Datanalytics

9.1-Introduction to Power BI

4 min read

Title: Introduction to Power BI-Class Transcript and Study Notes

Date: 2025-10-29 11:50

Status: Child

Tag: Power BI data modeling data transformation data analysis

Overview

  • First Power BI class in Data Analytics Job Ready Program
  • Focus: Foundational concepts, data flow, installation
  • Includes a mini-project demo
  • Participants: Professionals from finance, IT, sales, HR, and students

Power BI Fundamentals

Power BI is a reporting tool for connecting to various data sources, transforming data, creating visualizations, and sharing reports online.

Key Features and Comparisons

  • Data Visualization: Builds dashboards with multiple tables and sheets, similar to Excel but with dynamic drilling (e.g., category to subcategory).
  • Dynamic Reporting: Allows interaction, such as filtering by year or region, updating the entire report.
  • Relation to Excel:
    • Uses Power Query (data transformation) and Power Pivot (data modeling) from Excel.
    • Advanced Excel features like Pivot Tables, Power Map, and DAX are shared.
  • Participants’ Views:
    • Excel-like for data visualization.
    • Supports multiple tables and dynamic drilling.

Components of Power BI

Power BI includes:

  • Power Query: For data cleaning and transformation.
  • Power Pivot: For data modeling and relationships.
  • Power View: For visualizations.
  • Power BI Desktop: Main tool for building reports.
  • Power BI Service: For publishing and sharing.
  • Power BI Mobile: For access on devices.
  • Q&A: Natural language queries (e.g., “Show sales by continent” generates a map).

Data Flow in Power BI

The core process is: Connect → Transform → Model → Visualize → Publish/Share.

Step-by-Step Data Flow

  1. Connect Data Sources:

    • Databases: SQL Server, SAP, Oracle.
    • Files: CSV, Excel, Text, PDF.
    • Example: Raw sales data with invoice date, chain, country, category, unit price, cost price.

  2. Transform Data (Power Query):

    • Clean and reshape: Handle upper/lower case inconsistencies, trim spaces, calculate fields (e.g., profit = sales - cost).
    • Why needed: Raw data from 20+ tables may have inconsistencies; transform for accurate reporting (e.g., proper case product names).

  3. Model Data (Power Pivot/DAX):

    • Create relationships between tables (e.g., fact table: sales; dimension tables: country mapping, date mapping).
    • Add calculated columns/measures using DAX (Data Analysis Expressions).
    • Example: Map countries to continents (Bangladesh → Asia) for continent-wise sales.

  4. Visualize and Report:

    • Create dashboards with charts (column, bar, pie, treemap), scorecards (totals, percentages), and slicers (filters).
    • Interactive: Click category to drill to products; filter by financial year.
    • Examples:
      • Continent-wise sales bar chart.
      • Quarter-wise sales/profit combo chart.
      • Category-wise treemap.

  5. Publish and Share:

    • Upload to Power BI Service for online access.
    • Requires Pro license ($10/user/month) for sharing; Free for local work.
    • Supports collaboration across locations (e.g., CFO in New York views report from Chittagong).

Visual Data Flow Diagram (Conceptual)

Connect Data → Transform (Power Query) → Model (Power Pivot/DAX) → Visualize Dashboard → Publish/Share

Pricing and Licensing

PlanCost (per user/month)Features
Free$0Desktop app, local reports, no publishing.
Pro$10Publish and share reports online.
Premium$20Advanced features for large datasets.
Premium Per UserVariableFor enterprises with bulk usage.
  • Download: Microsoft website, Microsoft Store, or Office suite (auto-updates).

Installation Guide

  1. Download from Power BI Desktop.
  2. Run installer: Next → Next → Install.
  3. Launch: Select “Blank Report” to start.
  • System Requirements: Standard Windows setup; no complex configurations.

Mini-Project Demonstration: Sales Dashboard

Used an Excel file with three tables: Sales (fact), Country Mapping (dimension: country → continent), Date Mapping (dimension: date → financial year/quarter).

Steps in Power BI Desktop

  1. Connect Data:

    • Home → Get Data → Excel Workbook.
    • Select file → Choose sheets/tables (Sales, Country Mapping, Date Mapping).
    • Transform Data → Opens Power Query Editor.

  2. Transform in Power Query:

    • Select Sales table.
    • Add Column → Custom Column:
      • Total Units: = [Quantity] (or count logic).
      • Profit: = [Sales] - [Cost].
    • Close & Apply → Loads to Power BI.

  3. Data Modeling:

    • Model View → Drag to create relationships:
      • Sales[Country] → Country Mapping[Country] (one-to-many).
      • Sales[Invoice Date] → Date Mapping[Date] (one-to-many).
    • Ensures continent and quarter calculations propagate.

  4. Build Visualizations (Report View):

    • Column Chart: Axis = Continent, Values = Sales (from Country Mapping).
    • Combo Chart: Axis = Quarter (from Date Mapping), Column = Sales, Line = Profit.
    • Treemap: Group = Continent, Values = Sales.
    • Slicer: Field = Financial Year (filters entire report).
    • Bar Chart: Axis = Category, Values = Sales (legend = Chain for breakdown).

  5. Interactivity:

    • Slicer filters update all visuals (e.g., select FY 23-24 → Quarters Q1-Q4 appear).
    • Drill-down: Click continent → Shows country breakdown.

Challenges Addressed

  • Continent-Wise Sales: No direct column; used mapping table for lookup.
  • Financial Year/Quarter: Derived from date mapping; no raw column.
  • Profit Calculation: Added via Power Query (Sales - Cost).
  • Chain Breakdown: Legend in charts for Readymade vs. Accessories.

Sample Visuals Created

  • High-level KPIs: Total Sales, Profit %.
  • Continent/Quarter trends.
  • Category treemap.

Follow up

References

Backlinks