Why pacing on the GMAT fails

The GMAT Focus, like the classic GMAT before it, is a timed examination in which pacing is a measurable sub-skill. Candidates who reach question 18 of 23 in a section and realise they have only four minutes left are not facing a content problem; they are facing a pacing diagnosis. The fix is rarely "study harder" and almost always "redesign the time budget". This article walks through what to do when the clock runs out, before it runs out, and immediately after, on the Quant, Verbal, and Data Insights sections of the GMAT Focus Edition.

Diagnosing the timing failure: was it one slow item or a structural leak?

Before changing anything in your preparation strategy, a candidate who keeps running out of time on the GMAT needs to classify the failure mode. In my experience, almost every timing collapse on a timed examination like this one falls into one of two families, and the remedy is different for each.

The first family is the single-item trap. A candidate meets a hard Data Sufficiency stem or a tangled Critical Reasoning argument around question 7 or 8, decides to "just figure it out", and burns four minutes. The remaining 15 items now have 16 minutes instead of 20, and the section slowly bleeds out. The fix is not faster maths or better logic; it is a pre-committed rule about how long any single GMAT item is allowed to take. Most tutors working with serious candidates settle on a soft cap of roughly 2 minutes for Quant and Data Sufficiency, and slightly under 2 minutes for Verbal, before a forced skip is mandatory.

The second family is the structural leak. No single item is slow, but the candidate is doing 10 seconds of extra work on every item: rewriting the question, redrawing the figure, re-reading the passage, re-entering the calculator. Ten seconds across 23 items is almost four minutes, which is roughly 4–6 lost correct answers on a section scored on a 0–60 scale. This is the more common failure pattern at the V76 / Q78 / DI78 band, because the items themselves are not the bottleneck, the workflow is.

To diagnose which family you are in, pull the last two full-length GMAT Focus mock results. If your times look like 0:55, 1:05, 1:20, 6:40, 1:10, 1:15 … you have a single-item trap. If they look like 1:40, 1:45, 1:50, 1:55, 1:50, 1:55 … you have a structural leak. The triage plan differs. The single-item trap is cured by practising the skip-decision. The structural leak is cured by removing ten seconds from every item, which is the harder and more interesting problem to solve.

Three quick diagnostic numbers to record

Median time per item, broken out by section and by question position (1–6, 7–12, 13–18, 19–23).
Number of items you skipped, guessed, or left blank.
Number of items where you changed your final answer in the last 20 seconds.

These three numbers, kept in an error log, will tell you within a week whether pacing is your actual bottleneck or whether timing stress is masking a content gap. They are also the numbers I would ask any candidate to bring to a diagnostic session before I prescribed a preparation plan, because timing remediation is expensive in hours and only worth the spend if the diagnosis is right.

Building a minute-per-question budget that survives contact with a hard item

The standard time budget on the GMAT Focus is roughly 1 minute 55 seconds per item across a 23-item section, with a small buffer for the tutorial and the section review screen. That number is correct on paper and almost useless in practice, because it assumes an even distribution of difficulty, and the exam is built to violate that assumption. The Quant section is adaptive and contains items calibrated to find your ceiling; Verbal and Data Insights place denser items in the second half of the section. A flat budget fails at exactly the moments you can least afford to fail.

A more honest budget assigns time to phases of the section, not to individual items. For the 23-item Quant section, a working template looks like this:

Items 1–6: target 1:30 per item, 9 minutes total. These are the warm-up items, the section is still finding your level, and over-investing here is the most common Quant timing mistake.
Items 7–14: target 2:00 per item, 16 minutes total. This is the working middle where the bulk of the score is determined.
Items 15–23: target 2:10 per item, 20 minutes total. Allow a slight inflation here because the harder items live in this range.

The total comes to 45 minutes with a 1-minute buffer, which matches the official section length. Verbal and Data Insights follow the same shape, with one adjustment: Verbal passages take a fixed upfront cost, so the per-item budget for the first 4–5 items in a passage set is closer to 2:30, and the per-item budget on the shorter Verbal items (Critical Reasoning, Sentence Correction) drops to roughly 1:30. Data Insights has the widest spread because the section mixes 2-minute Data Sufficiency-style items with 3–4 minute Multi-Source Reasoning items. There, the budget is built around item families rather than around positions.

