CraftCourse logoCraftCourse
Get started
Roadmaps›Data Analyst roadmap

Data Analyst Roadmap

A data analyst turns raw, messy data into answers a business can actually act on: which product is losing customers, whether a marketing campaign worked, what's driving a metric up or down. The job is less about knowing every tool and more about a specific chain of skills working together, pulling the right data, analyzing it correctly, and explaining what it means to someone who doesn't want to see a query, they want a decision.

This is one of the more accessible entry points into a data career because the core tools, spreadsheets, SQL and a BI tool like Power BI, are genuinely learnable by a complete beginner in months, not years, and the skill transfers across almost every industry. The candidates who get hired quickly aren't the ones who know the most statistics theory, they're the ones who can take a vague business question, figure out what data actually answers it, and present that answer clearly, which is exactly the chain of skills this roadmap builds.

This path starts from zero and goes through spreadsheets and SQL as the non-negotiable foundation, into statistics fundamentals and Python for analysis, then data visualization principles, Power BI and dashboard design, and finally the A/B testing and stakeholder communication skills that separate someone who can run a query from someone a business actually trusts to answer their questions.

Drag or scroll to explore, click a node to learn more · 9 sections · 15 courses · free and self paced

How to use this roadmap

  1. Go through this roadmap in order if you're starting from zero, SQL and spreadsheets are assumed knowledge in nearly every later section.
  2. Don't rush past statistics fundamentals to get to the flashier tools, misreading a result because of a stats gap is one of the most common ways beginner analysts lose credibility with stakeholders.
  3. Each course's concepts are short, standalone lessons, work through them like a checklist rather than one long sitting.
  4. Practice on a real, messy dataset as you go, Kaggle and public government datasets work well, clean tutorial data doesn't prepare you for what a real job actually looks like.
  5. The stakeholder communication section near the end isn't a soft skill add-on, it's frequently the actual differentiator in data analyst interviews, don't skip it.
1

Data Analytics Landscape

What a Data Analyst Actually Does beginner: The real chain of work: pulling data, analyzing it, and explaining it to people who don't want to see the query.

  • Data Analyst vs Data Scientist: Answering defined business questions versus building predictive models.
  • The Analysis Workflow: The general path from a business question to a delivered answer.
  • Types of Analysis: Descriptive, diagnostic, predictive and prescriptive analysis, briefly.
2

Spreadsheets

Excel & Google Sheets Fundamentals beginner: The tool still used constantly for quick analysis and real business requests, not just a beginner stepping stone.

  • Formulas & Functions: The building blocks of any spreadsheet calculation or lookup.
  • VLOOKUP/XLOOKUP: Pulling matching data from another table or range.
  • Pivot Tables: Summarizing and aggregating data interactively without formulas.
  • Data Cleaning in Spreadsheets: Fixing inconsistent, duplicate or missing data before analysis.
3

SQL

SQL Fundamentals for Analysts beginner: How to actually pull the data you need out of a company's database.

  • SELECT & WHERE: Retrieving and filtering rows from a database table.
  • Aggregations (GROUP BY): Summarizing data into totals, averages and counts by category.
  • JOINs: Combining data from multiple related tables in one query.

Intermediate SQL for Analysis intermediate: The query patterns that show up constantly once analysis goes beyond simple aggregation.

  • Window Functions: Calculations across related rows, like running totals or rankings.
  • Common Table Expressions: Named subqueries that make complex analysis readable.
  • Subqueries: Nesting one query inside another for multi-step logic.
4

Statistics Fundamentals

Descriptive Statistics beginner: Summarizing a dataset's basic shape before drawing any conclusions from it.

  • Mean, Median, Mode: The three common ways to describe a dataset's center.
  • Standard Deviation & Variance: Measuring how spread out a dataset's values are.
  • Distributions: The shape describing how a dataset's values are spread.

Inferential Statistics Basics intermediate: Drawing conclusions about a larger population from a smaller sample, and avoiding common mistakes doing it.

  • Correlation vs Causation: Two variables moving together doesn't mean one causes the other.
  • Hypothesis Testing: A structured way to test whether an observed effect is likely real.
  • Statistical Significance: Whether a result is unlikely to have happened by random chance.
  • Confidence Intervals: A range likely to contain a true population value.
5

Python for Analysis

Python Fundamentals for Analysts beginner: Just enough Python to move into pandas, without needing a software engineering background.

  • Variables & Data Types: The basic building blocks of any Python program.
  • Lists & Dictionaries: Python's core structures for storing collections of data.
  • Loops & Functions: Repeating logic and packaging it into reusable pieces.

Data Analysis with pandas intermediate: The library that makes Python genuinely useful for real analysis work, beyond what SQL alone covers well.

  • DataFrames & Series: pandas's core structures for tabular data.
  • Filtering & Grouping Data: Slicing and summarizing data programmatically.
  • Merging Datasets: Combining multiple data sources into one DataFrame.
  • Handling Missing Data: Detecting and dealing with gaps in a real dataset.

