---
title: Data Analyst Roadmap 2026: Skills, Free Resources, Projects
url: https://www.masaischool.com/resources/data-analyst-roadmap
description: Data analyst roadmap in 8 steps: spreadsheets, SQL, statistics, a BI tool, Python, storytelling and a portfolio, with a free first resource for each skill.
updated: 2026-11-17
publisher: Masai (https://www.masaischool.com)
---

# Data Analyst Roadmap for India: What to Learn, in What Order, With Free Resources and Projects

> **Short answer**
>
> To become a data analyst, learn spreadsheets first, then SQL (Structured Query Language), then one dashboard tool such as Power BI, Tableau Public or Data Studio. Add basic statistics and Python with pandas, practise explaining results in plain words, and publish three projects built on real data. Two free Kaggle courses cover SQL and pandas in about seven hours.

This roadmap is for students, freshers and career switchers in India who want a first analyst job. It gives the order to learn things, a free first resource for each skill, what to build, and what hiring data says about 2027. A BI (business intelligence) tool is software for building dashboards. Every resource and number below was opened and checked on 11 November 2026.

Pay is covered on the [data analyst salary page](https://www.masaischool.com/resources/data-analyst-salary-in-india), so this page does not repeat it. The broader sequence of topics is on the [data analytics learning path](https://www.masaischool.com/resources/data-analytics).

1. **Spreadsheets** - clean, pivot, look up
2. **SQL** - joins, grouping, filters
3. **Statistics** - spread, sampling, tests
4. **One BI tool** - a dashboard end to end
5. **Python and pandas** - messy and large files
6. **Storytelling** - one chart, one decision
7. **AI-assisted work** - use it, then verify it
8. **Portfolio** - three projects, each written up
The order to learn the skills. SQL comes early because most analyst work starts with pulling data out of a database.

## What a data analyst does

A data analyst answers business questions with data: how many orders did we lose last month, which customers come back, did the new offer work. The output is usually a query, a dashboard or a short note that someone can act on. You pull data with SQL, clean it, summarise it, and explain what it means to people who do not read tables for a living.

The role is narrower than a data scientist's, which adds prediction and experiments (see the [data scientist roadmap](https://www.masaischool.com/resources/data-scientist-roadmap)), and different from a data engineer's, who builds the pipelines that deliver the data. Titles overlap in practice: business analyst, product analyst, BI developer and operations analyst often share most of the same skills, so read the posting, not just the title.

## Skills, why they matter, and the first free resource

| Skill | Why it matters | First free resource (checked 11 Nov 2026) |
|---|---|---|
| Spreadsheets | The quickest way to look at a new file, build a pivot table and check a number before you write any code | [Google Sheets training and help](https://support.google.com/a/users/answer/9282959) and [Excel help and learning](https://support.microsoft.com/en-us/excel), both free to read |
| SQL | How you get data out of a database. It is the tool you will use most often | [Kaggle Intro to SQL](https://www.kaggle.com/learn/intro-to-sql): about 3 hours, six lessons from SELECT to joins, no cost |
| Statistics | Tells you whether a difference is real or noise: averages, spread, sampling, confidence intervals, hypothesis tests | [Khan Academy Statistics and probability](https://www.khanacademy.org/math/statistics-probability), 16 units, or the free PDF of [OpenIntro Statistics](https://www.openintro.org/book/os/) |
| Python and pandas | Cleans and reshapes data that is too messy or large for a spreadsheet, and repeats the work with a script | [Kaggle Pandas](https://www.kaggle.com/learn/pandas): about 4 hours, no cost, then [10 minutes to pandas](https://pandas.pydata.org/docs/user_guide/10min.html) |
| Dashboards | Turns a query into something a manager opens every Monday | [Microsoft Learn Power BI training](https://learn.microsoft.com/en-us/training/powerplatform/power-bi) (Power BI Desktop is free but installs on Windows), [Tableau Public](https://www.tableau.com/products/public) (free, published work is public) or [Data Studio](https://docs.cloud.google.com/looker/docs/studio) (no cost) |
| Data storytelling | A correct number that nobody acts on is wasted work. Lead with the decision, then the evidence | [storytelling with data blog](https://www.storytellingwithdata.com/blog), free to read |
| Business metrics | Conversion, retention, churn and average order value are the vocabulary of most analyst questions | [Google Analytics demo account](https://support.google.com/analytics/answer/6367342): real store data, open to any Google user |
| Using AI tools carefully | AI tools write queries and chart code quickly, but you own the answer, so you must check row counts and logic | The [free AI tools](https://www.masaischool.com/resources/tools/free-ai-tools) list: pick one assistant, then check its SQL against your own row counts |