Why a phase budget beats a per-item budget

A per-item budget tempts the candidate to police the clock on every single question, which is mentally expensive and creates its own form of timing stress. A phase budget lets the clock absorb variance inside the phase, which is how the examination is actually designed. If a candidate is 90 seconds over budget by item 9, the phase budget permits a 15-second compression on items 10–14 to recover, with no individual item feeling rushed. If a candidate is 90 seconds under budget by item 9, the phase budget lets them spend the surplus on a careful re-read of a hard item at position 16. This is the difference between pacing as arithmetic and pacing as decision-making.

The skip-decision: when to give up on a GMAT item and protect the rest of the section

The single highest-leverage skill a slow test-taker can learn is the skip-decision. It is also the skill most candidates refuse to learn, because it feels like giving up. It is not. On a 23-item scored section, a candidate who spends five minutes on a single Data Sufficiency item and then gets it wrong has not lost one question, they have lost the next three questions as well, because those items were attempted in a panic. The skip-decision converts a guaranteed 4-point loss into a probable 1-point loss, which is a strictly better trade on a scored examination.

The skip-decision needs three components: a time trigger, a marking mechanism, and a return rule. The time trigger is the soft cap mentioned earlier, roughly 2 minutes for Quant, 2 minutes for Data Sufficiency, 1:45 for a Critical Reasoning or Sentence Correction item, and 2:30 for a Reading Comprehension item. The marking mechanism is the built-in flag function on the GMAT, which exists for exactly this purpose; using it well is the difference between a flag as a wish and a flag as a triage tag. The return rule is the contract: if you skip at item 9 with a flag, you may return to it after item 15, and only if the section timer says you have the surplus.

In practice, the return rule rarely fires. Most flagged items are abandoned because the candidate spent two minutes on them, did not converge on an answer, and the cost-benefit of another two minutes on a flagged item versus two minutes on a fresh, easier item is almost never favourable. Knowing this in advance is what makes the skip-decision psychologically tolerable. Candidates who resist skipping tend to have an internal rule that says "leaving an item blank is failure". The honest reframing is: leaving an item blank is a 20% expected value on that single question; staying and panicking is a 20% expected value on that question and the next two. Pick the 20%.

How to mark items so the review screen actually helps

The GMAT Focus review screen shows which items you flagged, which you left blank, and how much time you spent on each. Treat it as a per-item receipt. The convention I recommend to every candidate I work with is simple: flag only items where the skip-decision triggered on the time cap, not items where you simply hesitated. A flag that means "I hesitated" is noise; a flag that means "I spent 2:00 and had no convergence" is signal. After the section, the review screen should let you see, at a glance, how many forced skips you had and where they clustered. If they cluster in the first half, your phase budget is too generous at the front. If they cluster in the second half, your early pace is too slow.

Removing ten seconds from every item: the structural-leak fix

For candidates whose median time per item is already inside 2:00 but who still run out, the work is in workflow. Ten seconds saved per item, across 23 items, is roughly 4 minutes, which is enough to add 2–3 correct answers at the margin. The workflow changes that reliably produce this saving fall into three buckets: reading, drawing, and entering.

Reading. The first read of a GMAT item is for structure, not for content. Candidates who read once for content often re-read the stem, re-read the stimulus, and re-read the answer choices, which is three passes for the price of one. The fix is a one-pass read with three questions in mind, in this order: what is the question actually asking, what is the unit or scale, and what is the constraint. For Reading Comprehension, this becomes a passage-mapping pass rather than a content read. For Data Sufficiency, the structured read is a two-pass protocol: pass one extracts the question and the data, pass two evaluates each statement against the question.

Drawing. Many Quant and Data Insights items reward a clean diagram: a number line for inequalities, a Venn diagram for set problems, a tree for probability, a quick sketch for geometry. Candidates who draw in detail spend too long; candidates who skip drawing entirely waste time visualising in their head. The target is a 5-second sketch that captures the constraint, not the picture. If the diagram is taking 30 seconds, the diagram is wrong.

