The skills were there weeks before test day. The last 72 hours were a fight with my own nerves — and that turned out to be the real exam.The number that scared me three days outHere is the strange part of my story. For weeks my section scores had been landing at Q89, Q90, a V90 on my last full attempt. Then, three days from the exam, I sat a full-length mock and pulled a Q79. How do you score a 79 when 89 and 90 are your normal? I lay awake that night, woke at midnight, took another sectional, got a Q90, and finally slept. That gap between what I could clearly do and what I was doing under pressure is the whole reason this debrief exists.
Why I needed a platform that shows you where you are stuckI come from medicine, and I had lost a lot of touch with Algebra. My foundation was decent, but a decent base is exactly the kind that hides where you are actually leaking marks. I did not go comparison-shopping between options — I came in needing two things: tools that show me precisely what to fix, and a process I could repeat when my instincts failed. That is what made
e-GMAT, an online GMAT prep platform, fit my situation.
Algebra and Advanced Topics: trusting process over lucky answersMy old habit in Algebra was to back-solve — substitute the answer choices and reverse-engineer the equation. It worked until it didn't, and on a clock of roughly two minutes a question it failed me when I most needed it. So instead of chasing the right answer, I decided I would rather get the process right even on a question I got wrong, because that pays off later. To find where the leaks actually were, I lived in the data analytics on the portal — it logs every question and breaks accuracy down to the sub-topic — and once it pointed me at my weak spots, I ground the cementing quizzes after each topic until the method was automatic. My hard-question accuracy in Algebra moved from 65% to 87%, with my time per question flat or slightly down.
Algebra — hard-difficulty accuracy: 65% → 67% → 87% (time steady, ~2m18s)Advanced Topics was steeper. I used to force every probability and sets question through one formula and hope it fit. What broke that habit was the worked solutions sitting under each question — there is a subsection showing a couple of other ways the same question can be solved. Seeing two or three clean approaches let me pick the one I was actually comfortable with instead of defending one rigid framework. Hard-question accuracy in Advanced Topics went from 52% to 87%.
Advanced Topics — hard-difficulty accuracy: 52% → 87% → 87% (time down to ~2m18s)Drilling the weak spots until they heldWhen a sectional mock flagged that my number properties were shaky, I leaned on Neuron — the bank of official-style questions with worked solutions written by the experts rather than the standard ones. What made it work for me was the filtering: I could narrow down to one sub-topic, set the difficulty, and pull only the questions I had got wrong or not yet tried, then drill until I was clearing 80–90% of them. When a question genuinely stumped me, the "ask the expert" section under it — the internal forums — meant I could post the doubt and get an expert response in a few hours instead of losing a day to it. The questions sat right on par with what the real exam threw at me; on test day, with barely any time left on my last Quant question, the exact same one I had practised came up, values and all, and I just knew it. Scholaranium did the same at a broader level, with topic-level analytics behind it.
The fix that mattered most: a hard capEven with the skills in place, my timing was sabotaging me. I would get grossly involved in the first five or six questions, burn four or five minutes on them, and then have seven or eight minutes left for the last five — which is how silly errors creep in. The sectional mocks made this visible, and the fix was blunt: a hard cap of 2 minutes 30 seconds per question, with maybe 15–20 seconds of grace only if I was almost there. Past the cap, I bookmarked, took an intelligent guess, and moved on. I started finishing with three or four minutes to spare and circling back to the bookmarked ones with a clear head.
The part nobody warns you aboutOn verbal, what turned fluctuating scores into consistent ones was prethinking — deciding what the answer should look like before reading the options — on critical reasoning and reading comprehension. My sectional verbal climbed into the high V80s and a V90.
But the last phase was pure drama. Tuesday before a Thursday exam I scored a 645, the next morning a 625 — a number from the start of my prep. Something was clearly wrong, and it was not the process; consistent Q89s and V89s do not happen by fluke. This is where my mentor, Dhruv, mattered most. He did not hand me a new technique — he talked me down, even messaging me to say he would be brutal if that was what it took, because he did not want to lose me to nervousness. The reassurance, plus one simple question — what is the worst case, really? It is one attempt, and there is always another — is what let me stop catastrophising. I finally internalised it: shrink the exam to one question, give it 100% focus, hold the 2:30 cap, forget the rest. My final mock jumped to a 725. The difference between 625 and 725 was never the skill; it was mindset, confidence and the basics. If you are heading into your closing weeks, find someone who will catch the behavioral patterns your own nerves can't — it is worth more than any extra question you could grind.
What I would tell my earlier selfBuild the process, let the data point you, respect the clock — the skills only show on test day if your head lets them.


Attachment:
GMAT-Club-Forum-mg9yc9bg.png [ 28.4 KiB | Viewed 275 times ]
Attachment:
GMAT-Club-Forum-l4dn3enh.png [ 28.27 KiB | Viewed 275 times ]