Day 01 Part 1 — Power BI Basics¶

Power BI is Microsoft's business intelligence platform — and the most in-demand BI tool in the job market. This session takes you from zero to building your first interactive report: connecting data, shaping it in Power Query, modelling relationships, and writing your first DAX measures.

Learning Objectives¶

Understand the Power BI workflow: Get Data → Transform → Model → Visualise → Publish
Connect to data sources and clean data in Power Query
Build a data model with relationships
Write basic DAX measures
Create your first interactive report

The Power BI Ecosystem¶

Component	What it does
Power BI Desktop	Free Windows app where you build reports (your main tool)
Power BI Service	Cloud platform for publishing and sharing (app.powerbi.com)
Power BI Mobile	View reports on phone/tablet
Power Query	Built-in ETL engine for cleaning data
DAX	Formula language for measures and calculated columns

The Workflow¶

1. GET DATA      → connect to Excel, SQL, web, etc.
2. TRANSFORM     → clean in Power Query Editor
3. MODEL         → define relationships between tables
4. VISUALISE     → drag fields onto the report canvas
5. PUBLISH       → upload to Power BI Service to share

Step 1 — Get Data¶

Home → Get Data → choose source

Common sources: Excel, CSV, SQL Server, PostgreSQL, SharePoint, Web, Folder.

For analysts

Connect to a folder of CSVs (Get Data → Folder) to automatically combine all monthly files — and re-combine them on every refresh when new files are added.

Step 2 — Power Query (Transform)¶

Power Query records every cleaning step. When you refresh, all steps re-run automatically on the new data — this is the killer feature over manual Excel cleaning.

Common transformations: - Remove blank rows / columns - Change data types - Split columns by delimiter - Pivot / unpivot - Merge queries (join tables) - Group by (aggregate) - Add conditional columns

Set column data types early

The first thing to check in Power Query: are dates recognised as dates, numbers as numbers? Wrong types break everything downstream. Fix them in Power Query, not in the report.

Step 3 — Data Model and Relationships¶

In Model view, drag a field from one table to the matching field in another to create a relationship.

Star schema is the gold standard: - One central fact table (transactions, orders, sales) - Surrounding dimension tables (date, product, customer, region) - Relationships flow from the "one" side (dimension) to the "many" side (fact)

dim_date ──┐
dim_product├── fact_sales
dim_region ┘

Always create a dedicated Date table

Power BI's time intelligence (YTD, YoY, MoM) requires a proper date table. Create one with DateTable = CALENDAR(...), mark it as a date table, and relate it to your fact table's date column.

Step 4 — Your First DAX Measures¶

DAX (Data Analysis Expressions) is the formula language. Two types of calculations:

Calculated column (computed at refresh, stored, row context):

Order Size = IF(Sales[Total] > 300, "Large", "Small")

Measure (computed at query time, respects filters — prefer these):

Total Revenue = SUM(Sales[Amount])
Total Orders = DISTINCTCOUNT(Sales[OrderID])
Avg Order Value = DIVIDE([Total Revenue], [Total Orders], 0)

Measures over calculated columns

Use measures for aggregations (sums, averages, counts). They don't consume model memory and they respond to slicers and filters. Use calculated columns only for row-level categorisation you need to slice by.

Step 5 — Build a Report¶

Drag fields onto the canvas. Power BI picks a visual; change it in the Visualizations pane.

Essential visuals: - Card — a single big number (KPI) - Clustered column / bar — comparison - Line — trend over time - Table / Matrix — detail - Slicer — interactive filter - Map — geographic data

Practice Exercises¶

Warm-up 1. Connect Power BI Desktop to a CSV (use any project dataset). Load it. 2. In Power Query, change a text-date column to Date type and remove any blank rows. 3. Create a measure Total Revenue = SUM(...) and display it in a Card visual.

Main 4. Build a star schema: load a sales fact table and a separate date dimension. Relate them. 5. Create measures for Total Revenue, Order Count, and Average Order Value. 6. Build a report page with: a KPI card row, a revenue-by-month line chart, and a revenue-by-category bar chart. Add a slicer.

Stretch 7. Add a YoY growth measure using SAMEPERIODLASTYEAR. 8. Connect to a folder of multiple CSVs and combine them automatically.

Interview Questions¶

[Beginner] What is the difference between Power BI Desktop and Power BI Service?

[Beginner] What is Power Query used for?

[Mid-level] What is the difference between a calculated column and a measure? When do you use each?

[Mid-level] What is a star schema and why is it preferred in Power BI?

[Senior] Why does Power BI's time intelligence require a dedicated date table, and how do you set one up correctly?

← Week 01: Statistics · Next: Power BI Dashboards →