Entering. The on-screen calculator in Data Insights is a known timing trap. Typing "(0.0417 × 12,500) / 4" takes longer than estimating, and the candidate who insists on numerical precision on every item burns 20–40 seconds that they cannot recover. A working rule: use the calculator for division and square roots only, estimate everything else, and only confirm a numerical answer with the calculator when the estimate is close to two of the five answer choices.

A 3-item audit you can run tonight

Take any three Quant items you got wrong and time the audit. For each item, record: time spent on first read, time spent re-reading, time spent drawing, time spent calculating, time spent comparing answers. The category with the largest total is your leak. In my experience, the leak is re-reading roughly half the time, drawing or calculating roughly a third, and the first read the rest. Fixing the largest category reliably returns 8–12 seconds per item, which compounds across a section.

Section-specific pacing: Quant, Verbal, and Data Insights behave differently

Treating the three sections of the GMAT Focus as if they had the same pacing logic is one of the more expensive timing mistakes a candidate can make. They have different item counts, different item families, and different reasons that the clock tends to run out.

Quant (23 items, 45 minutes). The Quant section is adaptive, so the first 6–8 items are diagnostic and the back half is where the score is determined. The most common Quant timing failure is over-investing in the diagnostic items, then discovering the back half is calibrated harder and being short on the items that matter. The remedy is the phase budget above, and a hard rule that no item in the first 6 takes more than 1:30 even if the candidate is sure they could solve it faster with another 30 seconds. The second most common failure is calculator-style computation on items that reward estimation. The remedy is to do all simplification by hand and only use scratch work for the final arithmetic step.

Verbal (23 items, 45 minutes). Verbal pacing breaks differently. The bottleneck is the Reading Comprehension passage, which has a fixed cost regardless of how many items the passage carries. A 4-item RC passage is approximately 4 minutes of reading plus 4 minutes of items, which is 8 minutes for 4 items, or 2:00 per item. A 1-item RC passage is still 4 minutes of reading plus 1 minute of item, which is 5 minutes for 1 item, or 5:00 per item. Candidates who do not pre-budget for the passage cost run out of time on the shorter passages that appear in the second half of the section. The remedy is to time the passage, not the items, and to budget the Verbal section by passage, with the shorter passages getting a higher per-item rate.

Data Insights (20 items, 45 minutes). Data Insights has the widest variance in per-item cost, because Multi-Source Reasoning with three tabs is genuinely slower than a single-table Graphics Interpretation. The pacing fix is to triage the section by item family. Sort the section on first sight, do the cheap item families first (Graphics Interpretation, Table Analysis, Two-Part Analysis), and leave the expensive item families (Multi-Source Reasoning, Data Sufficiency with three statements) for the second pass. This sort-then-execute protocol is the single highest-leverage Data Insights timing change a candidate can make, and it is one of the few pacing moves that does not require any content work to deploy.

Comparative pacing targets across the three sections

Section	Items	Phase 1 target (items 1–6)	Phase 2 target (items 7–14)	Phase 3 target (items 15–23)	Common leak
Quant	23	1:30 / item	2:00 / item	2:10 / item	Over-investment in early diagnostic items
Verbal	23	2:00 / item (passage-heavy)	1:50 / item (mixed)	1:40 / item (CR / SC dense)	Under-budgeting for short RC passages
Data Insights	20	2:10 / item	2:15 / item	2:30 / item	Mixed per-item cost across item families

The numbers in the table are starting points, not laws. The right adjustment for most candidates who already finish in time is to leave the numbers alone and focus on workflow. The right adjustment for candidates who run out is to tighten Phase 1 by 15 seconds per item and to add the same 15 seconds to Phase 3, which protects the harder back half without changing the total.

Rehearsing the rescue: how to practise timing without practising panic

The mistake most candidates make in preparation is to take mock after mock and never isolate the timing variable. A mock taken under timed conditions measures what you did, not what you would have done with a different pacing decision, and so it does not generalise. Practising timing on the GMAT means practising decisions under artificial time pressure, and the only way to do that is to take single-section drills with the phase budget enforced.

