How to Become a Data Analyst in 2026: Skills & Portfolio Roadmap

You do not need a computer science background to work towards becoming a data analyst in 2026. By building practical skills in Excel, SQL, Python, data visualisation, and AI, you can start analysing data and creating projects that demonstrate your abilities. PW Earners — Data Analytics with AI can help you build these skills through practical learning.

The World Economic Forum's Future of Jobs Report 2025 lists Big Data Specialists among the fastest-growing roles, while Data Entry Clerks are among the fastest-declining. This makes one shift particularly important: move beyond simply handling data and develop the skills to analyse it, interpret it, and turn it into useful insights.

Why Data Analysis Is Different From Data Entry

The WEF's growing-versus-declining role data provides a useful way to understand what you should actually prepare for.

The WEF does not say that every data analyst role will grow or that a degree is unnecessary for every employer. Its findings instead show a broader shift in skills: AI and big data are rising rapidly, while analytical thinking remains the most sought-after core skill among surveyed employers, with seven in ten considering it essential in 2025.

That is the direction your learning roadmap should follow.

Can You Become a Data Analyst Without a Degree?

You can build data analytics skills without having a CS or data analytics degree. However, individual employers can still set their own education requirements, so you should not assume that a portfolio automatically removes degree requirements everywhere.

If you are entering analytics from another field, concentrate on building evidence that you can:

• clean and organise raw data,

• use spreadsheets confidently,

• query datasets using SQL,

• apply basic statistics correctly,

• identify patterns and trends,

• build understandable visualisations,

• use AI without blindly trusting its output,

• explain findings in simple language, and

• connect your findings to the original problem.

Your portfolio should eventually demonstrate these abilities rather than merely list them under a Skills heading on your resume.

What Skills Do You Need to Become a Data Analyst in 2026?

You do not need to learn every analytics platform available. Build a compact skill stack that allows you to complete an analysis from raw data to final recommendation.

You can build these skills progressively. There is little benefit in waiting until you have "completed" Excel, SQL, Python, Power BI, statistics, and AI before attempting your first project.

Data Analyst Roadmap for India: Where Should You Start?

If you are building a data analyst roadmap in India as a non-CS graduate or career-switcher, start with accessible tools before adding programming and AI.

Stage 1: Build Your Spreadsheet Foundation

Excel helps you understand rows, columns, data types, formulas, cleaning, aggregation, and basic visualisation without adding programming complexity.

Practise:

• sorting and filtering,

• removing duplicates,

• handling missing values,

• IF statements,

• SUMIF/SUMIFS,

• COUNTIF/COUNTIFS,

• XLOOKUP or other lookup functions,

• PivotTables,

• basic charts, and

• simple dashboards.

Milestone: Take a messy spreadsheet and turn it into a clean summary answering three or four specific questions.

Stage 2: Learn SQL for Asking Questions From Data

SQL moves you beyond manually inspecting spreadsheets and teaches you to retrieve the information you actually need.

Focus on:

• SELECT and FROM,

• WHERE,

• ORDER BY,

• GROUP BY,

• aggregate functions,

• JOINs,

• CASE statements,

• subqueries, and

• window functions as you progress.

Do not practise SQL only as isolated syntax exercises. Start with a question such as "Which product categories contributed the most revenue each quarter?" and work backwards to the query required.

Milestone: Take a dataset and answer 8–10 business questions using your own SQL queries.

Stage 3: Understand the Statistics Behind Your Analysis

You do not need advanced mathematics for every entry-level analytics task, but you need enough statistics to avoid misreading your results.

Build confidence with:

• mean, median and mode,

• percentages and percentage change,

• range and variance,

• standard deviation,

• distributions,

• correlation,

• outliers, and

• basic sampling concepts.

Milestone: Explain what your calculations mean in plain language instead of only reporting the numbers.

Stage 4: Learn Data Visualisation and a BI Tool

Choose Power BI or Tableau initially instead of dividing your time between several platforms.

Practise:

• importing data,

• transforming it,

• creating relationships,

• calculating useful measures,

• choosing appropriate charts,

• building filters,

• presenting KPIs, and

• designing a readable dashboard.

Milestone: Build a dashboard where someone unfamiliar with your dataset can identify the important findings without needing you to explain every chart.

Stage 5: Add Python and AI to Your Workflow

Once you understand the basic analysis process, Python and AI can help you handle more complex or repetitive work.

With Python, you can gradually practise:

• Pandas,

• importing datasets,

• cleaning data,

• filtering and grouping,

• exploratory analysis,

• automation, and

• basic visualisation.

With AI, you can experiment with:

• explaining unfamiliar SQL,

• troubleshooting formulas,

• suggesting approaches to cleaning,

• drafting queries that you then verify,

• documenting your analysis,

• brainstorming questions to investigate, and

• summarising findings that you have already validated.

The key rule is simple: AI can assist your analysis, but you remain responsible for checking it.

Why AI and Big Data Skills Matter in This Roadmap

The WEF places AI and big data at the top of its fastest-growing skills list, ahead of networks and cybersecurity and technological literacy. It also reports that 39% of workers' existing skill sets are expected to be transformed or become outdated between 2025 and 2030.

