GMAT Percentiles Updated (Aug 2026) - Full Analysis

Question

GMAC has updated the official percentiles today. This is the third August in a row that the update has landed at roughly the same point in the calendar, so the timing itself is no longer a surprise.

Link to the new percentiles on mba.com

Below is the full comparison against last year's table, followed by the section percentiles and what the movement actually says about the testing pool.

Earlier 
 https://gmatclub.com/forum/download/file.php?id=85354

Now 
 https://gmatclub.com/forum/download/file.php?mode=view&id=89518 
  SUMMARY

23 of the 61 total-score points moved. Every single one of them moved by exactly 1 percentile point.  Nothing moved by 2  or more, in either direction.

555 through 735 moved down 1 point.  14 score points, and this is the band that covers most competitive applicants. 
  255 through 485 moved up 1 point.  9 score points at the bottom of the scale. 
  Everything else held.  38 of 61 score points are unchanged, including the entire 745 to 805 range and the 495 to 545 range.

The section tables are the more interesting half of this release. Quant, Verbal and DI did not move together, and they did not move in the same direction. Full breakdown further down.

TOTAL SCORE - EVERY ROW THAT MOVED

Score  
  2026  
  2025  
  Delta  
 
  735 
 99% 
 100% 
 -1 
 
  685 
 95% 
 96% 
 -1 
 
  675 
 94% 
 95% 
 -1 
 
  665 
 91% 
 92% 
 -1 
 
  655 
 90% 
 91% 
 -1 
 
  645 
 86% 
 87% 
 -1 
 
  635 
 81% 
 82% 
 -1 
 
  625 
 78% 
 79% 
 -1 
 
  615 
 75% 
 76% 
 -1 
 
  605 
 69% 
 70% 
 -1 
 
  595 
 66% 
 67% 
 -1 
 
  585 
 60% 
 61% 
 -1 
 
  565 
 50% 
 51% 
 -1 
 
  555 
 47% 
 48% 
 -1 
 
  485 
 22% 
 21% 
 +1 
 
  455 
 16% 
 15% 
 +1 
 
  405 
 8% 
 7% 
 +1 
 
  395 
 7% 
 6% 
 +1 
 
  385 
 6% 
 5% 
 +1 
 
  355 
 4% 
 3% 
 +1 
 
  335 
 3% 
 2% 
 +1 
 
  305 
 2% 
 1% 
 +1 
 
  255 
 1% 
 0% 
 +1

Key anchors that did  not  move: 805 to 745 all remain 100th, 725 and 715 remain 99th, 705 remains 98th, 695 remains 97th, 575 remains 57th, 545 remains 42nd, 505 remains 27th.

ANALYSIS

The bad

675 is the only score that crossed a round threshold.  It fell from the 95th to the 94th percentile. If you have been writing "95th percentile" on a resume off a 675, that line is now out of date. 
  735 is no longer a clean 100th.  It now reports as 99th. The practical difference is nil, though the reported figure has changed. 
  The hit landed squarely on the applicant range.  555 to 735 covers the scores most people in this forum are actually reporting. All 14 points in that stretch went down.

The good

No score moved more than 1 point.  In the August 2024 update several scores moved further, and the speculation before that release was worse than the release itself. The same holds here. 
  705 held at the 98th percentile.  The score that took the most attention in previous updates did not move at all this year. 
  The top of the scale is completely untouched.  745 through 805 remain at 100th, exactly as in 2025. 
  This is normal drift, not a re-scaling.  A 1 point move in a cumulative percentile table means a fraction of a percent of the pool shifted across that score. It is a routine annual refresh of the reference group.

WHAT ACTUALLY CHANGED IN THE POOL

Reading the deltas one row at a time understates what happened. It is more useful to convert the cumulative percentiles into the share of test takers sitting inside each band. The reason for doing this is that a cumulative table hides where the mass moved, and the band view does not.

Band  
  2026  
  2025  
  Shift  
 
  205 - 405 
 8% 
 7% 
 +1 pt 
 
  405 - 505 
 19% 
 20% 
 -1 pt 
 
  505 - 605 
 42% 
 43% 
 -1 pt 
 
  605 - 705 
 29% 
 28% 
 +1 pt 
 
  Above 705 
 2% 
 2% 
 no change

It is interesting to note that the movement runs in both directions at once. The 605 to 705 band picked up a point of the pool and the 205 to 405 band also picked up a point, while the 405 to 605 middle gave up two points between them. The distribution did not inflate, it spread out. More candidates are now clustered just below 705 than a year ago, and slightly more are at the very bottom, with the middle thinning.

