POWER BI DATA MODELLING – INTERVIEW Q&A;
BEGINNER LEVEL
1. What is a Data Model in Power BI?
Answer: A data model is the structure of tables and relationships that organizes data for reporting.
Example: Sales and Customers tables linked by CustomerID.
2. What are Dimensions and Facts?
Answer: Facts contain measurable values (e.g., SalesAmount), while dimensions describe context
(e.g., Date, Product). Example: A Sales table is fact; Date table is dimension.
3. What are Relationships in Power BI?
Answer: Relationships define how two tables connect using keys. Example: Customer[CustomerID] →
Sales[CustomerID].
4. What is a Star Schema?
Answer: A star schema uses a central fact table connected to dimension tables. Example: FactSales
linked to DimDate, DimProduct, DimCustomer.
5. What is a Primary Key?
Answer: A unique column identifying each record. Example: CustomerID in Customer table.
INTERMEDIATE LEVEL
1. What is Cardinality in Relationships?
Answer: Cardinality defines relationship types—1:1, 1:* or *:*. Example: One customer → many sales.
2. What is Cross-filter Direction?
Answer: It defines how filters flow between tables: single or both directions. Example: Single-direction
from dimension to fact is recommended.
3. What is a Snowflake Schema?
Answer: A schema where dimensions are normalized into sub-dimensions. Example: DimCity linked to
DimRegion linked to FactSales.
4. Why is a Date Table important?
Answer: It enables time-intelligence functions (YTD, MTD). Example:
CALCULATE(SUM(Sales[Amount]), DATESYTD('Date'[Date])).
5. What is a Surrogate Key?
Answer: A synthetic key created to uniquely identify rows. Example: Auto-incremented CustomerKey.
ADVANCED LEVEL
1. What is a Many-to-Many Relationship?
Answer: A relationship where both tables contain non-unique values. Example: Multiple salespeople
assigned to multiple regions.
2. What are Role-Playing Dimensions?
Answer: One dimension serving multiple roles. Example: Date table used as Order Date, Ship Date,
Invoice Date.
3. What is a Composite Model?
Answer: A model combining DirectQuery and Import tables.
4. What is Data Normalization and Denormalization?
Answer: Normalization removes redundancy; denormalization improves performance by flattening.
Power BI prefers denormalized star schema.
5. Explain Relationship Inactivity.
Answer: Power BI allows only one active relationship; inactive ones can be activated using
USERELATIONSHIP().
SCENARIO-BASED QUESTIONS
1. Scenario: Sales report shows wrong totals. What will you check?
Answer:
- Relationship direction
- Cardinality
- Duplicates in key columns
- Incorrect model schema (snowflake instead of star)
2. Scenario: You need YTD calculations but the model has no date table.
Solution:
- Create a date table using DAX CALENDAR()
- Mark it as a Date table
3. Scenario: Performance is slow.
Fixes:
- Remove bi-directional filters
- Use star schema
- Reduce column cardinality
- Disable auto-date/time
4. Scenario: Two tables have many-to-many issues causing duplicate values.
Solution:
- Create a bridge table with unique keys.
5. Scenario: Need to analyze employee joining & resignation dates.
Solution:
- Use role■playing Date table with USERELATIONSHIP().
END OF DOCUMENT