This does not mean you should replace Excel, SQL, or statistics with AI prompts. Instead, build AI literacy on top of analytical fundamentals.

A useful division is:

An analyst who can generate SQL with AI but cannot tell whether the query is wrong has not solved the underlying skills problem.

How to Build a Data Analyst Portfolio Without Experience

For someone trying to become a data analyst without a degree, the portfolio needs to show more than attractive dashboards. It should provide evidence of how you think.

Build three different project types so each one proves a different part of your skill set.

Project 1: Sales and Business Performance Analysis

Example question: Which products, categories, regions, or months contribute most to sales performance?

What to demonstrate:

• spreadsheet or SQL cleaning,

• JOINs and aggregation,

• revenue and growth calculations,

• trend analysis,

• dashboard creation,

• KPI selection, and

• written business observations.

Public data sources: Look for suitable retail, e-commerce, or transactional datasets through sources such as Kaggle, government open-data portals, or public dataset repositories.

Portfolio output: Include your cleaned dataset or SQL queries, dashboard, 3–5 important findings, and a short explanation of what someone could investigate or act on next.

Project 2: Public or Government Data Analysis

Example question: How has a measurable indicator changed across states, districts, years, or demographic groups?

You could work with publicly available data related to transport, education, population, employment, agriculture, health systems, or another area that interests you.

What to demonstrate:

• handling larger real-world datasets,

• missing-value treatment,

• grouping and comparison,

• percentage calculations,

• trend analysis,

• careful interpretation, and

• clear data visualisation.

Public data sources: India's Open Government Data Platform (data.gov.in), ministry or department datasets, and other official statistical portals can provide useful material.

Portfolio output: Create a short analysis with a clear research question, methodology, charts or dashboard, findings, limitations, and source attribution.

Project 3: Customer or Marketing Analysis

Example question: Which customer segments behave differently, or how does campaign performance vary across channels or groups?

What to demonstrate:

• segmentation,

• SQL querying,

• calculated metrics,

• comparisons,

• dashboard filters,

• pattern identification, and

• recommendation writing.

Public data sources: Public marketing, customer behaviour, or e-commerce datasets on Kaggle and other open-data repositories can work well.

Portfolio output: Show how you moved from raw customer-level data to segments, findings, and practical recommendations.

What Should Every Data Analyst Portfolio Project Include?

Do not upload a dashboard with a title and expect it to explain your analytical ability. Structure each project like a small case study.

Use this seven-part format:

1. Problem: What question were you trying to answer?

2. Dataset: Where did the data come from, and what does it contain?

3. Cleaning: What quality problems did you identify?

4. Analysis: What formulas, SQL queries, statistics, or techniques did you use?

5. Visualisation: Why did you choose these charts or dashboard elements?

6. Findings: What did the analysis reveal?

7. Recommendation or Next Step: What could someone investigate or do based on the findings?

This is where you demonstrate the difference between working with data and analysing data.

When Should You Start Applying for Data Analyst Roles?

Do not wait until you have mastered every analytics technology. A better checkpoint is whether you can complete a small analysis independently and explain your decisions.

Before you begin applying, try to have:

• 3 strong portfolio projects covering different problems,

• working knowledge of Excel and SQL,

• at least one Power BI or Tableau dashboard,

• evidence of data cleaning and analysis,

• clear written insights,

• basic statistical understanding,

• examples of responsible AI-assisted work, and

• a resume that links directly to your relevant projects.

Remember that individual job requirements vary. Some employers may specify particular qualifications, tools, industries, or degrees. Read each role carefully rather than assuming one portfolio will satisfy every position.

Learn Data Analytics With AI Through PW Earners

If you want guided practice while following this roadmap, the PW Earners — Data Analytics with AI course can help you build analytics and AI skills together. The beginner-friendly course covers Microsoft Excel, SQL, Python, Statistics, Power BI, Tableau, data cleaning, visualisation, dashboard creation, and AI tools for analytics, along with assignments and capstone projects.

Use the course alongside your portfolio rather than treating course completion as the final goal. As you progress, focus on:

• applying Excel and SQL to datasets beyond class exercises,

• creating dashboards that answer a specific question,

• using Python for cleaning and analysing data,

• practising AI-assisted analytics while checking the results yourself,

• completing capstone and independent projects, and

• turning your strongest projects into portfolio case studies.

This approach can help you move beyond simply knowing analytics tools to showing how you use data to investigate a problem, find insights, and communicate your findings clearly.

Becoming a data analyst without a degree in 2026 is ultimately about moving from handling data to analysing it. The WEF's contrast between fast-growing big data roles and declining data-entry roles reinforces why your roadmap should prioritise Excel, SQL, statistics, visualisation, AI literacy, analytical thinking, and projects that demonstrate how you solve problems. Build these skills through practice, document your strongest work in a focused portfolio, and use PW Earners — Data Analytics with AI to develop your analytics and AI capabilities through guided learning while continuing to create projects of your own.