Two practical notes. Google renamed Looker Studio back to Data Studio in April 2026, so older tutorials use the earlier name for the same free tool: see the [dashboard guide](https://www.masaischool.com/resources/build/dashboard-google-sheets-looker-studio). Power BI Desktop installs from the Microsoft Store or as a Windows executable, so on a Mac a web-based tool such as Data Studio is the easier start. The Power BI Data Analyst certificate (PL-300) is optional and the exam is paid: Microsoft lists $165 USD, with the price varying by country.

## SQL you should be able to write from memory

Employers can see your SQL in a test; they cannot see how many tutorials you watched. Be able to group, count, filter groups and join two tables without looking anything up. This script builds a tiny orders table in memory using Python's built-in SQLite and answers two typical questions: revenue by month, and which customers ordered more than once.import sqlite3 con = sqlite3.connect(":memory:") con.executescript(""" CREATE TABLE orders (order_id INTEGER, customer TEXT, order_date TEXT, amount INTEGER); INSERT INTO orders VALUES (1, 'Asha', '2026-09-03', 1200), (2, 'Ravi', '2026-09-10', 800), (3, 'Asha', '2026-10-02', 1500), (4, 'Meera', '2026-10-05', 700), (5, 'Ravi', '2026-10-19', 900), (6, 'Asha', '2026-10-28', 400); """) monthly = """ SELECT substr(order_date, 1, 7) AS month, COUNT(*) AS orders, SUM(amount) AS revenue, COUNT(DISTINCT customer) AS customers FROM orders GROUP BY month ORDER BY month; """ for row in con.execute(monthly): print(row) repeat = """ SELECT customer, COUNT(*) AS orders, SUM(amount) AS spend FROM orders GROUP BY customer HAVING COUNT(*) > 1 ORDER BY spend DESC; """ print("repeat customers:") for row in con.execute(repeat): print(row)

It prints ('2026-09', 2, 2000, 2) and ('2026-10', 4, 3500, 3) for the months, then Asha with 3 orders and 3,100 spent, and Ravi with 2 orders and 1,700. Change the data and the question, run it again, and you are practising the same moves a real analyst test asks for. Then learn window functions and joins across three tables.

## What to build

Three projects with different shapes beat ten copies of the same dashboard. For each one, write the business question, the data source and the date you downloaded it, the method, one chart, the recommendation and the limits of the data.

- **A sales or operations dashboard.** Monthly revenue, top products, returns. Build it from a Google Sheet by following the [live dashboard guide](https://www.masaischool.com/resources/build/dashboard-google-sheets-looker-studio), or in Power BI or Tableau Public.
- **A customer analysis in SQL and pandas.** Repeat customers, churn, or segments. The [data analytics projects](https://www.masaischool.com/resources/projects/data-analytics-projects-with-ai) page has ten ideas with SQL, and the [Python data project ideas](https://www.masaischool.com/resources/projects/python-data-project-ideas) page has code-first versions. Masai's [15 data analytics projects for beginners](https://www.masaischool.com/blog/15-data-analytics-projects-for-beginners-2026/) is a longer list.
- **A website or product analysis.** Explore the Google Analytics demo account, or set up your own site with the [GA4 and Search Console guide](https://www.masaischool.com/resources/build/setup-ga4-search-console).

For data, use the [free datasets list](https://www.masaischool.com/resources/tools/free-datasets-for-projects) and India's [Open Government Data Platform](https://www.data.gov.in/), which showed a maintenance notice when checked on 11 November 2026. The [free data analytics tools](https://www.masaischool.com/resources/tools/free-data-analytics-tools) page covers the rest of the toolkit.

## 2027 outlook: demand, the entry level and AI

**Demand is selective, not absent.** Naukri's JobSpeak index for September 2026 stood at 3,053 against 3,002 a year earlier, up 2%, with fresher hiring up 1% and IT and software services down 4%. Its October reading, reported on 3 November, was up 14% year on year (2,836 against 2,480), but Info Edge's managing director Hitesh Oberoi said part of that reflects a soft festive base last year. JobSpeak counts hiring by sector and experience band, not analyst vacancies, so treat it as the backdrop. Masai's [data analyst demand in India](https://www.masaischool.com/blog/data-analyst-demand-in-india/) post sets out the readings month by month.

