Data Warehouse and Business Intelligence
Dashboards are only as trustworthy as the data underneath them. This program teaches you to design that foundation properly: dimensional models, slowly changing history, reliable ETL and ELT pipelines and a warehouse…
Course description
Dashboards are only as trustworthy as the data underneath them. This program teaches you to design that foundation properly: dimensional models, slowly changing history, reliable ETL and ELT pipelines and a warehouse that stays correct as the business changes.
You work on one end-to-end project, from raw source tables to a tested model that reports can rely on, and learn how to explain your design choices to both engineers and business teams.
Class schedule
- Days & time
- Saturday & Sunday
- Duration
- 12 weeks
What You'll Learn
Skills You'll Gain
Test Yourself
Profile operational sources
Explore the retail chain's POS, ERP and product API exports in PostgreSQL, write down grain, keys, volumes and quality problems, and propose which questions the warehouse must answer. You'll find the product API disagrees with the ERP on category. That's deliberate.
Hands-on Lab Task
Design and load a star schema
Model FactSales, DimDate, DimProduct, DimStore and DimCustomer at the right grain, generate surrogate keys and load them from staging with SQL.
Hands-on Lab Task
Implement a Type 2 dimension
Add valid_from, valid_to and is_current to DimCustomer, write the merge that closes old rows and inserts new versions, and prove with a query that a customer who moved from Lahore to Karachi still has last year's sales in Lahore.
Hands-on Lab Task
Build an incremental pipeline with batch control
Create a batch control table, load only new or changed rows each run, make the load safe to rerun and log row counts and status for every step.
Hands-on Lab Task
Add dbt-style models and tests
Split transformations into staging, intermediate and mart models, add not-null, unique and relationship tests, and generate lineage docs.
Hands-on Lab Task
Connect Power BI and tune it
Build a Power BI dataset over the marts, compare Import and DirectQuery, add an aggregation table and use query profiles to cut a slow report's time.
Hands-on Lab Task
Curriculum
Why would a company copy data out of the systems that run the business into a separate database that runs nothing? New students ask this every batch, and it's the right question. This weekend gives you the vocabulary and the big picture, OLTP versus OLAP, the layers of a platform, and who does what on a data team.
What we cover
- Operational databases versus analytical warehouses, purpose, workload, shape of the data
- Warehouse architectures, staging, integration and presentation layers, data marts, lakehouses
- Batch versus streaming, ETL versus ELT, and the modern data stack in one diagram
- Roles, data engineer, analytics engineer, BI developer, analyst, and what each is paid for in Pakistan and abroad
- The course case study, a retail chain with POS, ERP and a product API
In the lab
Explore the case study's source exports in PostgreSQL, list the business questions the warehouse must answer and sketch a first architecture.
Bring back next week
A one-page architecture proposal with a source inventory, the target questions and a layered diagram.
Batches
Pick a batch that fits your schedule, then use the enquiry form to reserve your seat.
No open batches right now, submit your details and we'll notify you when the next one starts.
An instructor will be assigned when the next batch is scheduled.
Student Reviews
No reviews yet, be the first to enroll and share your experience.
Requirements
- Working SQL: SELECT, JOIN and GROUP BY. If you can write a query that joins two tables, you are ready.
Fee by mode