That is the mechanism behind the headline. Scores in the 555 to 735 range each lost a percentile point because the group immediately beneath them got more crowded, not because anyone's performance was re-graded.

SECTION PERCENTILES (CURRENT)

The combined table below is the new 2026 figures for all three sections. The year on year section comparison follows it.

Score  
  Quant  
  Verbal  
  DI  
 
  90 
 100% 
 100% 
 100% 
 
  89 
 97% 
 99% 
 100% 
 
  88 
 95% 
 98% 
 99% 
 
  87 
 93% 
 97% 
 99% 
 
  86 
 91% 
 96% 
 99% 
 
  85 
 88% 
 93% 
 98% 
 
  84 
 85% 
 87% 
 97% 
 
  83 
 80% 
 81% 
 95% 
 
  82 
 75% 
 73% 
 93% 
 
  81 
 70% 
 64% 
 88% 
 
  80 
 64% 
 54% 
 83% 
 
  79 
 57% 
 46% 
 76% 
 
  78 
 50% 
 38% 
 69% 
 
  77 
 43% 
 30% 
 61% 
 
  76 
 38% 
 23% 
 52% 
 
  75 
 33% 
 18% 
 46% 
 
  74 
 28% 
 14% 
 40% 
 
  73 
 23% 
 11% 
 34% 
 
  72 
 20% 
 8% 
 29% 
 
  71 
 17% 
 6% 
 24% 
 
  70 
 14% 
 4% 
 20% 
 
  69 
 11% 
 3% 
 17% 
 
  68 
 9% 
 3% 
 14% 
 
  67 
 7% 
 2% 
 11% 
 
  66 
 5% 
 2% 
 10% 
 
  65 
 4% 
 2% 
 8% 
 
  64 
 3% 
 1% 
 7% 
 
  63 
 2% 
 1% 
 5% 
 
  62 
 2% 
 1% 
 5% 
 
  61 
 1% 
 1% 
 4% 
 
  60 
 1% 
 1% 
 3%

Mean scaled scores: Quant 77.92, Verbal 79.43, Data Insights 75.14. DI remains the lowest scoring section of the three.

SECTION YEAR ON YEAR - THE THREE SECTIONS SPLIT

This is the part of the release worth reading closely. 34 section percentile points moved in total, and they did not move together.

Quant - 12 moved, 10 of them upward

Score  
  2026  
  2025  
  Delta  
 
  88 
 95% 
 96% 
 -1 
 
  87 
 93% 
 94% 
 -1 
 
  76 
 38% 
 37% 
 +1 
 
  75 
 33% 
 32% 
 +1 
 
  74 
 28% 
 26% 
 +2 
 
  73 
 23% 
 22% 
 +1 
 
  72 
 20% 
 19% 
 +1 
 
  71 
 17% 
 15% 
 +2 
 
  70 
 14% 
 13% 
 +1 
 
  69 
 11% 
 10% 
 +1 
 
  68 
 9% 
 8% 
 +1 
 
  67 
 7% 
 6% 
 +1

Everything from 67 to 76 gained percentile. A mid range Quant score is worth  more  this year than last, which means fewer candidates are clearing those scores than a year ago. The only Quant losses are at 87 and 88, one point each.

Verbal - 11 moved, 9 of them downward

Score  
  2026  
  2025  
  Delta  
 
  88 
 98% 
 99% 
 -1 
 
  87 
 97% 
 98% 
 -1 
 
  85 
 93% 
 94% 
 -1 
 
  84 
 87% 
 89% 
 -2 
 
  83 
 81% 
 83% 
 -2 
 
  82 
 73% 
 74% 
 -1 
 
  81 
 64% 
 66% 
 -2 
 
  80 
 54% 
 56% 
 -2 
 
  79 
 46% 
 47% 
 -1 
 
  73 
 11% 
 10% 
 +1 
 
  65 
 2% 
 1% 
 +1

Verbal 80, 81, 83 and 84 each fell 2 points. Those are the four largest single moves anywhere in this release, total score included. The entire 79 to 88 stretch went down, so the upper middle of the Verbal pool is meaningfully stronger than it was in 2025.

