ERD (Entity Relationship Diagram) questions test your ability to design efficient database schemas, define relationships, and convert real-world business processes into structured database models.

These real-time scenario-based ERD questions are frequently asked in Infosys, Accenture, Capgemini, and Deloitte interviews — ideal for data analysts, BI developers, and DB designers.

🔹 Top ERD (Entity Relationship Diagram) Interview Questions (2025 Edition)

1️⃣ What is an ERD and why is it important?

Answer:
An Entity Relationship Diagram (ERD) is a visual representation of entities, their attributes, and the relationships between them.
It’s used to design the logical structure of a database before actual implementation.
✅ Helps ensure data integrity, reduce redundancy, and improve scalability.

2️⃣ What are the main components of an ERD?

• Entity: A real-world object (e.g., Customer, Order).

• Attribute: Property of an entity (e.g., Name, ID, Price).

• Relationship: Association between entities (e.g., Customer → places → Order).

• Primary Key (PK): Unique identifier of each entity.

• Foreign Key (FK): References PK from another table to form a relationship.

3️⃣ Explain the different types of relationships in ERD.

📌 Many-to-many relationships are usually implemented through a junction (bridge) table.

4️⃣ How do you convert an ERD into database tables?

1. Each entity becomes a table.

2. Each attribute becomes a column.

3. Primary keys and foreign keys are added to define relationships.

Example:
Entities:

• Customer (Customer_ID, Name, Email)

• Order (Order_ID, Order_Date, Customer_ID)

Relationship → One-to-Many

5️⃣ What is the difference between logical and physical data models?

✅ Accenture often tests this difference to check data modeling clarity.

6️⃣ What are normalization and denormalization?

• Normalization:
  Process of organizing data to reduce redundancy.
  Common forms:

  • 1NF: Atomic columns

  • 2NF: Remove partial dependency

  • 3NF: Remove transitive dependency

• Denormalization:
  Combining tables to improve query performance (used in reporting and data warehousing).

📌 Infosys interviewers often ask: “Why would you denormalize a data model?”

7️⃣ How do you represent optional and mandatory relationships in ERD?

• Mandatory Relationship: Shown with a solid line — e.g., Each Order must have a Customer.

• Optional Relationship: Shown with a dashed line — e.g., Each Customer may have a Discount record.

8️⃣ What is cardinality in ERD?

Cardinality defines the number of instances one entity can have in relation to another.
Example:

• A customer can place many orders → (1:N).

• An order belongs to one customer → (N:1).

✅ Cardinality determines foreign key placement in your database.

9️⃣ Real-Time Scenario: Design an ERD for an Online Learning Platform

Entities:

• Student, Course, Instructor, Enrollment
  Relationships:

• Student ↔ Enrollment (1:N)

• Course ↔ Enrollment (1:N)

• Instructor ↔ Course (1:N)

Junction Table: Enrollment(Student_ID, Course_ID, Enroll_Date)

✅ Common real-world scenario asked in Deloitte & Capgemini interviews.

💡 Pro Tip:

During interviews, when asked to “draw an ERD,” explain your approach before design:

“I start by identifying entities from the business use case, then define relationships, keys, and finally convert them into normalized database tables.”

This structured explanation shows professional thinking — and earns extra points.

🧩 Conclusion

These ERD (Entity Relationship Diagram) interview questions (2025 Edition) help you master database design fundamentals, relationships, and real-world modeling scenarios.

Companies like Infosys, Accenture, and Capgemini assess how well you can translate a business process into a clear, scalable data model — not just how you draw diagrams.

Keep practicing with sample case studies (e.g., Hospital Management, Banking System, E-commerce) to strengthen your ERD and normalization skills.

These 2025 interview questions cover the practical side of data warehouse design, ETL workflows, fact/dimension tables, and star-schema implementations commonly used in enterprise analytics projects.

🔹 Top Data Warehousing Real-Time Interview Questions (2025 Edition)

1️⃣ What is a Data Warehouse and how is it different from an OLTP system?

Answer:
A Data Warehouse (DW) is a centralized system for storing historical and analytical data, optimized for reporting.

✅ Common starting question in TCS interviews.

2️⃣ Explain the difference between Star Schema and Snowflake Schema.

