Learn Data Pipelines with Phoebe

How one order becomes a dashboard - in plain English

At 11:41pm someone buys a $42 coffee gift bundle. At 9am an executive glances at a revenue number for four seconds. Six sessions walk the road between those two moments - the pickup, the shelf, the cleanup room, the screen, and the mornings it all quietly breaks. Zero code, zero jargon left undecoded, one pipeline floor you get to break yourself.

6sessions
6stages, source to screen
0lines of code required
45min per session
🧭 Session 1 · easiest

Follow the order: how one purchase becomes a dashboard

The six-stage journey every dashboard number travels, why the Friday spreadsheet always dies the same death, and one of your own numbers traced back to its birthplace.

▶ Start hereThe map 🛒 Session 2 · easiest

Where data is born: sources and the overnight pickup

App databases, click events, files, and third-party feeds - and the trucks that collect them. Batch versus streaming, decoded with a milk run and a conveyor belt.

Sources + ingest 🧼 Session 3 · building

The cleanup room: raw to ready inside the warehouse

Why the shelf keeps an untouched copy of everything, how raw data becomes trustworthy in layers, and the ETL-versus-ELT story told through cheap storage.

Store + transform 📊 Session 4 · building

The last mile: dashboards, definitions, and who eats the data

Metrics are decisions wearing numbers. The serving layer, the schedule that runs the whole floor, and why two teams report different revenue with both pipelines healthy.

Metrics + serve 🔧 Session 5 · building

When pipelines break: stale, wrong, and how you find out

The two faces of every failure, the four causes behind most incidents, the safeguards that catch each - and Daybreak's pipeline floor, yours to break, stage by stage.

🔧 The floorInteractive 🤝 Session 6 · deepening

People, tools, and next steps: talking to your data team

Who does what on a data team, the full tool-name decoder table, how to ask for data so you actually get it - and the honest ladder to the deeper courses.

The decoder
easiest - start here building - the floor deepening - making it yours
☕ The running case: Daybreak, order #8412. One coffee-subscription order - $42, bought at 11:41pm on a Sunday - travels the whole course: picked up at 2am, shelved raw at 2:15, cleaned at 4:30, counted at 5:45, on screen at 6:05. Session 5 hands you the wrench and lets you break every stop on its route.
The plain-English promise. Every tool name you have overheard - Snowflake, dbt, Airflow, Fivetran, Looker - gets decoded into a one-line job description. No session requires code, and where a topic deserves real depth, we point honestly at the deeper course (learn-data-warehouse, learn-sql, learn-data-engineering) instead of pretending 45 minutes covers it.

Choose your focus 🧭

Everyone starts at Session 1 for the map. After that, follow what you need - or walk the whole road in order.

🔥 I have a dashboard I do not trust 5→ 6
🗣 I just want to talk to my data team 1→ 6
🛤 The full journey 1→ 2→ 3→ 4→ 5→ 6

The knowledge map 🧠

The whole course at a glance - hover a session to spotlight its concepts, click any node to jump in.