DI - 11 moved, every single one downward

Score  
  2026  
  2025  
  Delta  
 
  81 
 88% 
 89% 
 -1 
 
  77 
 61% 
 62% 
 -1 
 
  76 
 52% 
 53% 
 -1 
 
  75 
 46% 
 47% 
 -1 
 
  74 
 40% 
 41% 
 -1 
 
  73 
 34% 
 35% 
 -1 
 
  71 
 24% 
 25% 
 -1 
 
  70 
 20% 
 21% 
 -1 
 
  67 
 11% 
 12% 
 -1 
 
  63 
 5% 
 6% 
 -1 
 
  60 
 3% 
 4% 
 -1

DI is the only section with no upward movement at all. Eleven scores moved and all eleven went down, spread evenly from 60 all the way to 81. The whole DI field shifted up by a small, consistent amount.

WHAT THE SECTION SPLIT MEANS

Summing the deltas gives the cleanest read of it. Quant nets  +10  percentile points across all its moved scores, Verbal nets  -11 , and DI nets  -11 .

It is interesting to note that this is the second year in a row the same split has appeared. In the August 2025 thread the observation was that the minor inflation was showing up in Verbal and DI with nothing visible in Quant. That pattern has not only held, it has become clearer, with Quant now moving measurably in the opposite direction rather than staying flat.

Read plainly, candidates are getting better at Verbal and Data Insights and slightly worse at mid range Quant. The reason this matters for a current test taker is that it changes where a given scaled point is scarce, though it does not change any target worth setting.

The reassuring part

Despite 34 section points moving, the scaled score required to hit every standard percentile threshold is  identical  in both years, in all three sections.

Percentile  
  Quant  
  Verbal  
  DI  
 
  50th 
 78 
 80 
 76 
 
  60th 
 80 
 81 
 77 
 
  70th 
 81 
 82 
 79 
 
  80th 
 83 
 83 
 80 
 
  90th 
 86 
 85 
 82 
 
  95th 
 88 
 86 
 83

Not one of those 18 figures changed between 2025 and 2026. If you had a section target last week, it is the same target today.

THE SECTION FINDING WORTH ACTING ON

Comparing the three sections at the same scaled score shows how differently they are priced.

Target  
  Quant  
  Verbal  
  DI  
 
  80th percentile 
 83 
 83 
 80 
 
  90th percentile 
 86 
 85 
 82 
 
  95th percentile 
 88 
 86 
 83

A scaled 82 is the 93rd percentile in DI, the 75th in Quant, and the 73rd in Verbal. To reach the 90th percentile you need an 86 in Quant but only an 82 in DI, a gap of 4 scaled points for the identical percentile. The reason is straightforward, DI has the lowest mean of the three sections at 75.14, so the same raw performance clears a larger share of the field.

Two practical consequences follow from this.

DI is the cheapest percentile on the exam.  If your DI is sitting in the 70s and your Quant is already in the mid 80s, the next point of total score is far easier to find in DI. 
  Verbal 80 to 81 buys more percentile than any other single point on the exam.  That one scaled point moves you from the 54th to the 64th percentile, a 10 point jump. Verbal 81 to 82 is worth another 9. No single scaled point anywhere else on the exam pays that much, the best Quant equivalent is 7 points and the best DI equivalent is 9.

Verbal is also the steepest section overall through the middle. Going from 80 to 84 moves you from the 54th to the 87th percentile, 33 percentile points for 4 scaled points. The same 4 point climb in Quant is worth 21 points and in DI is worth 14.

QUESTIONS

My score report says one percentile and mba.com says another. Which is right? 
The published table is the current one. In past updates the percentile shown against an existing score was refreshed in the reporting system rather than frozen at the value from the test date, so expect the reported figure to follow the new table.

I have already submitted applications with the old percentile. Does that matter? 
A 1 point move does not change how an application reads. Adcoms work from the score, and a 675 that reads 94th instead of 95th is the same 675. There is nothing here that warrants contacting a school.

