Data Engineering & Warehousing·Intermediate
Data Modeling Best Practices for Data Warehousing
Most warehouse problems trace back to modeling decisions made early and never revisited. This course gives your team a shared, practical modeling discipline: how to choose the right grain, design conformed dimensions and fact tables, handle change over time, and avoid the anti-patterns that quietly erode performance and trust. It is platform-aware (Snowflake, Microsoft Fabric, Databricks, BigQuery) and closes with how modern teams express these models in code. Every module pairs concepts with worked examples on realistic data.
Duration
2 days
Level
Intermediate
Group size
1-30 people
Formats
On-site · Virtual · AICG facility
Agenda
A representative outline. Every session is tailored to your team's warehouse and workflows.
- 09:00Warehouse foundations: OLTP vs OLAP, architectures, and layered design
- 10:30Requirements, declaring grain, and source / data-quality analysis
- 13:00Dimensional fundamentals: star vs snowflake, conformed dimensions, the bus matrix
- 15:30Lab: design dimensions with slowly changing dimension (SCD) types on sample data
- 09:00Fact table design: transaction, snapshot, and accumulating facts plus measures
- 10:30Normalization trade-offs and the anti-patterns that hurt warehouses
- 13:00Quality, testing, governance, and performance on cloud platforms
- 15:30Capstone lab: model a star schema end to end and a 30-day adoption plan
Partial agenda shown. The full module list is shared when you inquire, then adjusted with your team lead.
Requirements
- Comfortable writing SQL (joins, aggregates, and CTEs)
- Basic familiarity with a relational or cloud data warehouse
- No prior formal data-modeling background required
What your team gets
- Full course workbook and reference materials
- Worked example datasets and model solutions
- Certificate of completion for each attendee
- 30 days of follow-up questions with your instructor
What you'll be able to do
- Choose the correct grain and translate business questions into a model
- Design star schemas with conformed dimensions and a clear bus matrix
- Implement slowly changing dimensions and the right fact-table type
- Recognize and refactor common modeling anti-patterns
- Apply modeling choices to cost and performance on cloud warehouses
Who it's for
- Data engineers and analytics engineers building or maintaining a warehouse
- BI developers and data analysts who design or consume warehouse tables
- Data architects standardizing modeling practice across teams
- Technical leads responsible for warehouse performance and cost