675 with DI 77 to DI 83, and the section I had barely touched

Six points of Data Insights. That is my whole improvement.

My first official attempt was a 615 carrying a DI77, and I finished at 675: V83, Q84, DI83, the 95th percentile. Verbal and quant each moved by a single point.

Two years of on and off prep, a course already finished once, and no way to test if any of it worked. I practised off PDFs and ran my own timer. I kept an error log with no idea whether it was any good, and there was nobody to help me figure out where exactly I was messing up. I had even bought this course back in 2021 and abandoned it after one mock, out of pure nerves.

e-GMAT starts at the diagnosis. The Personalised Study Plan takes your target, your test date and the hours you can honestly commit, then tells you what is fixable in the time left. PACE does the same at topic level, a diagnostic you take before studying, sorting lessons into leave alone and learn properly. I ran it after verbal and it confirmed I could skip what I was already good at. That saved me weeks.

Then nothing ran on instinct. The end-of-topic cementing quizzes carry a pass mark you clear or retake, and they are brutal. My hard Word Problems set opened at 30 percent, then 20, then 20 again, then 20. I am an engineer. That stung.

The error-log template turned it around. It does not ask which answer I got wrong, it asks at which step in my own process I came off the rails. Quant stopped being a knowledge problem and became a behavioural one. The same set read 50 percent the next day. Timed sectionals in Scholaranium took my quant from a 78 in early June into the high eighties, touching 90 twice, and Neuron OG did the same for CR with full reasoning behind every official question. My hard sets went from the low fifties to 70 inside a week.

DI is where e-GMAT is doing something I have not seen elsewhere. The course is built type by type and broad enough that you learn the thinking, never a template. Hard Data Sufficiency went 30 to 80, Graphical Interpretation went 40 to 80, and MSR went 60 to 83 on medium and 50 to 77 on hard. TPA on hard went 70 to 80, my strongest sub-section by then.

The change I would point at is not a number. On an MSR I used to open the question and start hunting, and now I read every tab first, the units, the time periods, where each figure sits. Three minutes in there feels insane. You get all of it back.

Last Mile Push was not really teaching. It was a mentor reading my dashboards weekly, naming habits I could not see, and telling me the numbers were sound when I did not feel it.

One warning. The mocks here are built to punish. Some will hurt.

If you know the content and you are plateauing anyway, this is the one to buy, especially if DI is the section you have been hoping will sort itself out.