Does this mean I need a higher score than last year? 
No. The score you need is set by the schools you are applying to, and the underlying scores have not changed. What changed is the percentile label attached to them, by one point, for scores between 555 and 735.

Why did the bottom of the scale go up while the middle went down? 
Those two movements are not in conflict. The 205 to 405 band and the 605 to 705 band each gained a point of the pool while the 405 to 605 middle lost two points between them. The distribution widened at both ends rather than shifting in one direction.

Which section should I be putting my time into? 
That is set by your own splits, not by this update. What the update does tell you is that DI remains the cheapest percentile on the exam and Verbal remains the steepest through the middle, and neither of those facts changed this year. The section score needed for every percentile threshold from the 50th to the 95th is the same as it was in 2025.

Verbal fell 2 points at four different scores. Is Verbal getting harder? 
The test is not getting harder. The Verbal pool is getting stronger, which is a different thing. A 2 point percentile drop at Verbal 84 means slightly more candidates now score at or above an 84 than did a year ago. Your own preparation is unaffected by that.

How often does this happen? 
The last three updates have all landed in August. On that pattern the next refresh would be expected around August 2027, though GMAC has not committed to a schedule.

Should I retake because of this? 
Not on the basis of a 1 point percentile move. If you were already considering a retake because your score is short of your target range, that reasoning is unchanged by this update.

RELATED THREADS

Official GMAT percentiles have been updated (Aug 2025)  
  GMAT Focus Percentiles Changed (Aug 2024)  
  GMAT Focus Score Calculator

If you want your own score checked against the new table, post it below with the section splits and I will add it to the thread.

bb · Answer

Thank you! It seems this was a fairly minor update was just some housekeeping items... seems more and more people are scoring higher and higher on the GMAT focus. And I guess it takes just a single person I’ll shift at the top to cascade everything down by one.

This could be a matter of 500 people’s scoring higher this year because the volume is low. 😬

This could be a matter of 500 people’s scoring higher this year because the volume is low. 😬

bb · Answer

Actually it’s two changes:
1. 735 - so people scoring at or above this mark is now significant enough to be a half a percent so GMA around it up to 1%. (that’s my assumption how they calculate things since they have 100 percentile) 
2. ⁠ 685 also added some folks and that pushed everyone else down, so just two non-uniform data points and both a fairly high. One is very high

yc168 · Answer

Thanks for the analysis.

If possible, I'd suggest that you can use the percentile numbers in the two concordance tables attached. That could be helpful to generate numbers with better precision.

Attachment:

GMAC_Concordance_Total_August2026.pdf [59.32 KiB]
Downloaded 19 times

Attachment:

GMAC_Concordance_Total_July2025.pdf [59.6 KiB]
Downloaded 8 times

bb · Answer

Oh oh. Does it mean one of our tables is incorrect? 😱

yc168 · Answer

Sorry, bb, I'm not sure. I didn't compare the numbers from the two sources. I just thought the percentile numbers in the pdf file have one additional decimal point, compared with the numbers GMAC published on the web page.

The 2026 concordance table is from the following link:
 https://blog.gmat.com/hubfs/07.Assessments/GMAT%20Exam/School%20Resources%202025/Concordance%20Tables%20July%202025/GMAC_Concordance_Total_July2025.pdf?__hstc=34420606.917035fa18385bd0b9c38f86d977398b.1740300708468.1786551945154.1786573700703.283&__hssc=34420606.1.1786573700703&__hsfp=422542f9c0b0b20f7de8f925ea426150

The 2025 concordance table is what I download before. Not sure if it is still available on the Internet...

bb · Answer

Thank you Got it. That’s where we got them too. Will doublecheck anyway.

bhanu29 · Answer

Thanks for posting this, I sourced my analysis from the official page, which is in sync with these tables.

GmatTest2026 · Answer

Thanks for posting the detailed analysis. I can see that one user ID points to tables that ultimately say the same things or don't add much new value to the main breakdown. Honestly, it doesn't add much substance to the detailed analysis you provided.

Also, simply suggesting the use of the concordance tables attached without providing clear reasoning or explaining how they added value over the primary data makes me feel that the comment made by this user ID is a bit redundant.