A working rehearsal protocol has four steps. First, take a 10-item Quant drill with a 20-minute hard cap and the phase budget published on a sticky note. Second, score the drill, then mark every item that exceeded 2:00. Third, redo only the over-budget items, this time with a 2:00 hard cap and a stopwatch. Fourth, repeat the original 10-item drill the next day. The second attempt is almost always faster, and the speed-up comes from one of two places: the candidate learned a faster method for the slow items, or the candidate learned that the slow items were not worth the time on the first pass. Both are wins.

For Verbal, the rehearsal protocol is similar but the unit of practice is the passage, not the item. Take a 4-item RC passage with a 7-minute hard cap for the whole passage, including reading. If the cap is exceeded, redo the passage with a 6-minute cap. The aim is to find the reading speed that lets you finish the passage plus all four items inside the budget, and the only way to find it is to repeat the same passage three or four times under tightening caps. Repetition under a cap is what trains the skip-decision, because the candidate learns, on the third repetition, that the third sentence of paragraph two was not load-bearing and can be skimmed.

The rehearsal trap to avoid

Do not rehearse the rescue protocol on the day before the examination. The rescue protocol is a preparation plan, not a test-day move, and practicing it under fatigue the night before the exam produces a candidate who is tired, demoralised, and now convinced that timing is a hopeless problem. Rehearse the rescue in the middle weeks of preparation, when there is still time to absorb the discomfort. By test day, the rescue protocol should be a habit, not a tactic.

Test-day moves: what to do when the clock actually runs out mid-section

Despite all preparation, some candidates will reach question 19 of 23 with the section timer reading four minutes and four items remaining. At that point, the question is no longer pacing; it is triage. The triage protocol has three steps, in order.

First, do not look at the clock again. Looking at the clock on the last four items is a known timing-stress amplifier and tends to cost 5–10 seconds per item in re-orientation cost. Trust the time you have, work the items at the pace you have been working, and stop negotiating with the timer.

Second, do not change your guessing strategy. Candidates who run out of time often switch from process-of-elimination to random guessing, which is faster but yields the same expected value on items where they had no convergence. If you have been using elimination, keep using elimination. If you have been skipping, keep skipping. The pattern that got you to item 19 will get you through item 23.

Third, if the timer hits 30 seconds with an item unfinished, click the best answer you can identify and move on. The GMAT does not penalise wrong answers beyond the standard scoring, and an unfinished item is, mechanically, a wrong answer. Picking one of the remaining choices takes 2 seconds and converts the item from a guaranteed loss to a probable loss, which is the same trade the skip-decision makes earlier in the section. The rule that "leaving an item blank is failure" should be re-read at this moment as "leaving an item blank is the only outcome I can still control, and the timer has just taken that option away".

Common pitfalls and how to avoid them

Treating the diagnostic items as free time. Items 1–6 in Quant are still timed, and the habit of over-investing there is the most common reason candidates run out by item 17. Cap them at 1:30 each.
Refusing to skip. The skip-decision is the highest-leverage timing skill on the examination. Practise it during preparation, not on test day.
Re-reading the stem. A one-pass structural read saves 10–15 seconds per item. Train the read on practice items until it is automatic.
Over-drawing diagrams. A 5-second sketch beats a 30-second picture. If the diagram is taking longer than the algebra would, stop drawing.
Using the calculator for estimation. The on-screen calculator in Data Insights is a known time sink. Estimate first, confirm only when the estimate is close.
Rehearsing the rescue the night before. Triage protocols are preparation, not test-day moves. Practise them in the middle weeks.

Putting it together: a 3-week pacing rescue plan

For a candidate who has identified pacing as the bottleneck through the diagnostic in section one, a focused three-week plan is usually enough to recover 4–6 correct answers per section. The plan has three pieces, one per week, and each piece targets one of the failure modes above.

Week 1: Single-item trap. Take a 23-item Quant drill each day, with the phase budget enforced and the 2:00 per-item cap active. Mark every forced skip. After the drill, redo every forced skip with no time cap, and record whether the answer was actually attainable in 2:00. For most candidates, roughly half the forced skips were attainable, and the other half were genuine hard items. The attainable skips are the ones to drill; the genuine hard items are the ones to flag and budget for.

