The GMAT Focus reports a single scaled score running from 205 to 805, and the four numbers candidates ask about most often — 555, 605, 655 and 705 — sit on very different parts of that scale. The first sits clearly below the median for admitted MBA classes, the second lands at a respectable but uncompetitive band, the third puts a candidate into the conversation at many target programmes, and the fourth opens the door at the most selective business schools. Each jump between these bands represents not just ten or twenty extra correct answers, but a shift in question-type mastery, time budgeting, and error profile. This article walks through what each score means in programme-fit terms, what kind of candidate typically lands there, and the concrete preparation moves that close the gap from one band to the next.
How the GMAT Focus scale is built and why four scores can say so much
The GMAT Focus quant section runs 0 to 60 in 1-point increments, the verbal section runs 0 to 60 in 1-point increments, and the data insights section runs 0 to 60 in 1-point increments. The final total is the sum of those three, producing the 205 to 805 composite. There is no longer a separate Integrated Reasoning or Analytical Writing Assessment, and the test adapts section by section in a short adaptive format with 21 items per section. Knowing the mechanics matters because the four headline scores of 555, 605, 655 and 705 correspond to very different distributions of correct answers across the three sections, and the skill ceiling is not the same in each.
For a candidate to land on 555, the rough shape is a quant around the high teens, verbal in the high teens, and data insights near the mid-teens. That profile usually means the test-taker is missing roughly half the items on the test, often with the heaviest damage on multi-source reasoning, two-part analysis, and the harder graphics interpretation stems. A 605 candidate is averaging around a 78% hit rate across the three sections with quant in the low 20s, verbal around 20, and data insights around 20. The 655 line typically requires section scores in the mid-20s, with quant often the strongest section and data insights the weakest. The 705 band usually requires at least one section in the high 40s, with the other two in the high 30s, meaning the candidate is missing perhaps 6 to 9 items across the whole exam.
In practice, the difference between bands is rarely about working harder; it is about working on the right handful of question types and stopping the bleeding on a specific error pattern. The 555-to-605 jump is usually about pacing and careless errors. The 605-to-655 jump is about Data Sufficiency and Multi-Tab Reasoning discipline. The 655-to-705 jump is about the last 10% of inference stems, the trickiest data interpretation items, and Two-Part Analysis. Each of those is a learnable, drillable skill set, and the score band tells you which one to attack first.
What a 555 actually signals: foundations, not intelligence
Most candidates who land on 555 are not weak students. They are usually working professionals who prepared for a few weeks, used only the official starter material, and walked into the test centre still mixing up Data Sufficiency statements, running out of time on the last four quant items, or misreading the question type on Data Insights prompts. In admissions terms, a 555 is below the median for most full-time MBA cohorts and it is below the published middle 80% range for most top-15 programmes. A candidate at this band should usually not apply with that score in hand unless the rest of the application is unusually strong — a 3.8+ GPA, a selective employer, a clear post-MBA narrative.
The diagnostic story for a 555 candidate almost always has three or four of the following items. First, untimed practice has hidden a pacing problem that surfaces only under exam conditions. Second, the candidate has not internalised the structure of Data Sufficiency, so they spend too long on each stem and confuse the two statements. Third, Multi-Tab Reasoning feels like a new item type even after studying, so the candidate reads all three tabs from scratch every time. Fourth, two-part analysis is being treated as two independent questions, which doubles the time cost. Fifth, graphics interpretation charts are read top-to-bottom rather than axis-to-axis, which costs 30 to 45 seconds per item.
The fix is rarely 'study more material'. It is a four-week reset. Week one should be a diagnostic under timed conditions, scored honestly, with every wrong item tagged by question type. Week two should be a deliberate re-drill of the four item families that carry the most weight: Data Sufficiency, Multi-Tab Reasoning, Table Analysis, and Two-Part Analysis. Week three should reintroduce timed sets of 10 mixed items per section, with a strict minute-per-question budget. Week four should be two full-length adaptive simulations, ideally on the official practice platform, so the candidate experiences the real branching before test day. Most 555 scorers who complete that cycle move to the 605 to 625 range on the retake.