Thank you!

yc168 · Answer

I was in a hurry earlier today so probably did not make myself very clear. The analysis done by bhanu29 in the opening post was great. I just thought the data with an additional decimal point would provide a bit more nuance. I'll give two examples below:

Example 1:  For a score of 755, the 2025 percentile was 99.9% while the updated 2026 percentile is now 99.8% (There may still be rounding in these numbers, but let's take them as they are for lack of better data.)

People might think - what's the big deal about a 0.1% difference? But if you look at it from the opposite direction, people scoring 755 were the top 0.1% previously, now they are top 0.2%. Assuming there were 100,000 exams taken in the past year, for simplicity of calculation (The number was 93,196 total exams taken in Test Year 2025.), there were 100 persons scoring 755 or above. Now there are 200 persons scoring 755 or above, doubling from pervious year. The increase is actually quite dramatic, much more than what the 0.1% difference in percentile indicates on the surface.

Not to mention that the percentile update is based on data over the past 5 years, sort of like an weighted average or moving average. If we are to look at the data only for the most recent year, it's likely that the change will be even more pronounced than the change in the weighted/rolling average.

Example 2:  For a score of 705, the 2025 percentile was 98.0% while the updated 2026 percentile is now 97.5%

Using the same logic and assumption as in Example 1 above, there used to be 2,000 persons scoring 705 or above in a single year, now there are 2,500 people scoring 705 or above. The percentage increase is not as dramatic as in Example 1, but that's 500 more people scoring 705 or above (including 100 more scoring 755 or above), so 400 more scoring between 705 and 755.

Granted the analysis above might not be relevant for some test takers, but for people scoring or aspiring to score a 705+ in GMAT, I hope that's a bit helpful to get a better feel of how competition is intensifying. As to why/how so many more people are scoring 705+, that will be a huge discussion for another day.

bhanu29 · Answer

For applicants I would still say the practical takeaway does not change. If your GPA is decent, anything above 705 does not add much, and the official score report rounds the percentile to a whole number anyway. I wish a higher score counted for more, but MBA admissions really are holistic. What the shift does mean is that the score stops being your differentiator earlier than it used to, so the time is better spent on essays and recommendations.

It does matter tiny bit, if your GPA doesn't show your actual ability.

It does matter tiny bit, if your GPA doesn't show your actual ability.

GmatTest2026 · Answer

Thank you for providing those examples! They really help clarify the logic behind the analysis and make it much easier to digest. My earlier question was just to get a bit more context, so I truly appreciate you taking the time to expand on it.
Thank you!

yc168 · Answer

Agree with most of your comments. Given the context of your full analysis, I think it makes total sense to use the whole-number percentiles in your post. Using the percentile numbers with additional decimal point is just my own personal preference, and it was presumptuous for me to suggest that in your post.

The analysis I was thinking indeed did not fit into your original post. So I'll just post one of them below as supplemental information for whomever interested. This is the comparison of the percentile numbers between 2026 and 2025, across the higher-end of the score range:

Score  
   2026  
 
   2025  
 
  Delta  
 
  805 
 100.0% 
 100.0%  
 - 
 
  795 
 100.0%  
 100.0%  
 - 
 
  785 
 100.0%  
 100.0%  
 - 
 
  775 
 100.0%  
 100.0%  
 - 
 
  765 
 99.9% 
 99.9%  
 - 
 
  755 
 99.8%  
 99.9%  
 -0.1% 
 
  745 
 99.6%  
 99.7%  
 -0.1%  
 
  735 
 99.4%  
 99.5%  
 -0.1%  
 
  725 
 98.9%  
 99.1%  
 -0.2% 
 
  715 
 98.6%  
 98.8%  
 -0.2%  
 
  705 
 97.5%  
 98.0%  
 -0.5% 
 
  695 
 96.9%  
 97.4%  
 -0.5%  
 
  685 
 95.1%  
 95.8%  
 -0.7% 
 
  675 
 94.0%  
 94.8%  
 -0.8% 
 
  665 
 91.2% 
 92.1% 
 -0.9% 
 
  655 
 89.6% 
 90.5% 
 -0.9%  
 
  645 
 85.6%  
 86.7% 
 -1.1%

In addition to around 500 more people scoring 705 or above (concentrating on 705), as mentioned in my previous post, there are around 600 more people scoring between 645 and 695 (fairly evenly distributed), for 2026 compared with 2025.