Week 2: Structural leak. Take the 3-item workflow audit on every wrong answer from the previous week. Pick the largest leak category, and run a 20-item drill that targets only that category. For reading, this means timed reads of Quant stems. For drawing, it means 5-second sketch drills on a small set of canonical geometry, number line, and Venn diagram templates. For entering, it means Data Insights drills with the calculator disabled except for division and square roots. The aim of week 2 is to remove 8–12 seconds per item from the workflow.

Week 3: Section-specific pacing. Run one full section per day at exam pace, with the phase budget and the per-item cap active. Quant, Verbal, and Data Insights each get two timed runs across the week. After each run, review the section for two questions: where did the skip-decision trigger, and where did the workflow leak. Enter the answers in the error log, and review the log at the end of the week. By the end of week 3, the rescue protocol should be a habit, and the test day should feel like another timed run rather than a crisis.

What to track in the error log to know it is working

Two numbers matter. The first is the number of forced skips per section, which should drop week on week. The second is the median time per item, which should also drop, but more slowly and with a floor. The floor is roughly 1:30 for a Quant item and 1:40 for a Verbal item; pushing below the floor usually means the candidate is rushing the easy items, which is its own form of timing failure. When the number of forced skips stops dropping, the candidate has reached the genuine difficulty of the section, and the work shifts from pacing to content.

Conclusion and next steps

Running out of time on the GMAT Focus is not a verdict on a candidate's ability; it is a pacing diagnosis with a known set of remedies. Classify the failure as a single-item trap or a structural leak, build a phase budget that survives contact with a hard item, practise the skip-decision until it is automatic, and remove the workflow leak that costs ten seconds on every item. Section by section, the protocol is the same shape but the item families and the bottlenecks differ, so the rehearsal has to be section-specific. Three weeks of focused work, tracked in an error log, is enough to convert a timing collapse into a paced, controlled section.

TestPrep Europe's diagnostic assessment is a natural starting point for candidates who want a personalised pacing audit and a per-section preparation plan built around the protocol above.

Frequently asked questions

How many minutes per question should I budget on the GMAT Focus?

A flat per-item budget of 1:55 is correct on paper but fails in practice, because the examination places denser items in the second half of each section. Use a phase budget instead: roughly 1:30 per item for Quant items 1–6, 2:00 for items 7–14, and 2:10 for items 15–23. Verbal and Data Insights need family-level budgets rather than position-level budgets, because Reading Comprehension passages and Multi-Source Reasoning items have a fixed cost that a flat per-item rate cannot capture.

Is it better to skip a hard GMAT question or to keep trying?

On a scored timed examination, the skip-decision is almost always the better trade once an item has exceeded roughly 2:00. Staying and panicking costs the next two or three items in the section, so a single 5-minute sink typically becomes a 4-point loss rather than a 1-point loss. Use the in-exam flag function as a triage tag, not a wishlist, and pre-commit to a return rule so the skip is psychologically tolerable.

Why do I run out of time on Data Insights but not on Quant?

Data Insights mixes item families with very different per-item costs. Multi-Source Reasoning with three tabs is genuinely slower than a single-table Graphics Interpretation item, and a candidate who does not triage by family will overspend on the expensive items early in the section and run out before the cheap items appear. The fix is to sort the section on first sight, do the cheap families first, and leave the expensive families for the second pass.

Should I use the on-screen calculator on every Data Insights item?

No. The on-screen calculator is one of the most expensive timing traps in Data Insights, and typing a long expression often takes longer than estimating the same expression. Use the calculator for division and square roots only, estimate everything else, and only confirm a numerical answer with the calculator when the estimate is close to two of the five answer choices.

How long does it take to fix a pacing problem on the GMAT?

For most candidates, a focused three-week plan is enough to recover 4–6 correct answers per section. The first week targets the single-item trap, the second week targets the structural workflow leak, and the third week runs full sections at exam pace to consolidate the habit. Track the number of forced skips and the median time per item in an error log; when both stop improving, pacing is no longer the bottleneck and content work takes over.