roots (unlike square roots) don't produce a pair. That's a definitive answer: is not greater than . Sufficient alone. Answer: B.
The mechanism worth internalizing: an even-power equation (, …) almost always needs a sign check, because two different values satisfy it. An odd-power equation (, …) has a unique real root, so it resolves sufficiency on its own far more often. Test-takers who don't separate these two cases lose points to the same error repeatedly.
What Is the C-Trap, and Why Do Strong Test-Takers Fall Into It?
The C-trap is picking C — "you need both statements" — when the real answer is D, because one statement was actually sufficient on its own and you stopped checking too early after spotting surface-level ambiguity.
Worked example. What is ?
- (1)
- (2)
Statement (1): means or . Most test-takers stop right there, see two possible values of , and mark it insufficient. But the question doesn't ask for — it asks for . And either way. The sign ambiguity that felt disqualifying is irrelevant to the actual question being asked. Statement (1) is sufficient alone.
Statement (2): directly gives . Also sufficient alone.
Correct answer: D — each statement alone is sufficient. A test-taker who reflexively treats " could be positive or negative" as automatic insufficiency, without re-checking it against what the question actually asks, will mark C (or even E) and lose the point.
The fix: every time a statement produces more than one possible value, re-read the question stem before ruling the statement out. Ask specifically: does the ambiguity survive into the thing being asked about? If the question asks for , , or a range condition, sign or magnitude ambiguity in itself often washes out.
Why Does an Answer of "No, Definitely Not" Still Count as Sufficient?
This is the second most common Data Sufficiency error, and it's a framing mistake, not a math mistake: a statement that lets you answer "No" with certainty is just as sufficient as one that lets you answer "Yes." Sufficiency means the statement settles the question one way or the other — the question never has to resolve in the affirmative.
mba.com's own Data Insights prep guidance makes this the first strategy point for Data Sufficiency: "Decide whether the problem allows only one value or a range of values. Remember: you are only determining whether you have enough data to solve the problem" — not what that data actually says.
Worked example. Is between 5 and 15?
- (1)
- (2)
Statement (1): . Is 8 between 5 and 15? Yes, definitively. One value, one answer. Sufficient.
Statement (2) looks like it gives more information — a whole range instead of one number — but that's the trap in the other direction. could be 4 (not between 5 and 15) or could be 10 (between 5 and 15). Two different answers to the same question. A wide range that straddles both a "yes" region and a "no" region is not sufficient, no matter how much it narrows down in absolute terms. Not sufficient. Answer: A.
The lesson runs in both directions: a single value is sufficient if it gives one definite answer (even a "no"), and a range is sufficient only if every value inside it produces the same yes/no answer — not just a smaller range.
What Other Data Sufficiency Traps Does GMAC Flag Directly?
mba.com's official Data Insights prep-strategy page names one more trap specific to Data Sufficiency: geometry figures. Its guidance is explicit — "Avoid making unwarranted assumptions based on geometric figures. Figures are not necessarily drawn to scale." A triangle that looks equilateral on screen might not be; an angle that looks like 90° might be labeled unknown. Any conclusion you draw from how a figure looks, rather than from the values and relationships explicitly stated, is not valid data — treat unlabeled visual proportions as noise.
Quick-Reference: The Full AD/BCE Decision Grid
| Statement (1) alone | Statement (2) alone | Together | Answer |
|---|---|---|---|
| Sufficient | Sufficient | — | D |
| Sufficient | Not sufficient | — | A |
| Not sufficient | Sufficient | — | B |
| Not sufficient | Not sufficient | Sufficient | C |
| Not sufficient | Not sufficient | Not sufficient | E |
Where Does Data Sufficiency Fit on the GMAT Focus Edition?
Data Sufficiency is one of five question types inside the Data Insights section, alongside Multi-Source Reasoning, Table Analysis, Graphics Interpretation, and Two-Part Analysis. Per mba.com's official exam-content page:
- Data Insights: 20 questions, 45 minutes, on-screen calculator allowed.
- Quantitative Reasoning: 21 questions, 45 minutes, no calculator.
- Total exam: 64 questions across three sections in 2 hours 15 minutes, with one optional 10-minute break and test-taker-chosen section order.
The calculator distinction matters strategically: because Data Insights (and therefore Data Sufficiency) allows a calculator, arithmetic itself is rarely the bottleneck there the way it can be in Quant. The skill being tested is judgment about what counts as enough information — exactly the AD/BCE and C-trap mechanics above — not raw computation speed.
Data Sufficiency questions recur throughout the section, so a systematic elimination habit pays off repeatedly, not just on one question. Simon Flynn's GMAT Cohort coaching program works through exactly this kind of error pattern live, statement by statement, against real GMAT-level problems until it's automatic under time pressure.