I spend three hours commuting every day—ninety minutes from Kalyan to Bandra, ninety minutes back. That's where I think. Last month, on the local train, a junior analyst from my team asked me: "Should I learn Python or get better at Excel first?"
I didn't answer immediately. Because the honest answer isn't what most people expect.
For two years, I've watched Indian data professionals—especially those trying to break into finance or analytics—make the same choice wrong. They pick Python because it sounds cooler. Or they stick with Excel because it's "safer." And they end up mediocre at both.
I've done both. I studied Economics at university, spent my first job drowning in Excel sheets at Morningstar, and then spent a year learning Python because I thought I had to. Here's what I wish someone had told me straight up: the question itself is flawed. But if you're forced to choose one first, there's a clear winner for Indian millennials in data roles.
And it's not the one you're probably thinking about.
Excel Is Still The Language Most Indians Speak
Let's start with the uncomfortable truth.
Every single company I know in Mumbai uses Excel. Not as a backup tool. As the primary tool. Your boss sends you data as an .xlsx file. Your reporting happens in Excel. Your financial models are built in Excel. Your fund manager at that ₹50 crore NBFC? Still using macros from 2008.
I'm not exaggerating. At Morningstar, I've seen analysts with fifteen years of experience who've never written a single line of Python. They're brilliant. They can pull together a complex multi-sheet analysis in two hours that would take most Python developers a full day to code.
Why Excel Wins in the Real Indian Workplace
Excel is the lingua franca. Every stakeholder understands it. Your manager doesn't need you to explain what a .ipynb file is. Your client doesn't care if your model is "scalable." They care if the numbers are right and if they can modify it themselves in a month when requirements change (spoiler: they will).
There's also friction with Python adoption in India. Companies move slower here. Infrastructure is often older. Approval for new tools takes three months and two committee meetings. A CFO's decision to let you use Jupyter notebooks? That's 2025 conversation. Maybe 2026.
But here's the thing that matters most: Excel teaches you data thinking. When you build a model in Excel, you're forced to understand each calculation. You see your errors immediately. You're not abstracting logic into functions you can't debug easily. For someone learning for the first time, this is gold.
The Hidden Advantage of Excel Fluency
When you're really good at Excel, you can do something that impresses people. You can take messy, real-world data—the kind that comes from Tally exports or bank statements with formatting issues—and clean it, reshape it, and analyze it in minutes. Not hours. Minutes.
I've seen it happen. Someone comes back from meeting with raw data. A good Excel person delivers insight before lunch. That person becomes irreplaceable. And then they get a raise. And then they can afford better coffee on the local.
Also: Excel skills translate immediately into ₹. You can freelance with Excel (Upwork, Toptal). You can build models for startups during weekends. You can help your uncle's business clean up their books. Python? Much harder to monetize quickly unless you're already senior.
Python Is For When You're Outgrowing Excel
But—and this is a big but—Excel has limits. Real limits.
When you're working with 50,000 rows, Excel starts feeling slow. With 500,000 rows? Your laptop dies. Add multiple data sources, complex transformations, and automated reporting that needs to run daily, and Excel becomes a bottleneck. You spend more time updating formulas than thinking about problems.
That's when Python isn't optional anymore. It's necessary.
When You Actually Need Python
I'll give you real scenarios where Python beats Excel completely:
Scenario 1: You work at a fintech startup (like Groww or CRED) and need to process transaction data for millions of users daily. Excel? Dead on arrival.
Scenario 2: You're analyzing market data from Zerodha. You need to pull data from APIs, clean it, backtest a strategy, and visualize results—all automatically, every morning. Excel macros will make you want to quit.
Scenario 3: You're at a bigger company where someone else owns data infrastructure, but you need to pull from databases, merge datasets that live in different places, and create reproducible reports. Python and tools like Jupyter notebooks let you version-control your work. Excel files? Someone always ends up with "Final_FINAL_v3_REAL.xlsx"
Python scales. Excel doesn't. Full stop.
The Learning Curve Problem Nobody Mentions
Here's what surprised me: Python is harder to learn if Excel is all you've ever known. Not because of the syntax. Because of the mental model.
