The Practical Data Analyst.
Learn to solve business problems with data in the AI era — from spreadsheet fundamentals and SQL fluency to metric definitions, validation and influencing the decision.
Don't learn to compete with AI at writing SQL. Learn to become the analyst who can frame the problem, work confidently with the data, catch bad analysis and make the decision better.
Build the analyst layer by layer.
Start with analytical thinking and data foundations, build spreadsheet fluency, then learn SQL in two steps: core querying first, multi-table analytical SQL second.
Foundations
Badge lockedUnderstand the business question, the data model, the spreadsheet workflow and the query logic before touching a dashboard.
Think Like an Analyst
Turn vague stakeholder questions into measurable analytical problems, hypotheses and decision criteria.
Data Foundations
Learn grain, keys, event data, dimensions, facts, relationships and how business systems become tables.
Spreadsheet Foundations
Use practical Excel and Google Sheets formulas, cleaning, dates, validation, lookups, pivots and Power Query patterns.
SQL Foundations
Learn SQL from zero: select rows and columns, filter, sort, handle nulls, aggregate and group business data.
Advanced SQL for Analytics
Use joins, CTEs, window functions, deduplication and control queries to solve multi-table analytical problems safely.
Applied SQL Analysis
Use reusable SQL analysis patterns for cohorts, retention, funnels, churn, RFM, LTV, attribution, experiments, inventory and risk.
Business Metrics
Define metrics precisely so revenue, conversion, retention and churn mean the same thing to everyone.
Analysis
Badge lockedTurn raw data into trustworthy evidence and communicate it without hiding behind a chart.
Data Quality & Validation
Find duplicate rows, broken joins, missing data, bad denominators and results that fail simple sanity checks.
Visualization & Storytelling
Choose charts for the question, design useful dashboards and explain what a decision-maker needs to know.
Statistics for Analysts
Use distributions, confidence, variance, seasonality and correlation carefully in everyday analysis.
Python for Analysts
Use pandas and notebooks for cleaning, exploratory analysis and work that becomes awkward in SQL.
AI-Native Analytics
Badge lockedUse AI to move faster while keeping human ownership of definitions, validation and decisions.
AI-Assisted SQL & Analysis
Give AI the schema, business context and analytical brief it needs to generate useful first-pass work.
Evaluate the AI
Audit generated SQL, calculations and explanations for double counting, logic errors and invented conclusions.
Analyst Workflows & Agents
Design repeatable analytical workflows that combine metrics, verified queries, business rules and AI.
Business Domains
Badge lockedLearn how analytical thinking changes when the business model, customer journey and operating constraints change.
Product Analytics
Analyze activation, funnels, cohorts, retention, engagement and feature adoption.
Marketplace Analytics
Reason about demand, supply, matching, cancellations, incentives and geographic performance.
Marketing Analytics
Measure acquisition, CAC, ROAS, funnel conversion, attribution limits and customer quality.
E-commerce Analytics
Break down revenue, AOV, repeat purchase, categories, discounts, returns and refunds.
Operations Analytics
Investigate SLA, capacity, turnaround time, backlog, productivity and service quality.
Experiments & Causal Thinking
Separate correlation from causation and reason about A/B tests, controls, bias and confounding.
Career Lab
Badge lockedWork like an analyst when the request is ambiguous, the deadline is real and the answer has to influence a decision.
Stakeholders & Communication
Clarify requests, push back when needed, communicate uncertainty and turn evidence into recommendations.
Analytics Interviews
Practice SQL screens, metric questions, product cases, dashboard critiques and analytical communication.
Capstone: The Analyst Case
Solve one ambiguous business problem from raw data to SQL, visualization, executive memo and recommendation.
CASE ARCADE
Break the analysis on purpose, run the checks, and learn to recognize the failure pattern before it happens in a real dashboard.
Reading teaches the rule. Breaking the metric teaches the instinct.
Every lab has a red failure action and a green validation action. Predict what will happen before you click.
Grain Under Pressure
Join order-level revenue to item-level rows and watch a perfectly valid SUM become wrong.
The Join Explosion
Run a payment-attempt join that quietly multiplies order value, then diagnose it with control queries.
Metric Contract Debugger
Keep the same numerator but swap the denominator and watch the cancellation story change completely.
Dashboard Trust Incident
A city dashboard says orders collapsed 34%. Inject duplicates, null mappings and stale data, then decide whether to escalate.
Funnel Drop-off Detective
A redesign appears to hurt activation. Run a funnel diagnostic and separate real product friction from a tracking break.
Experiment Integrity Check
Treatment conversion looks better, but the assignment split is wrong. Run integrity checks before reading lift.
AI SQL Red Team
An AI-generated query runs and returns a polished answer. Run validation to uncover three analytical bugs.
EVERY LESSON FOLLOWS A REAL WORKFLOW
No isolated syntax drills. Each lesson starts with a business context and ends with a decision or communication task.
THE CAPSTONE
The final project is not “build a dashboard.” It is an ambiguous analyst request that has to become a defensible recommendation.
The Analyst Case.
You receive the business context, database, metric definitions and stakeholder request. You decide what to investigate, use SQL/Python/AI where useful, validate the result and present what the business should do next.