When I started as a data analyst three years ago, I thought I had to master everything. Excel, Python, Tableau, SQL, Power BI — I signed up for paid courses, bought subscriptions I barely used, and convinced myself that spending money on tools would make me better. Spoiler: it didn't.
What actually happened? I wasted ₹15,000 on a Udemy course teaching Power BI when I could've learned the same thing free on YouTube. I paid for a Tableau Public premium upgrade (which doesn't even exist, by the way — I was genuinely confused). And I spent three months learning advanced Excel VBA when 80% of my actual job just needed basic formulas and pivot tables.
The real turning point came during my first month at Morningstar. A senior analyst named Priya asked me what tools I was comfortable with. I rattled off my paid subscriptions like they were badges of honor. She laughed — not unkindly — and said, "Dagale, you know Google Sheets is free and does 90% of what you'll need, right?" That conversation changed everything.
Here's what I've learned: the best tools aren't always the most expensive ones. They're the ones you actually use. And as a data analyst commuting from Kalyan to Bandra every day, I realized that the tools you can access from anywhere — browser-based, cloud-first, genuinely free — are often more valuable than the desktop software sitting on your laptop that you rarely touch.
Let me walk you through the tools I actually use now. Not the aspirational tools I thought I *should* use. The ones that've saved me time, made my work cleaner, and cost me absolutely nothing.
Google Sheets — The Tool I Kept Underestimating
Here's my honest mistake: I treated Google Sheets like a training-wheels version of Excel. Something you outgrow once you get serious. I was wrong. Completely wrong.
Why I Used to Dismiss It
When you work at an investment research firm like Morningstar, Excel feels like the professional choice. It's what Bloomberg terminals and institutional investors use (well, partially). Google Sheets felt... amateurish? Corporate? I don't know. The bias was there.
I also believed the myth that Sheets couldn't handle large datasets. That it was slow. That it lacked the formulas I needed. None of that is true for 95% of real-world analysis work, but I didn't know that yet.
What Changed for Me
Two things: first, I started using Sheets for quick exploratory analysis before moving to Python or SQL. Just to understand the data shape. And I realized it was often *faster* than firing up Excel because there's no installation lag, no file bloat, no "recovering unsaved changes" nonsense.
Second, I discovered IMPORTRANGE and QUERY functions. QUERY in particular is honestly more intuitive than a pivot table for someone who knows SQL — it literally lets you write SQL-like syntax to analyze data. Suddenly, Sheets wasn't a stepping stone to "real" tools. It was a real tool.
The collaboration aspect is underrated too. When I needed to send a financial model to a colleague in Mumbai, she could edit it in real-time, leave comments, and I could see her changes instantly. No email attachments. No version control chaos. No "Final_Final_FINAL_v3.xlsx" nonsense.
Python and Jupyter Notebooks — Free, But I Learned Wrong
This is where I really messed up. I learned Python the way you're *supposed* to learn it in college — with IDE theory, variable types, OOP principles. By lecture 8, I was deep in class inheritance and it felt pointless.
What I should've done: install Anaconda (free), open Jupyter Notebook (free), and just start analyzing data with pandas and NumPy. That's it.
The Anaconda + Jupyter Combo
Anaconda is a free distribution of Python that comes pre-loaded with most libraries you'll actually need — pandas, NumPy, Matplotlib, Scikit-learn. No dependency hell. No hours lost to "import errors". Just download, install, done.
Jupyter Notebooks are browser-based coding environments where you can write code, see results, add visualizations, and write markdown notes all in one place. It's how data analysts actually work — exploring, iterating, learning as you go. Not writing perfect production code from line 1.
What I Use It For Now
Quick data cleaning. Feature engineering. Building analysis workflows that I can document and share with teammates. When I need to pull data from APIs or databases and transform it before dumping into Sheets or Power BI, Python is faster than clicking through spreadsheet menus.
And here's the thing: Jupyter Notebooks are reproducible. I can save my analysis with all the code and outputs, someone else can open it next month, re-run it with updated data, and get the same results. That's powerful. Excel files? Good luck knowing what formula was in that cell six months ago.
The learning curve isn't as steep as people think. If you can write Excel formulas, you can learn basic pandas in a week. Seriously.
SQL and DBeaver — The Combo That Surprised Me
I used to think SQL was only for database administrators. Something data analysts learned eventually, but not urgently. That changed fast.
Most datasets worth analyzing live in databases, not Excel files. To access them, you need SQL. And DBeaver is a free database client that lets you connect to pretty much any database (PostgreSQL, MySQL, SQL Server, even SQLite) and write queries without paying Jetbrains ₹12,000 for DataGrip.
Why This Matters for Analysts
Here's the practical truth: if you can't write SQL, you're dependent on someone else to extract data for you. You're slow. You're waiting for a data engineer to fulfill your requests. At a startup or mid-size company, that's a bottleneck nobody can afford.
Once you can write SQL, you're independent. You can answer your own questions, pull data when you need it, and move fast. And DBeaver makes it smooth — autocomplete, syntax highlighting, ability to visualize query results instantly.
The Learning Path
You don't need to learn database optimization or normalization first. Learn SELECT, WHERE, JOIN, GROUP BY, and ORDER BY. That's 80% of what you'll actually write. The rest you'll pick up on the job.
I learned SQL through Mode Analytics' free SQL tutorial (still one of the best resources out there), then practiced on my company's databases with guidance from a senior analyst. Two months in, I was writing complex queries independently.
Visualization Tools — Power BI Free vs. Looker Studio vs. Simple Charts
This is where everyone gets confused. There are so many choices, and the industry pushes you toward paid tools like Tableau or Power BI Premium.
