Between March and August 2004, I collected 47 job rejections for accounts work. I had just passed my 12th in Commerce through Patrachar — no coaching, no tuitions, self-studied — and I was trying to get into an accounts role, any accounts role, that would let me build a career instead of a series of odd jobs.
47 times, the answer was no.
The cyber cafe method
What changed the outcome wasn't talent showing up late. It was a system, built out of necessity, that I only recognised as a system years later.
After every single rejection, I went to a cyber cafe — paid by the hour, which meant every minute of research had to count — and searched, as specifically as I could, the exact questions that had caused that day's rejection. Not "accounting interview questions" in general. The actual question I'd fumbled, word for word.
I wrote the answers by hand in a notebook. I practiced them out loud, alone, until they didn't sound rehearsed anymore. And then I walked into the next interview slightly less unprepared than the one before it.
By the 47th rejection, I had, without ever holding a real accounts job, accumulated something close to a year or two of working knowledge — pulled out of 47 separate hiring processes, each one a small, brutal, free education.
The part most people skip
During those same months, I was also teaching mathematics to 29 students who had failed their compartment exams. 28 of the 29 passed. I mention this not as a side note, but because it's the same method pointed in a different direction: find exactly what's broken, drill it specifically, don't move on until it's fixed.
The 48th interview said yes. It was a CA firm, and it was the door that eventually led to Sara Textiles, to Rosmerta Autotech, to twenty-two years in operational finance, and eventually to building 7AM & Realtime CFO™ — a product whose entire philosophy is "find exactly what's leaking, not what's generally wrong."
Every rejection was curriculum. I just had to be willing to read it that way.
I think about this system often when I talk to founders about their MIS. Most businesses don't fail from one big mistake. They fail the way I nearly did — from treating each miss as bad luck instead of as a specific, fixable, well-documented lesson sitting right there in the data — usually in an Excel sheet or a SQL table nobody looked at closely enough.
That same drill-it-specifically habit is what eventually turned into 22+ years of hands-on Excel, VBA and Power Query work — the same habit, now automated with Python, SQL and Claude, and running underneath every model I build today.