What a 605 signals: a competitive but not yet convincing score
A 605 on the GMAT Focus is a respectable score. For many regional MBA programmes, online MBAs, and specialised master's degrees, a 605 is comfortably above the median and a candidate can apply with confidence. For top-25 full-time MBAs, 605 is at or just below the middle of the admitted range, and the candidate will need other parts of the application to lift the file. The 605 band is the score I think of as 'workable', because it usually means the candidate is performing reliably on easier stems and has a defined ceiling on the harder ones.
Reading the 605 in section terms, the candidate typically has a quant in the low 20s, a verbal in the high teens to low 20s, and data insights in the high teens to low 20s. The quant tends to be the strongest because the question pool there is the most drillable. The verbal tends to be the trickiest because reading comprehension and critical reasoning are sensitive to inference vocabulary that is hard to memorise. Data insights is where the ceiling shows up: the candidate is strong on straightforward data interpretation and sorting-and-filtering tables, but loses points on the multi-source and two-part items.
For a 605 candidate, the next 30 to 50 points come from a narrower set of moves. First, fix the last two quant items that get skipped. Most 605 scorers run out of time around question 19 or 20, and the fix is a 90-second cap per item with a 10-second triage pass at question 15 to mark which of the remaining six to attempt first. Second, drill inference stems until 'must be true' versus 'could be true' versus 'must be false' becomes automatic. Third, build a 25-second routine for opening a Multi-Tab Reasoning prompt: skim the question, identify which tab is answer-relevant, and ignore the other two until the second pass. Each of these is small, but the points add up. A disciplined 605 candidate who executes these three moves for four to six weeks is realistically looking at a 635 to 655 on the retake.
What a 655 signals: a competitive score with a defined next band
The 655 band is where applications start to look strong across a wide range of programmes. Top-25 MBAs will read a 655 as competitive, many top-15 programmes will consider it within range, and specialised master's programmes at selective business schools will treat it as above the median. A candidate at 655 can usually move forward with the application rather than retaking, although a retake is still worthwhile if the candidate believes the next 30 to 50 points are achievable in a single prep cycle.
Reading a 655, the section scores are typically quant in the mid-to-high 20s, verbal in the mid-20s, and data insights in the mid-20s. The interesting feature of this band is that the candidate is now scoring more than half marks on the harder stems. Data Sufficiency is being handled with a recognisable method rather than a guess. Multi-Tab Reasoning is being triaged rather than read front-to-back. Two-Part Analysis is being treated as one question with two answer slots, not as two independent questions. The remaining losses are concentrated in inference vocabulary, the trickiest data interpretation charts, and the rare quant item that requires an unusual geometry or number-property insight.
The 655-to-705 gap is the hardest gap on the test, and it deserves honest framing. It is rarely closed in less than 8 to 12 weeks of focused work, and it usually requires the candidate to give up some untimed practice in favour of timed mixed sets. Two areas tend to drive most of the movement. The first is the inference stem in verbal: candidates who score in the 655 range still lose one or two items per section on 'supported' versus 'true' versus 'inferred' stems, and that vocabulary work is mechanical. The second is the harder graphics interpretation items, where two y-axes or stacked bars need to be read in a specific order. If a 655 candidate spends three weeks on inference stems and three weeks on advanced graphics, with weekly mixed timed sets, the realistic ceiling is in the 685 to 705 range.
Reading your own 655: a five-minute diagnostic
Before deciding whether to retake, a 655 candidate should do a 30-minute diagnostic. Pull the last two full-length practice tests, sort every wrong item by section and question type, and rank the four weakest families. In my experience, the 655 candidate usually has one family responsible for 40% of the losses and three families responsible for the rest. A targeted 4-week drill on that dominant family, with a weekly mixed set to maintain the others, tends to add 15 to 30 points on a retake. A scattergun approach that touches everything tends to add nothing.