Data Visualization with Python intermediate: Producing charts programmatically for deeper, more repeatable exploratory analysis.

  • Matplotlib Basics: Python's foundational plotting library.
  • Seaborn for Statistical Plots: A higher-level library built for statistical visualizations.
6

Data Visualization Principles

Data Visualization Principles intermediate: The design choices that make a chart actually communicate something, instead of just displaying numbers.

  • Choosing the Right Chart Type: Matching a chart type to the specific comparison being shown.
  • Avoiding Misleading Visuals: Common chart design choices that distort how data is perceived.
  • Color & Accessibility in Charts: Using color deliberately and readably for all viewers.
7

Power BI & Dashboards

Power BI Fundamentals intermediate: Connecting to data and building the reports most entry-level analyst roles expect fluency in.

  • Connecting Data Sources: Pulling data into Power BI from databases, files and other sources.
  • Data Modeling & Relationships: Connecting multiple tables correctly inside Power BI.
  • DAX Basics: Power BI's formula language for calculated measures.

Dashboard Design intermediate: Laying out a dashboard so a busy stakeholder can actually find the answer they need in seconds.

  • Dashboard Layout Principles: Organizing visuals so the most important information stands out.
  • Interactivity & Filters: Letting users explore data themselves within a dashboard.
  • Designing for the Audience: Tailoring a dashboard's detail level to who will actually use it.
8

A/B Testing

A/B Testing Fundamentals intermediate: Comparing two versions of something and knowing whether the difference is actually real.

  • Test Design Basics: Structuring an experiment so its results can be trusted.
  • Sample Size Considerations: Why a test needs enough data to detect a real effect reliably.
  • Interpreting Test Results: Reading a result correctly and avoiding common misreadings.
9

Stakeholder Communication

Translating Business Questions into Analysis intermediate: Turning a vague stakeholder ask into a concrete, answerable analytical question.

  • Clarifying Ambiguous Requests: Asking the right follow-up questions before starting an analysis.
  • Scoping an Analysis: Deciding what's realistically answerable given time and data.

Presenting Findings to Non-Technical Audiences advanced: Explaining what an analysis actually means to someone who doesn't want to see the query.

  • Leading with the Insight: Stating the conclusion first, not the analytical process.
  • Storytelling with Data: Structuring findings as a clear narrative, not a data dump.

Not satisfied with the roadmap? Create your own!

Sage

Frequently asked questions

Yes, and don't skip it. Excel and Google Sheets remain extremely common in real business environments, especially for smaller analyses, quick stakeholder requests and any team without a full data infrastructure. Interviewers frequently test basic spreadsheet fluency directly, and it's often the fastest tool for a genuinely small, one-off question.

Learn SQL first, it's how you'll actually get data out of a company's database in almost every entry-level analyst role, and it's a more universally required skill in job postings than Python. Learn Python with pandas after SQL, for analysis that goes beyond what SQL alone handles well, like more complex statistical work or repeated analysis workflows.

You need a solid working understanding of descriptive statistics, distributions, correlation versus causation, and basic hypothesis testing, enough to correctly interpret results and avoid common misreadings. You don't need graduate-level statistical theory, entry-level roles rarely require anything beyond what this roadmap's statistics section covers.

Power BI is a strong first choice given its very wide adoption across company sizes and its tight integration with Excel, which most beginners already know. Tableau remains common in certain industries and is worth knowing conceptually exists, but learning one BI tool deeply and being able to explain the general concepts, like relationships and measures, transfers quickly if a job uses a different tool.

Working consistently, most complete beginners following this roadmap reach a genuinely job ready level, including 2 to 3 portfolio projects with a real dashboard, in 4 to 6 months. This is generally faster than becoming job ready as a developer or engineer, since the tools involved have a shorter path to real competency.

No, data analyst hiring leans heavily on demonstrated skills through a portfolio, SQL and Excel tests during interviews, and the ability to walk through how you'd approach a business question. A strong portfolio with 2 to 3 real projects, including a dashboard and a written analysis, is often more persuasive to hiring managers than a degree alone.

A strong first project pulls a real, messy public dataset, cleans and analyzes it with SQL and Python, and presents findings through a Power BI dashboard with a short written summary aimed at a non-technical audience. The written summary and dashboard, not just the analysis code, are what most differentiate a portfolio piece employers actually remember.

A/B testing compares two versions of something, like a webpage or email, by showing each to a different group and measuring which performs better statistically. Data analysts are frequently asked to design, analyze or interpret A/B test results, and understanding the statistics behind it, like statistical significance, prevents common mistakes like calling a result meaningful when it's actually just noise.

The data analysts who get hired quickly and get trusted with real business questions are the ones who can move fluidly through the whole chain, pulling the right data, analyzing it correctly, and explaining it clearly, not just the ones who know the most SQL syntax. Working through this roadmap in order, and building real projects with messy data along the way, is what turns this list of tools into the practical judgment that gets you hired.