Data Modelling for Business Analysts

Instructor: Alaa Hady · Language: English or Arabic · Suggested duration: 1–2 days · Level: Intermediate (Level 200)

Objectives

• Design intuitive, high-performance semantic models tailored for business reporting.
• Apply Kimball dimensional modelling principles (Star Schema) in Power BI.
• Configure date tables, hierarchies, and role-playing dimensions accurately.
• Eliminate common modelling traps (flat tables, bi-directional chaos, snowflaking) that break DAX.

Outline and contents

1. Translating business questions into dimensional model requirements (defining the grain)
2. Fact tables (transactional, periodic snapshot, accumulating snapshot) vs. Dimension tables
3. Star Schema vs. Snowflake vs. Single Flat Table: why Star Schema wins in VertiPaq
4. Relationships: 1:, 1:1, and *: cardinality, surrogate keys, and single vs. bi-directional cross-filtering
5. Designing a dedicated Calendar/Date table and handling multiple dates (Order Date vs. Ship Date via USERELATIONSHIP)
6. Resolving Many-to-Many (:) scenarios safely with bridge tables and shared dimensions
7. Budget vs. Actuals modelling: handling data at different granularities
8. Model usability: display folders, hierarchies, synonyms for Copilot/Q&A, data categories, and hiding technical keys
9. Validating model integrity and reconciling totals against source systems
10. Hands-on lab: Refactoring a slow, broken flat-file model into a clean Star Schema

Target audience

Power BI developers, business analysts, report authors, and data professionals building semantic models.

Expected outcomes

Participants can architect and validate a scalable Star Schema model, handle multi-fact and multi-granularity requirements cleanly, and keep DAX measures simple and accurate.