Here's my honest take: for most analysis work, you don't need them.
Looker Studio (Google's Free Alternative)
It connects directly to Google Sheets, Google Analytics, and SQL databases. You can build interactive dashboards, share them as links, and your stakeholders can filter data themselves. Is it as polished as Tableau? No. But it's free, it works, and it covers 85% of what you'll need in a typical analyst role.
I built a portfolio tracking dashboard using Looker Studio that pulls data from a Google Sheet I update monthly. It automatically refreshes, looks clean enough to show in meetings, and took me two hours to build. If I'd waited for a Tableau license approval, I'd still be waiting.
Power BI Free vs. Paid
Power BI has a free desktop version. It's legitimately good — you can build complex visualizations, connect to multiple data sources, and publish to the web. The catch? You can't collaborate easily and the free tier for sharing is limited.
For personal learning or internal use, Power BI free is solid. But if your organization already uses it and you need the paid version to share dashboards, that's a decision above your pay grade.
When to Use Simple Charts Instead
Honestly? Most of the time. I used to think every analysis needed a dashboard. Now I realize that a well-designed scatter plot in Python's Matplotlib or Seaborn, exported as a PNG, often communicates better than an interactive dashboard.
Over-engineering visualization wastes time. Your stakeholders often just need to see the trend, the outliers, the correlation. A simple chart answers that. A dashboard makes them click around and get confused.
| Tool | Cost | Best For | Learning Curve |
|---|---|---|---|
| Google Sheets | Free | Quick analysis, collaboration | Very Low |
| Jupyter Notebooks | Free | Data cleaning, exploration | Medium |
| DBeaver | Free | Database queries | Medium |
| Looker Studio | Free | Interactive dashboards | Low-Medium |
| Power BI (Free) | Free | Complex visualizations | Medium |
| Matplotlib/Seaborn | Free | Static charts | Medium |
The Unsung Heroes — Git, VS Code, and Documentation
These aren't glamorous. Nobody talks about them at career conferences. But they've genuinely changed how I work.
Git and GitHub — Version Control You'll Actually Need
Here's what I got wrong: I thought Git was for software engineers. For "real" coders. Not for analysts.
Then I had a nightmare scenario. I'd written a complex analysis script, tweaked it for three days, realized the original version was better, and had no way to get it back. I had five versions saved locally and couldn't remember which was which. Lost an entire day.
Git is version control. You write code, you "commit" it with a message explaining what changed, and you can go back to any previous version instantly. GitHub is where you store it in the cloud so you don't lose it if your laptop dies.
The learning curve is real (I won't lie), but the free tier of GitHub is perfect for analysts. And most teams use it now anyway, so learning it early is an investment that pays off.
VS Code — A Code Editor That Doesn't Suck
When I learned Python, I was using basic text editors or bloated IDEs. VS Code changed that. It's free, it's lightweight, and it has extensions for everything — Python, SQL, Jupyter notebooks, formatting tools. Once you set it up, you won't look back.
The integrated terminal is huge. Write code in VS Code, run it immediately, see errors, fix them, done. No switching between windows.
Documentation — The Tool Everyone Ignores
This is the most underrated tool in data work. A README file (just a .txt or .md file) explaining what your analysis does, how to run it, what the output means — this saves future you and your teammates hours.
I used to think documentation was busywork. Now I realize it's the difference between looking like someone who's organized and someone who's chaotic. When you hand off a project to a colleague, good documentation is the gift that keeps giving.
My Perspective
Last month, I was having chai with my colleague Rohit (who works in data engineering) and I mentioned how I'm now writing most of my exploratory analysis in Jupyter instead of Excel. He laughed and said, "Welcome to 2020, Dattatray. I've been waiting for you."
What surprised me most was how condescending it wasn't. He was genuinely happy that I'd figured out a better workflow. And then he added something that stuck with me: "The tool doesn't matter. Shipping analysis fast and documenting it well does."
That's when I realized my mistake wasn't just wasting money on paid tools. It was fixating on the *tool* instead of the *outcome*. I spent energy on which platform to use when I should've spent it on asking the right questions and communicating answers clearly.
I used to think that mastering paid, premium tools would make me valuable. What actually makes me valuable? Being independent. Not waiting for others to extract data. Writing code that's reproducible. Explaining my methodology so someone else can verify my work six months later. None of that requires expensive subscriptions.
If I could start over, I'd learn in this order: SQL (so you can get your own data), Python basics (so you can clean and analyze), Google Sheets (so you can explore quickly), and visualization (so you can communicate). Paid tools come later, only if your job specifically requires them.
Final Thoughts
Being a data analyst in 2024 is actually good fortune. The best tools are free. They're cloud-based, they're powerful, and they're built by the smartest engineers in the world. Google, Meta, and the open-source community have given you everything you need to do serious work.
Don't fall into the trap I did — thinking that money spent = progress made. The analyst who ships clean analysis using Google Sheets and Python is more valuable than the analyst with every paid subscription who's still waiting for the perfect tool to start.
Learn these tools deeply. Not superficially. Master one visualization tool. Get genuinely comfortable with SQL. Really understand how to structure a Jupyter notebook. That depth matters more than breadth.
And if you're just starting out? Don't feel pressured to know everything. I still don't. But I know how to learn quickly, I know where to find answers, and I know that the best tool for the job is usually the free one that's already in your browser.
Start with Sheets. Move to SQL and Python. The rest will follow naturally.
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 • 05 September 2026
.png)
.png)
0 Comments