Data Analyst
Become a production-ready data analyst — SQL, Excel, statistics, Python/pandas, visualization with Tableau/Power BI, A/B testing, and storytelling. 16 phases from fundamentals to advanced analytics.
Your Progress
Complete all 16 phases to master this track
Become a production-ready data analyst — SQL, Excel, statistics, Python/pandas, visualization with Tableau/Power BI, A/B testing, and storytelling. 16 phases from fundamentals to advanced analytics.
Phase 1: Analyst Role and Data Lifecycle
Understand the analyst role, stakeholders, and how data moves through its lifecycle.
Analyst Role and Responsibilities
Map what data analysts own versus data scientists and engineers on a team.
Data Lifecycle Stages
Trace records from collection and storage through processing, analysis, and retention.
Framing Stakeholder Questions
Translate a vague request like 'why did sales drop' into a testable analysis brief.
Metric Definition and Governance
Define canonical metrics so teams stop reporting conflicting revenue or user numbers.
Triaging Analysis Requests
Prioritize incoming requests by impact, effort, and data readiness using a simple rubric.
Phase 2: Spreadsheets and Excel
Build reliable spreadsheets with formulas, lookups, pivots, and repeatable import steps.
Workbook Structure and Formulas
Build tidy sheets with cell references, ranges, and core functions like SUMIFS and IFERROR.
Lookup Functions in Excel
Join tables in spreadsheets using XLOOKUP, INDEX-MATCH, and robust not-found handling.
Pivot Tables for Quick Analysis
Slice, group, and summarize large exports with pivot tables and calculated fields.
Data Validation and Error Guards
Constrain inputs with dropdowns, rules, and conditional formatting to prevent bad entries.
Power Query Basics
Repeatable import, cleanup, and shaping steps that refresh instead of manual rework.
Phase 3: Data Cleaning and Transformation
Turn messy exports into analysis-ready data through profiling and documented cleaning steps.
Handling Missing Values
Audit blanks and nulls, then choose deletion, imputation, or flagging with clear rationale.
Duplicates and Outliers
Detect repeated keys and extreme values using grouping, quantiles, and domain thresholds.
String and Date Normalization
Standardize casing, whitespace, categories, and mixed date formats into canonical values.
Schema and Type Consistency
Align column types, units, and keys so joins and aggregations behave predictably.
Documented Cleaning Pipelines
Package cleaning steps as scripted, reviewable transformations with a change log.
Phase 4: SQL SELECT, WHERE, and ORDER
Query precisely with SELECT, WHERE, pattern filters, and deterministic ordering.
SELECT and Column Control
Retrieve exact columns, alias them, and compute derived fields in the projection.
WHERE Clause Filtering
Restrict rows with comparison operators, ranges, lists, and boolean logic.
ORDER BY and LIMIT
Sort results deterministically and cap output with LIMIT for top-N reporting.
Operators and Pattern Matching
Use LIKE, ILIKE, and range predicates to slice text and numeric segments.
Null Semantics in SQL
Handle three-valued logic so counts and filters do not silently drop or include rows.
Phase 5: SQL Joins and Grouping
Combine tables with joins and summarize them with grouped aggregations correctly.
Inner and Outer Joins
Combine tables with INNER, LEFT, RIGHT, and FULL joins matched on business keys.
Join Cardinality and Fan-out
Diagnose inflated totals caused by one-to-many joins and duplicated fact rows.
GROUP BY Aggregations
Summarize rows by dimensions using COUNT, SUM, AVG, MIN, MAX, and DISTINCT counts.
HAVING vs WHERE
Filter aggregated groups with HAVING while WHERE prunes rows before grouping.
Many-to-Many Bridge Tables
Model campaigns, products, or tags linked many-to-many without double counting.
Phase 6: Subqueries, CTEs, and Window Functions
Handle multi-step logic with subqueries, CTEs, and window functions for advanced SQL analysis.
Subqueries in Practice
Nest queries for filters, comparisons, and exists checks when CTEs are unavailable.
CTE Chains and Readability
Stage multi-step logic with WITH clauses that read top to bottom like a pipeline.
Window Function Basics
Compute across related rows with OVER, partitions, and ordering without collapsing rows.
Ranking and Offset Windows
Rank rows and fetch previous or next values using ROW_NUMBER, RANK, and LAG.
Running Totals and Cohort Metrics
Build cumulative sums, moving averages, and first-touch dates per user or entity.
Phase 7: Descriptive Statistics
Summarize datasets with measures of center, spread, shape, and association.
Measures of Central Tendency
Compare mean, median, and mode and pick the right one for skewed business data.
Spread and Variability
Quantify dispersion with range, variance, standard deviation, and IQR.
Distribution Shape
Read skewness, kurtosis, and histograms to understand how values are spread.
Correlation and Covariance
Measure linear association between metrics and interpret its limits correctly.
Exploratory Data Analysis Reports
Structure an EDA deliverable with profiles, segments, anomalies, and open questions.
Phase 8: Probability Essentials
Apply probability rules, Bayes reasoning, and simulation to quantify uncertainty.
