The fastest route into data analytics in India is SQL first, then Excel and Power BI, then Python with pandas — in that order — followed by a portfolio of three real analyses. Entry-level data analysts typically earn ₹4–6 LPA, rising to ₹8–15 LPA with three to five years of experience, and the role is hiring across BFSI, e-commerce, healthcare and GCCs rather than only in tech companies.
Data analyst is the most accessible serious tech career in India right now: no engineering degree required, tooling you can learn in months, and hiring spread across banking, insurance, e-commerce, healthcare, logistics and every GCC in the country. It is also a role drowning in bad advice — twelve-tool roadmaps and "become a data scientist in 90 days" courses. This guide is the boring, effective version: three tools, one portfolio, in the right order.
Learn the tools in this order
1. SQL — the non-negotiable core
Every analyst job runs on SQL. Learn SELECT with filtering and sorting, aggregations with GROUP BY and HAVING, all the JOIN types against realistic multi-table schemas, subqueries and CTEs, and window functions — ROW_NUMBER, RANK, LAG, running totals — which are the single most-tested topic in 2026 analyst interviews. Practise on real datasets, not toy tables: load an e-commerce dataset into free PostgreSQL and answer business questions against it. Six to eight weeks of daily practice is a realistic budget.
2. Excel and Power BI — where the business lives
Excel still matters — pivot tables, XLOOKUP, and clean spreadsheet habits — because your stakeholders live there. Then Power BI: data modelling with fact and dimension tables, Power Query for cleaning, DAX for measures (start with CALCULATE, FILTER and time intelligence), and dashboard design that answers a question rather than decorating a page. The PL-300 (Microsoft Power BI Data Analyst) certification is the one Indian employers actually name in job posts and is worth taking once you can build a multi-page report from raw data unaided.
3. Python with pandas — the extender
Python earns its place third. With pandas you clean messy files SQL never sees, automate the weekly report nobody enjoys, and pull data from APIs. Learn the language basics, then pandas (reading files, filtering, groupby, merge, pivot), plus matplotlib or seaborn for quick charts. A little statistics belongs here too: distributions, correlation versus causation, and enough hypothesis testing to know when a difference is noise.
The portfolio that gets interviews
Three projects, each answering a real business question, each published where a recruiter can click:
- A SQL-heavy analysis — for example, cohort retention and revenue breakdowns on a public e-commerce dataset — written up with the queries visible.
- A Power BI dashboard on Indian public data (government open-data portals, RBI releases, census data) with a one-page summary of the decisions it supports.
- A Python cleaning-and-analysis notebook that takes a genuinely messy dataset from raw to insight, with your reasoning narrated between cells.
Three finished, explained projects beat ten certificates. In interviews you will be asked to walk through one of them — choose subjects you can talk about with energy.
Domain knowledge: the underrated multiplier
Two analysts with identical tooling are separated by domain fluency. If you target banking, learn what NPA, CASA ratio and credit risk mean; for e-commerce, know AOV, CAC, conversion funnels and cohort retention; for healthcare, understand claims and utilisation data. Employers routinely choose the candidate who speaks the business's language over the one with a longer tool list, because the first month of an analyst job is spent asking questions — and domain fluency makes those questions better. Pick one domain for your portfolio and let all three projects live in it; the coherence itself becomes a talking point.
Mistakes that keep applicants stuck
The common failure patterns are predictable: collecting course certificates instead of finishing projects; learning Python before SQL and stalling in both; building dashboards that display data without answering a question; and applying only to roles titled "data analyst" while ignoring MIS executive, reporting analyst and business analyst postings that involve identical work and convert into analytics titles within a year. The job market rewards finished, explained work in a specific domain — everything else is preparation theatre.
Salaries and who is hiring
- Entry level (0–2 years): ₹4–6 LPA in metros; BFSI, e-commerce and GCCs at the top of the band, small agencies below it.
- 3–5 years: ₹8–15 LPA for analysts with strong SQL-plus-BI and domain fluency; MIS-to-analytics converts land here too.
- Senior analyst / analytics lead: ₹16–28 LPA, with specialist paths (marketing analytics, risk analytics, product analytics) at the upper end.
Mumbai's BFSI concentration makes it one of the strongest analyst markets in the country — banks, insurers, NBFCs and their GCCs hire continuously, and they prize candidates who can explain numbers to non-technical stakeholders in clean English or Hindi. That communication skill is tested in every interview loop; practise presenting your portfolio out loud.
A six-month schedule
Months one and two: SQL daily, one dataset per week. Month three: Excel plus Power BI fundamentals, first dashboard shipped. Month four: DAX depth and the PL-300 attempt, second project. Months five and six: Python with pandas, third project, and applications — apply from month five onward rather than waiting to feel ready, because analyst interviews are themselves the best syllabus. Working professionals should halve the weekly load and double the calendar rather than skipping the portfolio months.
If you want the structured version, RST Forum runs data analytics training — SQL, Power BI through the PL-300 blueprint, and Python — in our Mumbai classroom and live online across India, with practice labs on the Skillopia platform and instructors who have taught working professionals since 1997. Over a million engineers have trained with us, and every batch starts the same way: an opening session you can sit in on, so you can judge the teaching before you commit.
Frequently asked questions
Can I become a data analyst without an engineering degree?
Yes. Data analytics is one of the most degree-flexible tech roles in India — commerce, science and even arts graduates get hired if they can demonstrate SQL, Power BI and a portfolio of real analyses. BFSI and consulting employers in particular value domain understanding as much as tooling.
SQL or Python — which should I learn first?
SQL, without question. It appears in effectively every data analyst job description, interviews test it hands-on, and day-to-day analyst work is more SQL than anything else. Python with pandas comes third in the sequence, after SQL and Power BI, and extends what you can do rather than replacing either.
Is Power BI or Tableau more in demand in India?
Power BI, by a clear margin. Microsoft's enterprise footprint — Office 365, Azure, Teams — has made Power BI the default in Indian BFSI, manufacturing and GCC environments, and the PL-300 certification is the one employers name in job posts. Tableau remains common in some MNCs and is easy to pick up later; the concepts transfer directly.
What salary can a data analyst starting out expect in India?
Typically ₹4–6 LPA in metros for a first analyst role, with BFSI, e-commerce and GCC employers at the upper end and small agencies below it. With three to five years and strong SQL-plus-BI skills, ₹8–15 LPA is a realistic band; analytics leads and specialists go higher.
Do data analysts need machine learning?
Not for the core role. Analysts describe and explain what happened; data scientists build predictive models. Basic statistics — distributions, correlation, hypothesis testing — is expected, and a little scikit-learn is a nice-to-have, but SQL, BI and business communication get you hired. You can grow toward data science later from a working analyst seat.
Written by
Software & Data Faculty
Full stack, Python & data engineering track leads
The software development and data teaching team — placeholder profile, rename to a named trainer in the CMS before launch.