Why I Build Systems, Not Spreadsheets
The Night That Changed Everything
Three years ago, I was sitting at my desk at 11 PM on a Tuesday. My team and I were manually reconciling vendor delivery data across four spreadsheets, trying to figure out why our open-to-buy for womenswear was off by an amount large enough that nobody could explain it.
A big number. The kind that turns a quiet evening into a crisis.
Turns out someone had copy-pasted the wrong column from an old file. One wrong column. Four hours of detective work. That was the night I decided: never again.
The Spreadsheet Trap
Here is something nobody tells you when you start a career in retail buying. The entire industry runs on Excel. I am not exaggerating. Buying plans, vendor scorecards, assortment matrices, margin trackers. All of it lives in spreadsheets that one person "owns" and everyone else copies.
The problem is not Excel itself. Excel is a brilliant tool. The problem is that spreadsheets reward heroics.
When your buying calendar is a spreadsheet, the person who stays latest wins. When your vendor evaluation is a spreadsheet, the person who remembers to update it is the hero. When your display allocation is a spreadsheet, you are one accidental overwrite away from chaos.
I led a women's-fashion category at scale across 530+ India stores. That is not a spreadsheet problem. That is an engineering problem.
What I Actually Built
Over the past few years, I have built six operational systems at Max Fashion. Not apps. Not startups. Operational systems. Workflows and automations that replaced manual drudgery with repeatable processes.
The Buying Calendar Automation. Our planning cycle used to take a season. It was mostly waiting. Waiting for data, waiting for approvals, waiting for someone to consolidate numbers. I mapped every step, identified the bottlenecks, and built automated handoffs. Now it takes a sprint.
The Vendor Scoring System. We work with hundreds of vendors. Evaluating them used to be subjective. Whoever shouted loudest got the next order. I built a scoring model that weighs the dimensions that matter. Now the data talks, not the loudest person in the room.
The Display Logic Framework. Which products go to which stores? With 530+ locations across different cities, climates, and customer profiles, you cannot eyeball this. I built allocation logic that reads each store on its own terms. It is not perfect, but it is a thousand times better than gut feel multiplied by 500.
Systems Compound, Heroics Exhaust
This is the core principle I operate on: systems compound, heroics exhaust.
When you pull an all-nighter to fix a data problem, you are a hero for one night. But tomorrow there will be another fire. And next week. And next month. You will burn out, or your best people will leave, or both.
When you build a system that prevents the data problem from happening, you are boring for one quarter. But next quarter, that system is still working. And the quarter after that. And it frees you up to think about the next problem instead of relitigating the last one.
I have never seen a career built on all-nighters that did not eventually collapse. I have seen plenty of careers built on good systems that kept compounding.
The Takeaway
If you are drowning in spreadsheets and late nights, pick the one process that eats the most time and causes the most errors. Map it end to end. Find the step where a human is doing something a rule could do. Automate that one step.
Then do it again. And again.
Systems compound. Heroics exhaust. Build accordingly.