• Star Schema:

  • One fact table linked to multiple dimension tables.

  • Simpler, faster queries.

• Snowflake Schema:

  • Dimensions normalized into sub-dimensions.

  • Reduces redundancy, but increases joins.
    ✅ Capgemini often asks which schema you used in your project.

3️⃣ What are Fact and Dimension Tables? Give examples.

• Fact Table: Stores quantitative data (e.g., sales amount, quantity).

• Dimension Table: Stores descriptive attributes (e.g., customer, product, region).
  Example:

• Fact_Sales: Date_Key, Product_Key, Sales_Amount

• Dim_Product: Product_Key, Category, Brand

4️⃣ What are Slowly Changing Dimensions (SCD)?

SCDs handle historical changes in dimension data.
Types:

• Type 1: Overwrite old data (no history)

• Type 2: Add new record (keeps full history)

• Type 3: Add new column (partial history)

Example (Type 2):

✅ Real project question from Infosys DW team.

5️⃣ What is a Data Mart and how is it related to a Data Warehouse?

A Data Mart is a subset of a Data Warehouse — focused on a specific business function.
Example:

• Finance Data Mart

• Sales Data Mart
  ✅ TCS often checks if you understand data segregation for analytics.

6️⃣ How do you load data incrementally in a Data Warehouse?

Incremental load = only new or changed data since last refresh.
Techniques:

• Using Last Updated Date columns

• Change Data Capture (CDC)

• Comparing hash keys
  Example:

7️⃣ What is ETL and what tools are commonly used?

ETL: Extract → Transform → Load.
Popular tools:

• Informatica PowerCenter

• Talend

• SSIS

• Oracle Data Integrator (ODI)

• Azure Data Factory / AWS Glue

✅ Capgemini focuses on data orchestration and automation concepts.

8️⃣ What is Data Staging and why is it used?

The staging area temporarily stores raw extracted data before transformation.
Advantages:

• Reduces load on source systems.

• Supports incremental data validation.

• Simplifies error correction during ETL.

9️⃣ What are Aggregate Tables and why are they used?

Aggregate tables store pre-calculated summaries (e.g., monthly sales totals).
They:

• Improve query performance.

• Reduce computation on live queries.
  Example:

✅ Infosys uses this example in Power BI + Oracle DW interviews.

🔟 How do you ensure data quality in ETL processes?

• Validate source vs. target counts.

• Check for duplicates and nulls.

• Maintain audit logs.

• Use error handling tables for rejected records.

📌 Practical question in real-time DW project interviews.

💡 Pro Tip:

When explaining a project scenario, always describe the data flow:

“We extracted sales data from SAP, transformed it in Informatica, and loaded it into an Oracle Data Warehouse for Power BI dashboards.”

It adds credibility and aligns with real enterprise environments.

🧩 Conclusion

These Data Warehousing real-time interview questions (2025 Edition) reflect what top companies like Infosys, TCS, and Capgemini look for in ETL and BI professionals.

Focus on understanding data modeling concepts, SCD types, and incremental loading techniques — they form the foundation of every data warehouse project.

Consistent practice with ETL tools and SQL-based transformations will help you confidently answer any scenario-based question.

These 2025 interview questions are collected from real candidate experiences and cover both conceptual and scenario-based topics.

🔹 Top ETL & Data Warehousing Interview Questions (2025 Edition)

1️⃣ What is ETL and why is it important?

Answer:
ETL stands for Extract, Transform, Load — the process of moving and preparing data for analysis.

• Extract: Pull data from multiple sources (databases, APIs, files).

• Transform: Clean, validate, and structure it.

• Load: Store in a data warehouse or reporting system.
  ✅ Common question in TCS and Cognizant technical rounds.

2️⃣ What are the key differences between OLTP and OLAP systems?

3️⃣ Explain the Star Schema and Snowflake Schema.

• Star Schema: Fact table linked to dimension tables — simple and fast for reporting.

• Snowflake Schema: Dimensions are normalized into sub-dimensions — reduces redundancy but increases joins.

✅ Infosys often asks which schema you used in your last project and why.

4️⃣ What are fact tables and dimension tables?

