Gigsouk
Post a job

How to Become a Data Analyst

A practical path from zero to your first analyst job: the skills to learn, in order, the projects to build, and how to apply.

Data analysts turn raw data into answers a business can act on: why sales dipped, which customers are likely to churn, whether a new feature worked. It's one of the most accessible ways into tech — many analysts come from finance, marketing, operations, science or teaching — because the core skills can be learned in months, not years.

You don't need a degree in computer science. You need to be good at SQL, comfortable with spreadsheets and basic statistics, able to visualise results, and — most importantly — able to explain what the numbers mean. Here's the order to learn them in.

The five-phase roadmap

  1. 1

    Spreadsheets and data thinking

    2–4 weeks

    Get comfortable turning a business question into numbers.

    • Lookups (XLOOKUP, INDEX/MATCH), SUMIFS and pivot tables
    • Cleaning messy data: types, duplicates, blanks
    • Basic charts and when to use each one

    Build: A monthly sales report from a public dataset, with a one-paragraph summary of what changed and why.

    Excel cheat sheet →CSV cleaner →

  2. 2

    SQL — the core skill

    6–8 weeks

    Answer any reasonable business question from a database on your own.

    • SELECT, WHERE, GROUP BY, HAVING and every JOIN type
    • CTEs, subqueries and CASE WHEN
    • Window functions: ROW_NUMBER, RANK, LAG, running totals
    • Date handling and avoiding duplicate-row fan-out

    Build: Answer 20 business questions on an e-commerce dataset — revenue trends, top customers, retention cohorts.

    SQL joins visualizer →SQL case studies →SQL formatter →

  3. 3

    Statistics for analysts

    3–4 weeks

    Know when a difference is real and how to explain uncertainty.

    • Distributions, mean vs median, variance and percentiles
    • Sampling, confidence intervals and hypothesis tests
    • A/B testing: significance, power and sample size
    • Correlation vs causation

    Build: Analyse a real (or simulated) A/B test and write a recommendation with the confidence interval.

    A/B test calculator →Sample size calculator →Descriptive stats →

  4. 4

    Visualisation and storytelling

    3–4 weeks

    Make findings obvious to someone who has 30 seconds.

    • One BI tool in depth: Tableau, Power BI or Looker
    • Dashboard design: headline metrics first, then drill-downs
    • Writing an insight: the answer, the evidence, the recommendation

    Build: A published, interactive dashboard on a topic you care about, with a short write-up.

  5. 5

    Portfolio, Python and the job search

    4–8 weeks

    Prove you can do the job, then get in front of employers.

    • Python with pandas for cleaning and automation (a strong plus, not always required)
    • 2–3 portfolio projects, each with a clear question, method and conclusion
    • Resume bullets with measurable results
    • SQL, statistics, case and behavioural interview practice

    Build: A portfolio page linking your projects, and a resume tailored to each application.

    Skills checklist →Resume bullet builder →Resume ↔ job matcher →Interview questions →

Total: roughly 4–7 months part-time. Phases overlap — keep practising SQL throughout.

What makes a portfolio project stand out

  • A real question — “Which neighbourhoods have the slowest 311 response times, and why?” beats “Exploring NYC data”.
  • Messy data you cleaned yourself, with your steps documented.
  • A clear conclusion and recommendation up front — not just charts.
  • Your SQL or code in a public repo, so reviewers can see how you work.

Applying for your first analyst job

  • Search beyond “Data Analyst”: Business Analyst, Reporting Analyst, Operations Analyst, Marketing Analyst and BI Analyst roles use the same skills.
  • Tailor your resume to each job's keywords — and check it with our resume ↔ job matcher.
  • An internal move is often easiest: volunteer for reporting work in your current role and put the results on your resume.

Ready to look? Browse data & AI jobs, entry-level data jobs, remote data jobs or jobs that use SQL, and see what data roles pay.

Frequently asked questions

How long does it take to become a data analyst?

Roughly 4–7 months of part-time study for most people, depending on your starting point and how much time you can give it.

Do I need a degree to become a data analyst?

Not necessarily. Many employers hire on demonstrated skills — SQL, spreadsheets, statistics and a portfolio of real projects.

What should I learn first?

Spreadsheets, then SQL — the most important technical skill for analysts — followed by statistics, visualisation and Python.

More free tools

All tools →

Put your skills to work

Thousands of open roles from companies hiring now — apply directly with the employer.

Data & AI jobsBrowse all jobs