Data Analytics Academy
Start Day 1
Open source · MIT licensed · 0 prerequisites

Data Analytics Academy

Five days. Twenty hours. From "I've opened Excel a few times" to shipping an end-to-end analytics project on real data — Excel → SQL → Python → Power BI → AI.

20 hours Beginner level Self-paced
Get started · Day 1 → See the syllabus
What you'll learn

Six concrete skills you'll walk away with

Clean a messy CSV in Excel.

Tables, lookups, conditional aggregation, dates, pivots — the formula library every analyst leans on daily.

Query a real database with SQL.

Joins across 4+ tables, CTEs, aggregations, and the empty-string-vs-NULL gotchas that catch most beginners.

Reproduce SQL analysis in pandas.

Load, filter, group, and merge with row-count sanity checks — the habits that prevent silent wrong answers.

Build a 1-page interactive Power BI dashboard.

Data model, DAX measures, slicers — a deliverable a non-analyst will actually click through.

Direct an AI assistant to do real analytics.

The prompting templates, plan mode, and verification habits that make Claude Code a force multiplier instead of a confidently-wrong oracle.

Ship a stakeholder-ready 1-page report.

Findings backed by traceable numbers, embedded charts, and recommendations a head of operations can act on.

Skills you'll gain

The toolbox

Excel TablesXLOOKUPSUMIFSPivot tablesSQL SELECTGROUP BYJOINCTEsSQLitepandas DataFramesgroupby + aggpd.mergeJupyterPower Query (M)Data modelingDAX measuresCALCULATECross-filteringClaude CodeAI promptingPlan modeVerification habitsTheme extractionStakeholder reports
Details to know

Practical facts

Free, MIT-licensed

No signup, no email, no ads. Fork the repo and customise for your team.

20 hours, self-paced

Five days × four hours, but stretch or compress to suit your schedule.

Beginner-friendly

No coding background assumed. Day 3 starts from "import pandas as pd".

Real data, real capstone

~100K Olist orders threaded through every day; ships as a public 1-page report.

Curriculum

5 modules in this course

Each module is one day of teaching plus one hour of capstone application. Click a module to see its lessons.

Module 1 · Excel

4 hours · 6 lessons

Tables · structured refs · XLOOKUP · IFERROR · SUMIFS · pivot tables
  1. Tables & references ~30 min
  2. Math & aggregation ~30 min
  3. Logic & lookups ~30 min
  4. Dates, text & cleaning ~30 min
  5. Pivot tables & charts ~30 min
  6. Self-test ~15 min
Open the module →

Module 2 · SQL

4 hours · 6 lessons

SELECT · GROUP BY · JOIN · CTEs · NULL semantics · SQLite quirks
  1. SELECT, FROM, WHERE ~30 min
  2. Aggregation & GROUP BY ~30 min
  3. Joins ~40 min
  4. CTEs & multi-step queries ~30 min
  5. NULLs, types & SQLite quirks ~25 min
  6. Self-test ~15 min
Open the module →

Module 3 · Python

4 hours · 6 lessons

pandas · DataFrames · groupby · pd.merge · the row-count assertion
  1. Notebooks & loading data ~25 min
  2. Selecting & filtering ~30 min
  3. Aggregation & groupby ~30 min
  4. Merge & combine ~25 min
  5. Cleaning, dates, plots ~30 min
  6. Self-test ~15 min
Open the module →

Module 4 · Power BI

4 hours · 6 lessons

Power Query · the data model · DAX measures · interactive dashboard
  1. Get Data & Power Query ~30 min
  2. The data model ~25 min
  3. Building visuals ~30 min
  4. DAX measures ~30 min
  5. Interactivity, layout, polish ~25 min
  6. Self-test ~15 min
Open the module →

Module 5 · Claude Code

4 hours · 6 lessons

Prompting · plan mode · verification · text analytics · the final report
  1. Getting started ~20 min
  2. Prompting that works ~25 min
  3. Plan mode & verification ~25 min
  4. AI-assisted querying ~30 min
  5. Text analytics & the report ~30 min
  6. Self-test ~15 min
Open the module →
Sample lesson · Day 1 Lesson 1

Read a real lesson before you commit

curriculum / day1_excel / 01_tables.md

Lesson 1 — Tables & references

Time: ~30 min. You'll be able to:

  • Convert a range to an Excel Table with Ctrl+T and reference its columns by name
  • Use structured references like [@col] and Sales[Revenue] instead of B2:B1000
  • Tell when to use $A$1, A$1, $A1, and A1 — and stop guessing
  • Sort, filter, and freeze panes without breaking your formulas

Tables are the single most under-used Excel feature. Convert any data range to a Table with Ctrl+T (Windows) or ⌘+T (Mac). The moment you do, three things happen:

  1. Headers freeze automatically when you scroll past the first row.
  2. Formulas auto-extend when you add a row at the bottom — no copy-paste.
  3. You can reference columns by name: =SUM(Sales[Revenue]) instead of =SUM(B2:B1000).
Where this course fits

Free intensive vs paid certificates

You are here

This course

$0

  • 5 days · ~20 hours of teaching
  • Real dataset (~100K Brazilian e-commerce orders)
  • Capstone is a public 1-page report you can share
  • MIT licensed; fork the repo, customise for your team
  • No signup, no email, no ads, no tracking
  • Includes a full day on AI-assisted analytics

Paid certificate (Coursera / IBM / Google)

$400–$4,000

  • 6 months · ~240 hours
  • Toy datasets per module
  • Capstone is a PDF certificate
  • Curriculum is fixed; can't adapt it
  • Email signup, marketing reminders, paid upsells
  • AI usually a final-week module, not the workflow

Different tools for different jobs. Paid certs are better for formal hiring credentials. This course is better for fast portfolio impact and team upskilling.

Frequently asked

Questions students actually ask

How long does the course take?
Five days at four hours each — twenty hours of teaching and one hour of capstone application per day. It is self-paced, so you can stretch it across a fortnight or compress it into a long weekend.
Do I need coding experience?
No. Day 1 starts from "what is a Table" in Excel. Day 3 starts from import pandas as pd. If you have opened a spreadsheet a few times you are qualified.
Is it really free?
Yes. MIT licensed, no signup, no email capture, no ads, no upsell. Source on GitHub.
Can I do this on a Mac?
Mostly. Days 1, 2, 3, 5 work natively on macOS. Day 4 (Power BI Desktop) is Windows-only — alternatives noted in the lesson.
Is there a certificate?
No paper certificate. The capstone — a 1-page stakeholder report on real data — is the certificate. Push it to your GitHub, link from LinkedIn.
What if I get stuck?
Every lesson ends with a "Common pitfalls" section and a "Try it yourself" drill with collapsible solution. Open an issue for anything not covered.
Who built this and why?
Originally written for an internal analytics academy and released publicly under MIT. See What a data analyst actually does for the framing.
After the course

Where to go from here

Finished all five days plus the capstone? What to learn next — the books, courses, and habits that take a fluent beginner to a working analyst.

About

Built in the open

Built for an internal analytics academy and released publicly under MIT. Source, issues, contributions: github.com/scripts-and-tables/daa. Spotted a typo or have a better worked example? Open a PR.