Become a performance analyst.

Who this guide is for

Beginners and early-career analysts who want to work with business KPIs, operational performance, commercial questions or management reporting. You do not need a computer science degree; curiosity, careful reasoning and consistent practice matter. Comfort with spreadsheets is useful but can be learned along the way.

Suggested weekly time: 5–7 hours. If you have less time, keep the same sequence and stretch the schedule. Treat the roadmap as a guide, not a job guarantee.

Stage 1 · Foundations (Weeks 1–4)

Learn: business questions, rows and columns, data types, spreadsheet formulas, cleaning, summary statistics and clear charts.

Build: a one-page KPI scorecard and a short note that explains one trend and one limitation.

Resources: Beginner course, Google Sheets training, Khan Academy statistics.

Stage 2 · SQL and trustworthy metrics (Weeks 5–10)

Learn: SELECT, filtering, aggregation, joins, CTEs, window functions, date logic and reconciliation. Practise writing queries from a business question rather than copying syntax.

Build: a repeatable report with a KPI dictionary, validation checks and documented assumptions.

Resources: SQLBolt interactive lessons, PostgreSQL tutorial, and the site’s SQL course pages.

Stage 3 · BI and dashboarding (Weeks 11–16)

Tool to learn first: Microsoft Power BI. Practise importing data, modelling relationships, writing basic DAX measures, designing drill-downs and checking that totals reconcile.

Build: a dashboard for a clear audience and decision. Include definitions, refresh date, target comparisons and a short executive summary.

Resources: Microsoft Learn: Power BI.

Stage 4 · Analysis and automation (Weeks 17–24)

Learn: Python with pandas, data-quality checks, segmentation, variance analysis, descriptive statistics and simple automation.

Build: an analysis notebook or script that can be run again on a fresh input file.

Resources: Python tutorial, pandas getting started.

Stage 5 · Portfolio and consulting communication (Weeks 25–36)

Learn: hypothesis-led analysis, issue trees, prioritisation, stakeholder questions, limitations and clear writing.

Build: two end-to-end case studies with code, sample data, a visual, calculations, recommendations and a section on what the evidence cannot prove.

Resources: The Analyst projects and Google’s technical writing courses.

Stage 6 · Job readiness (Weeks 37–52)

Review relevant job descriptions, identify repeated skill requirements, practise timed SQL and business cases, and prepare honest project walkthroughs. Tailor your CV to evidence you can explain and demonstrate. Quantify outcomes only when they are real and substantiated.

A weekly routine (5–7 hours)

• 2 hours: learn one topic and take short notes.
• 2 hours: solve a practical task using sample data.
• 1 hour: check calculations, assumptions and data quality.
• 1–2 hours: explain the result in writing and improve portfolio evidence.

What makes a strong portfolio project?

It has a clear question, documented dataset provenance, reproducible code, meaningful checks, a useful visual, an actionable conclusion and honest limitations. Synthetic data is entirely acceptable for demonstrating a method when it is labelled plainly.

Last updated: 17 October 2026