What a 705 signals: a top-percentile score with specific risk
A 705 on the GMAT Focus is a top-percentile result. It is at or above the median for admitted classes at most top-10 MBA programmes, and it is comfortably above the middle 80% range at many top-25 programmes. For the candidate, a 705 means the application moves on its own strength without the test being a soft spot in the file. The 705 is also the score where the cost-benefit of a retake changes: another 20 to 30 points gives diminishing admissions returns and consumes weeks of prep time that could be spent on essays, recommendations, and interview prep.
Reading a 705, the section scores typically show quant in the high 40s, verbal in the high 30s to mid-40s, and data insights in the high 30s to mid-40s. The candidate is missing perhaps 6 to 9 items across the whole exam. The errors are no longer clustered in one family; they are spread across inference stems, the trickiest two-part analysis prompts, the rare quant word problem with an unusual structure, and the occasional data interpretation item where the chart has been misread. The candidate has internalised pacing, treats the test as a 90-minute decision-making exercise rather than a knowledge test, and uses a 25-second triage on the first 12 items of every section.
The risk for a 705 candidate is over-investment. There is a temptation to retake and push for a 735 or 745, but the marginal admissions benefit of those extra points is small relative to the time cost. My own advice at this band is to lock the score, move the prep energy into the application narrative, and only retake if a specific programme publishes a hard floor above 705. Even then, the candidate should run a one-week refresher rather than a full prep cycle, because the test is now a maintenance activity, not a learning activity.
Mapping the four scores to programme fit
The relationship between a GMAT Focus score and a programme's admitted range is the most important practical question a candidate has. The table below translates the four headline scores into programme-fit language, using the published middle 80% ranges for representative tiers of business programmes.
| GMAT Focus score | Typical section shape | Programme tier where it sits at or above the median | Recommended next move |
|---|---|---|---|
| 555 | Q high teens, V high teens, DI mid teens | Online MBAs, regional full-time MBAs, many specialised master's | Four-week diagnostic reset; do not apply yet |
| 605 | Q low 20s, V high teens to low 20s, DI high teens to low 20s | Strong regional MBAs, top-25 specialised master's, many European MBAs | Six-week targeted drill on the dominant loss family |
| 655 | Q mid-to-high 20s, V mid-20s, DI mid-20s | Top-25 full-time MBAs, many top-15 specialised programmes | Decide between retake and applying; both are reasonable |
| 705 | Q high 40s, V high 30s to mid-40s, DI high 30s to mid-40s | Top-10 full-time MBAs, most selective specialised programmes | Lock the score, move energy to application narrative |
Two caveats. First, the section shapes in the table are typical patterns, not exact thresholds — a candidate with a 605 and a balanced profile can outperform a candidate with a 605 and a 35 in quant with a 12 in data insights. Second, programme medians shift over time, so a candidate should always read the most recent class profile of each target programme and treat the table as a translation guide, not a guarantee.
Question-type gaps that separate one band from the next
The four score bands are not equally far apart. The 555-to-605 jump covers roughly the same ground as the 605-to-655 jump, but the 655-to-705 jump is twice the work. The reason is that each band is gated by a specific question-type frontier, and crossing the frontier requires a different kind of preparation.
From 555 to 605. The frontier is basic Data Sufficiency and the easier Data Interpretation items. A 555 candidate usually cannot reliably distinguish statement (1) alone, statement (2) alone, both together, or neither, and they are losing 3 to 5 items per section because of that confusion. Drill item: any Data Sufficiency prompt where the answer is (A) or (E) — the most common patterns. Time target: 1 minute 30 seconds per item, then down to 1 minute 15 seconds.
From 605 to 655. The frontier is Multi-Tab Reasoning and the harder sorting-and-filtering tables. A 605 candidate usually reads all three tabs of a multi-tab prompt before looking at the question, which doubles the time cost. The fix is to read the question first, identify the relevant tab, and only read the second tab if the first one does not resolve the question. The harder sorting tables require a 25-second triage on column headers before reading any row. Drill item: any Multi-Tab prompt with a clearly answer-relevant tab.