Conditional Probability Basics
Apply addition, multiplication, and Bayes rules to event-based business questions.
Bayes Reasoning for Metrics
Interpret conversion rates and error rates as updated beliefs with base rates.
Expectation and Variance
Compute expected value and variance for discrete outcomes and simple forecasts.
Common Probability Distributions
Recognize Bernoulli, binomial, Poisson, and normal distributions in real data.
Monte Carlo Simulation
Estimate uncertain outcomes by resampling random inputs thousands of times.
Phase 9: Sampling and Distributions
Design samples and sampling distributions that support valid inference.
Sampling Methods
Choose random, stratified, cluster, or systematic samples that match the question.
Sampling Bias and Representativeness
Spot selection bias, survivorship bias, and coverage gaps in observed data.
Central Limit Theorem
Use the CLT to justify normal approximations for means and proportions.
Building Confidence Intervals
Construct intervals for means and proportions with correct level and margin.
Phase 10: Inferential Testing and Confidence Intervals
Run hypothesis tests, build intervals, and balance errors and power rigorously.
Hypothesis Testing Framework
State null and alternative hypotheses, choose a test, and interpret the p-value.
T-Tests in Practice
Run one-sample, paired, and two-sample t-tests for means with assumption checks.
ANOVA for Multiple Groups
Compare means across three or more segments and follow up with post-hoc tests.
Type Errors and Statistical Power
Balance false positives, false negatives, sample size, and detectable effect size.
Multiple Comparison Corrections
Control false discovery when running many tests across segments and metrics.
Phase 11: A/B Testing
Design, analyze, and read out controlled experiments that drive product decisions.
Experiment Design
Define hypothesis, randomization unit, duration, and success criteria before launch.
Metrics and Guardrails
Pick a primary metric, secondary metrics, and guardrails that protect the business.
Sample Size and Power
Calculate required sample size from baseline rate, MDE, alpha, and power.
Sequential Testing and Peeking
Avoid inflated error rates from repeated checks using fixed horizons or always-valid methods.
Experiment Readout
Analyze results, check validity, and write a decision-focused readout document.
Phase 12: Python and pandas Basics
Load and manipulate data in Python using pandas for repeatable analysis workflows.
Python Environment Setup
Install and manage Python, Jupyter, and virtual environments for analysis work.
Core Python Data Structures
Use lists, dictionaries, sets, and comprehensions to shape analysis data in memory.
Series and DataFrames
Create tabular structures, inspect them, and select columns with dtypes in mind.
Filtering and Selecting Rows
Query rows with boolean masks, query strings, and safe chaining patterns.
Reading and Writing Data
Load CSV, Excel, and databases and export results with correct parsing options.
Phase 13: pandas Aggregation and Reshaping
Aggregate, join, reshape, and optimize dataframes for real analysis tasks.
GroupBy and Aggregation
Split, apply, and combine operations with named aggregations and multiple metrics.
Merging and Joining Frames
Combine datasets with merge, validate relationships, and control join direction.
Melt and Pivot Reshaping
Switch between wide and tidy layouts using melt, pivot_table, and stack operations.
Rolling and Expanding Windows
Compute moving averages, cumulative metrics, and lag features over time series.
Performance and Vectorization
Speed up analysis by avoiding Python loops and using apply sparingly.
Phase 14: Visualization Principles
Design clear, honest, and accessible charts that communicate analytical findings.
Choosing Chart Types
Match line, bar, scatter, and distribution charts to the analytical question.
Color, Hierarchy, and Annotation
Guide attention with deliberate color, sizing, and direct labels on charts.
Dashboard Layout Design
Arrange KPI, trend, and detail zones so questions are answered in reading order.
Auditing Misleading Charts
Catch truncated axes, cherry-picked ranges, and dual-axis tricks before publishing.
Accessible Visualization
Choose colorblind-safe palettes and readable contrast for inclusive dashboards.
Phase 15: Tableau and Power BI Dashboards
Build performant, interactive Tableau and Power BI dashboards for stakeholders.
Tableau Calculated Fields
Build ratios, level-of-detail expressions, and conditional logic inside Tableau.
Power BI DAX Measures
Write time intelligence and filter-aware measures with CALCULATE in Power BI.
Interactive Filters and Drill-Through
Add slicers, parameters, and drill-through pages that respect user context.
Dashboard Performance and Publishing
Optimize queries, data models, and refresh schedules for reliable distribution.
Dashboard QA and Review
Run reconciliation checks and a stakeholder review before announcing a dashboard.
Phase 16: Storytelling and Capstone Analysis
Communicate recommendations and ship a portfolio-ready capstone analysis.
Narrative Structure for Insights
Organize findings into situation, complication, question, and resolution arcs.
Executive Summary Writing
Compress a full analysis into a one-page summary busy leaders will actually read.
Framing Recommendations
Turn analysis into specific actions with owners, expected impact, and trade-offs.
End-to-End Capstone Analysis
Deliver a full analysis from raw data through cleaning, SQL, Python, and visuals.
Portfolio and Presentation Review
Package projects into a portfolio and present findings with confidence and clarity.