**Longer-range view.** The World Economic Forum's Future of Jobs Report 2025, a survey of employers, lists big data specialists among the fastest-growing roles in percentage terms to 2030. That is a global result and says nothing about how easy a first job is.

**The entry level and AI.** A Stanford Digital Economy Lab paper, revised on 12 August 2026, uses US payroll data from ADP through June 2026. It finds employment of workers aged 22 to 25 in AI-exposed occupations 19% below where it would be had it kept pace with less-exposed peers, mainly through reduced hiring, with no comparable gap for experienced workers. The authors call these early descriptive indicators, not causal estimates. Our reading: first-draft queries and chart code are what AI tools now do quickly, so a junior analyst is judged on defining the metric, checking the data and explaining the result. Build projects that show those three things.

## Interview preparation

- **SQL screen.** Expect joins, GROUP BY with HAVING, and often a window function. Practise on your own tables, then on a [coding practice site](https://www.masaischool.com/resources/tools/coding-practice-websites) that has SQL sets.
- **A metric case.** Example: orders fell 10% week on week, what do you check? Say how you define the metric, then split by time, channel, city and product, then rule out a data fault such as a broken tracking tag, and only then suggest causes outside the product.
- **A project walk-through.** Pick one project and explain the question, the data, the decision it supports and what you would do differently. Interviewers test whether you understand your own work.
- **Spreadsheet and statistics basics.** Be ready to explain a pivot table, a lookup, an average versus a median, and what a confidence interval says.
- **Plain-language explanation.** You may be asked to explain a result to a non-technical manager in two sentences.

## Frequently asked questions

### How do I become a data analyst with no experience?

Learn spreadsheets, SQL and one dashboard tool, then build three projects on real data and write each up with a question, a method and a recommendation. Projects are your experience. Apply for analyst, business analyst and junior BI roles, and keep SQL practice going while you apply.

### Which is more important for a data analyst, SQL or Python?

Learn SQL first. It is how you retrieve and aggregate data from databases, and it is the usual first test. Python with pandas comes next for messy files, large datasets and repeatable work, though some analyst roles never need it, so check the postings you are targeting.

### Do I need a computer science or statistics degree to be a data analyst?

Postings differ, so check the ones you want. Whatever your degree, a portfolio with SQL and a dashboard is the evidence you control. Basic statistics is worth learning on its own because it stops you reporting noise as a finding.

### How long does it take to learn data analytics?

The two Kaggle courses for SQL and pandas are about 7 hours of lessons in total, but lessons are only the start. Most people need several months of regular practice and a few finished projects before their work looks convincing to an employer.

### Is Excel still needed for a data analyst?

Yes for most roles. Spreadsheets are still where many teams look at data first, and many tests ask for pivot tables and lookups. Treat Excel or Google Sheets as the first skill, not the last, and add SQL and one dashboard tool quickly after it.

### Will AI replace data analysts?

Evidence so far points to changed entry-level hiring rather than disappearance: a Stanford study of US payroll data finds weaker employment among younger workers in AI-exposed jobs. Analysts who define metrics, check data and explain results remain useful. Use AI tools daily, and always verify what they produce.

### What is the difference between a data analyst and a data scientist?

An analyst explains what happened and why using SQL, spreadsheets and dashboards. A data scientist adds prediction and experiment design, using Python and machine learning. Many data scientists start as analysts, and the skills overlap heavily, so SQL and statistics serve both roles.

## Sources checked

- [Intro to SQL: 3 hours, six lessons, no cost](https://www.kaggle.com/learn/intro-to-sql)
- [Pandas course: 4 hours, no cost](https://www.kaggle.com/learn/pandas)
- [Naukri JobSpeak, September 2026 data](https://www.naukri.com/blog/naukri-jobspeak-october-2026-white-collar-hiring-ai-ml-growth/)
- [Stanford paper revised 12 Aug 2026](https://digitaleconomy.stanford.edu/publications/canaries-in-the-coal-mine/)
- [WEF fastest-growing roles](https://www.weforum.org/publications/the-future-of-jobs-report-2025/)
- [Looker Studio renamed Data Studio](https://cloud.google.com/blog/products/data-analytics/looker-studio-is-data-studio)