From 655 to 705. The frontier is Two-Part Analysis and the inference stems. A 655 candidate can usually handle a Two-Part Analysis with a clean structure, but loses 1 to 2 per section on the prompts where the two answer slots are not symmetric. The fix is to label the two parts explicitly, work them in parallel, and check that the answer is consistent across both parts. Inference stems require vocabulary work: 'supported', 'must be true', 'could be true', 'must be false'. A 655 candidate who scores these correctly 80% of the time can usually push to 90%+ with three weeks of deliberate practice.
How long does it take to move from one band to the next?
The honest answer depends on the candidate's starting diagnostic and on whether they are studying full-time or alongside work. The rough pattern I have seen is as follows. A 555-to-605 move takes 4 to 6 weeks of part-time study, 20 to 30 hours per week, with a clear four-week reset structure. A 605-to-655 move takes 6 to 10 weeks, 15 to 25 hours per week, with a target on the dominant loss family and a maintenance set on the rest. A 655-to-705 move takes 8 to 14 weeks, 15 to 25 hours per week, with inference vocabulary and Two-Part Analysis as the dominant work.
Two practical notes on timing. First, the first 30 points of movement are the cheapest, and the last 20 points are the most expensive, so a candidate should retake after each band of movement rather than waiting for a single large jump. Second, the test adapts section by section, so the candidate who trains under timed conditions sees a different adaptive path on test day than the candidate who trains untimed. A 705 candidate who only ever studied untimed is rarer than a 705 candidate who studied timed from week one.
Common pitfalls and how to avoid them at each band
Each band has a characteristic error pattern, and the patterns are different. Knowing the pattern is half the fix.
- At 555: Studying the wrong material. The official starter material is enough to reach 605, but it is not enough to reach 655. The fix is to identify the four question families above and drill those, not to read more general material.
- At 605: Mistaking content review for practice. The candidate spends three hours a week reading a guide and 30 minutes a week doing practice items, which is the inverse of what the test rewards. The fix is to flip the ratio: 30 minutes of review, 3 hours of timed practice, with every wrong item tagged by question type.
- At 655: Mistaking a strong quant for a strong test. A candidate with a 35 in quant and 20s in verbal and data insights is at 655 but is not a 705 candidate. The fix is to drill the two weaker sections rather than polish the strong one.
- At 705: Retaking for diminishing returns. The 705 is already a competitive score, and another 20 points rarely changes the admissions outcome. The fix is to lock the score and move prep energy into the application.
- Across all bands: Confusing a full-length practice test with real preparation. A practice test is a measurement, not a study session. The candidate should sit a test, then spend 2 to 3 times the test length reviewing the wrong items.
Putting it together: a reading of your own score
When a candidate sees a 555, 605, 655 or 705, the score is a starting point, not a verdict. The 555 says: foundations need work, the test is not yet a strength, and a four-week reset will move the score. The 605 says: the basics are working, the next 30 points come from a defined set of question-type drills, and a six-week cycle is realistic. The 655 says: the candidate is competitive at a wide range of programmes, the next 30 to 50 points come from a more targeted drill, and a retake is a judgement call against application timing. The 705 says: the test is no longer the weak point, the application narrative matters more than another 20 points, and prep time should be redirected. Each score has a clear reading, a clear next move, and a clear honest framing of how long the move takes.
For most candidates reading this, the practical question is not 'what does 605 mean in the abstract' but 'what does 605 mean for me, given my target programmes and my timeline'. The answer to that question lives in the diagnostic: which sections are above the band average, which sections are below, and which question family is responsible for the largest share of the losses. Once that is clear, the prep cycle writes itself, and the next band of movement is a matter of weeks rather than months.
TestPrep İstanbul's diagnostic assessment is a natural starting point for candidates who want a reading of their own 555, 605, 655 or 705 and a written prep plan that targets the specific question-type frontier between their current band and the next one.
Conclusion and next steps
The four headline scores on the GMAT Focus — 555, 605, 655 and 705 — are not just numbers on a transcript. They are signals about the candidate's current relationship with the test, the question families that are still costing points, and the kind of prep cycle that will move the score. Treat the score as a diagnostic, identify the dominant loss family, build a 4 to 12 week cycle around it, and retake with a clear target band. The next move is the diagnostic, not the retake.