• Fact Table: Stores measurable data (e.g., sales, revenue).

• Dimension Table: Stores descriptive attributes (e.g., product, region, customer).
  Example:

• Fact_Sales: sales_amount, date_key, product_key

• Dim_Product: product_key, category, brand

5️⃣ What are slowly changing dimensions (SCD)?

Used to manage changes in dimension data over time.
Types:

• Type 1: Overwrite old data.

• Type 2: Keep history with new record + effective dates.

• Type 3: Keep limited history using extra columns.

Example in SQL:

6️⃣ What tools are commonly used for ETL processes?

• Informatica PowerCenter

• Talend

• SSIS (SQL Server Integration Services)

• Apache NiFi

• Azure Data Factory / AWS Glue

✅ Cognizant prefers scenario-based questions on Azure Data Factory pipelines.

7️⃣ How do you handle data quality issues in ETL?

• Implement data validation rules in transformation stage.

• Use lookup checks for referential integrity.

• Apply deduplication logic using keys.

• Maintain error tables to log invalid records.

8️⃣ What is data staging in ETL?

The staging area is a temporary storage layer used to hold raw data before transformation.
It helps:

• Reduce load on source systems

• Manage incremental loads

• Audit data lineage

9️⃣ How do you perform incremental data loads?

By loading only new or changed records since the last update.
Common methods:

• Using timestamps or last modified dates

• Maintaining control tables to store last run times

• Applying CDC (Change Data Capture)

Example:

🔟 Explain ETL performance optimization techniques.

• Push transformations to source database.

• Use bulk load instead of row-by-row inserts.

• Index join columns and partition large tables.

• Parallelize ETL jobs using threads or data partitions.

✅ Infosys interviewers love real-time performance tuning examples.

1️⃣1️⃣ What is a data mart and how is it different from a data warehouse?

• Data Warehouse: Enterprise-level, multiple subject areas.

• Data Mart: Department-level, focused on a single business process.
  Example:

A “Finance Data Mart” might be created from the larger enterprise warehouse.

1️⃣2️⃣ What are surrogate keys in data warehouses?

Surrogate keys are system-generated IDs (e.g., identity columns) used instead of business keys.
They simplify joins and improve query performance.

💡 Pro Tip:

When answering ETL or warehousing questions, include a mini project example like:

“In my Cognizant project, we used Informatica to extract data from Oracle and Salesforce, applied business rules in staging, and loaded it into a Snowflake warehouse for Power BI dashboards.”

It adds credibility and shows real-world exposure.

🧩 Conclusion

These ETL and Data Warehousing interview questions (2025 edition) are frequently asked at TCS, Infosys, and Cognizant.
Focus on explaining data flow, transformation logic, and performance strategies — recruiters value clarity and practical experience over memorized definitions.

Consistent hands-on practice with ETL tools and SQL will set you apart in 2025 interviews.

Here’s a sample methodology to approach business requirements for BI dashboard development:

1. Stakeholder Meetings: Begin by meeting with key stakeholders to gather detailed information on their goals, KPIs, and expectations. This is the foundation for understanding the "why" behind the data.

2. Requirement Documentation: Use a requirement-gathering template to document essential details, such as data sources, reporting frequency, and the specific metrics or KPIs to focus on. This documentation will act as a blueprint throughout the project.

3. Iterative Prototyping: Develop an initial dashboard prototype and review it with stakeholders to gather feedback. This iterative approach ensures that the end product aligns closely with the users’ needs.

4. Alignment with Business Goals: Continually align the dashboard features and visuals with business goals. For instance, if the focus is on sales performance, the dashboard should highlight KPIs related to revenue, customer acquisition, and retention rates.

For a more in-depth look at BI interview questions and answers, check out our comprehensive SQL & BI Interview Questions and Answers guide.

Conclusion:
Following these steps not only helps create impactful dashboards but also enhances stakeholder trust and engagement. With a structured approach, BI professionals can ensure they deliver insights that truly support business decision-making.

What is the normalization in DBMS and what is the purpose of it? 

can you explain how many types of fact tables available in data warehousing? 

What is the concept of slowly changing dimensions in the data warehousing? 

Can you please answer this question what is the major difference between star schema and snowflake schema in data warehousing?