Also agreed that MBA admissions are holistic. One minor disagreement would be that I think the point beyond which scores stop being differentiating will become higher over time, not lower (if that's what you meant by earlier). Take the percentiles over time for score 645 as an example:

Jan. 2024 
 Jul. 2025 
 Aug. 2026 
 
  89.3% 
 86.7% 
 85.6%

645 was almost a 90% score in the early days of GMAT Focus, and has now dropped to around 85%. Percentile ranking for a given score will most likely continue to drop over time, as people are getting better and better at this edition of the test. When more people are scoring at the higher-end of the range and mean/median GMAT scores published in school profiles rise, most likely the bar will be heightened as well, imho.

bhanu29 · Answer

Thanks for the analysis. I agree that over time the arithmetic mean will increase, which is natural xD (that’s why I am publishing this analysis). I meant, rn given the present circumstances score above 705 doesn’t matter much because of the holistic nature of the application.

	Target Test Prep - Save up to $599 $400 Discount + $99 Bonus GMAT Club Tests
	e-GMAT: 60% Off on GMAT Focus Get to 735+ from any starting point
	Magoosh increases your score, guaranteed - 15% off ALL Magoosh GMAT plans with code GMATCLUB15 + Free GMAT Club Tests.
	Manhattan Prep: $99 Bonus Score Higher on Your GMAT Exam. Guaranteed.
	Experts' Global - ~~$1720~~ $960 End-to-End GMAT Prep + App Support: 150+ Videos, 2000+ Questions, 15 FLTs +
	GMATWhiz - $299 ~~$499~~ Experience the course through a free trial.
	GMAT Club Tests - Free Included with every course purchase of $149+ View all deals

GMAT Percentiles Updated (Aug 2026) - Full Analysis

Prep Toolkit

Top 5 GMAT Debrief Videos

615 to 715 in 15 Days (Ria's Strategies)

More than Quant/Verbal Abilities: Speed vs. Accuracy

745 with Only Self-Prep and GMAT Club

From 555 to 765: 210-point Improvement

Perfect 805 Score Debrief by Julia

My Rewards

Score	2026	2025	Delta
735	99%	100%	-1
685	95%	96%	-1
675	94%	95%	-1
665	91%	92%	-1
655	90%	91%	-1
645	86%	87%	-1
635	81%	82%	-1
625	78%	79%	-1
615	75%	76%	-1
605	69%	70%	-1
595	66%	67%	-1
585	60%	61%	-1
565	50%	51%	-1
555	47%	48%	-1
485	22%	21%	+1
455	16%	15%	+1
405	8%	7%	+1
395	7%	6%	+1
385	6%	5%	+1
355	4%	3%	+1
335	3%	2%	+1
305	2%	1%	+1
255	1%	0%	+1

Band	2026	2025	Shift
205 - 405	8%	7%	+1 pt
405 - 505	19%	20%	-1 pt
505 - 605	42%	43%	-1 pt
605 - 705	29%	28%	+1 pt
Above 705	2%	2%	no change

Score	Quant	Verbal	DI
90	100%	100%	100%
89	97%	99%	100%
88	95%	98%	99%
87	93%	97%	99%
86	91%	96%	99%
85	88%	93%	98%
84	85%	87%	97%
83	80%	81%	95%
82	75%	73%	93%
81	70%	64%	88%
80	64%	54%	83%
79	57%	46%	76%
78	50%	38%	69%
77	43%	30%	61%
76	38%	23%	52%
75	33%	18%	46%
74	28%	14%	40%
73	23%	11%	34%
72	20%	8%	29%
71	17%	6%	24%
70	14%	4%	20%
69	11%	3%	17%
68	9%	3%	14%
67	7%	2%	11%
66	5%	2%	10%
65	4%	2%	8%
64	3%	1%	7%
63	2%	1%	5%
62	2%	1%	5%
61	1%	1%	4%
60	1%	1%	3%

Percentile	Quant	Verbal	DI
50th	78	80	76
60th	80	81	77
70th	81	82	79
80th	83	83	80
90th	86	85	82
95th	88	86	83

Target	Quant	Verbal	DI
80th percentile	83	83	80
90th percentile	86	85	82
95th percentile	88	86	83