In Excel, everything is visual. You see the data. You see the results. Python lives in abstraction. You write code, run it, and trust it worked. If you're not used to thinking in functions, loops, and data types, it feels like staring into a void.
I spent four months learning Python (honestly? it was frustrating). But those four months came after five years of Excel mastery. I already understood data logic. I understood why you'd transform data in steps. I just needed to learn the syntax.
If I'd tried Python first? I'd probably have quit after two weeks and convinced myself I'm "not a coding person."
The Head-To-Head Comparison
| Criteria | Excel | Python |
|---|---|---|
| Job Market in India (2024) | Essential for 95% of roles | Required for 40% of roles; highly valued in 20% |
| Learning Curve | 2-3 months to be useful | 3-6 months to build real things |
| Immediate Applicability | Can use in job from week one | Useful after 2-3 months of practice |
| Scalability | Breaks with large datasets (100K+ rows) | Handles millions of rows easily |
| Flexibility | Limited; you're constrained by cell-based logic | Unlimited; you can automate anything |
| Freelance Income Potential | ₹30,000-₹1,00,000 per project | ₹1,00,000-₹5,00,000+ per project |
| Debugging Difficulty | Easy; errors are visible | Hard; errors hide in logic |
| Maintenance Over Time | Gets messy quickly as complexity grows | Stays clean if written well |
My Perspective
I want to be honest about something that changed my thinking.
In my Economics degree, we had this one professor who said: "Master your tools so deeply that you stop thinking about the tool and start thinking about the problem." That stuck with me. For years, I thought this meant learning as many tools as possible. But it actually means the opposite.
It means becoming so fluent in one tool that it becomes invisible to you. And for most of us starting out, that tool should be Excel.
What surprised me? How many senior analysts I've met who are still primarily Excel users. They make ₹25+ lakhs annually. They run entire finance functions. They're not "behind" because they don't code. They're ahead because they understand business problems deeply and can solve them faster than anyone else.
But here's what I got wrong initially: I thought once you learned Excel really well, Python would be harder. It wasn't. It was easier. Because I already had the logic framework built. I just needed to learn the syntax. That part took two weeks once I actually wanted to do it.
So my honest take? Start with Excel. Get genuinely good at it (not just VLOOKUP good). Then layer Python on top when you hit Excel's limits. That's the realistic path for most Indians in data roles.
The Real Answer: Learn Excel First
If you're between 22 and 28, trying to break into data analysis or finance, and you've never seriously used either tool?
Learn Excel first.
Here's why, practically:
You'll get hired faster with strong Excel skills than with mediocre Python. You'll contribute value immediately. You'll understand data problems better because you're forced to manually work through them. And then, in 12-18 months when you're tired of the limitations, you'll be ready for Python—with a real understanding of why you need it.
The people who skip Excel and jump to Python often burn out because they're learning abstract logic without any concrete problem to solve. They build "Hello World" programs and tutorials but never use Python for a real job. Then they convince themselves coding isn't for them.
It's the wrong order.
The right order is: Excel → Problems → Python → Automation → Career Growth.
Not: Python → Tutorials → Frustration → Quit.
Final Thoughts
That analyst who asked me on the train? I told her to spend the next three months really learning Excel. Not surface-level. Deep. Build models. Learn to optimize. Understand what "good" looks like. Then, if she still feels limited, Python will make sense and will be easier to learn.
She texted me three months later. She'd landed a role at a ₹500 crore fintech startup. Not because she learned Python. Because she was so efficient with Excel in her interview that they hired her on the spot and said, "We'll teach you to code once you're here."
That's the move.
So here's what I'd suggest: Stop debating online forums. Stop watching YouTube videos comparing the two. Pick Excel. Block out 90 days. Get genuinely good. Build something that impresses you. Then reassess.
You won't regret it. I promise.
Dattatray Dagale
Data Analyst • Blogger • Mumbai
I'm a data analyst from Kalyan, Maharashtra, working at Morningstar. I write about personal finance, career growth, and everyday life for Indian millennials — the stuff I wish someone had told me earlier.
Written by Dattatray Dagale • 31 October 2026
.png)
.gif)
0 Comments