Intermediate dbt
Write dbt tests that catch bad data, track history with snapshots, add sources and seeds, and automate project runs with dbt build.
Inconsistent data in dbt models, sources and seeds needs catching before it reaches downstream analysis, and changes in warehouse data need tracking over time. You validate with built-in, singular and reusable tests, record historical changes as slowly changing dimensions with snapshots, and run models, tests and snapshots together with dbt build.
What You Will Achieve
- Define built-in tests on a dbt model and use the failures to locate bad data
- Turn a one-off validation query into a custom test that several models can share
- Declare raw data as dbt sources for lineage and manage small, static datasets as seeds
- Add a snapshot that keeps SCD2 history and run the whole project with a single dbt build command
Program Overview
Testing Data Quality in dbt
Builds a validation workflow that catches inconsistent data in models, sources and seeds before it reaches downstream analysis.
- Built-in tests and where to define them
- Singular tests, such as a trip duration check
- Custom reusable tests written with Jinja
- Troubleshooting test failures and verifying test queries
Implementing dbt in Production
Covers the features a production project relies on to trace raw data, hold static reference data, keep history and automate runs.
- dbt sources and data lineage
- dbt seeds for small, static datasets
- SCD2 with dbt snapshots and the snapshot process
- Automating runs with dbt build and its limits
Certificate Awarded (Paid)
Job Outlook
Testing, snapshots and automated builds in dbt are part of data engineering and analytics engineering work, where warehouse transformations feed reports and analysis that depend on accurate, traceable data.
Good for: dbt users who already build models and want tests, sources, seeds and snapshots in place before their projects run in production.
Keep in mind: The title says intermediate while the course is rated Advanced, and the first chapter names documentation, though its lessons are all about testing; documentation comes up only through source lineage.
Bottom line: A focused course that pairs dbt testing with snapshots, seeds and dbt build, the pieces a project needs once real data depends on it.
Frequently Asked Questions
How does a singular test differ from a reusable test in dbt?
A singular test is a query written for one specific check, such as verifying trip durations in a model. A reusable test is a custom test defined with Jinja that can be applied across several models. One lesson takes an existing singular test and updates it into a reusable one.
What part do snapshots play in the course?
Snapshots are how the course handles slowly changing dimensions (SCD2), keeping a record of how rows in the warehouse change over time. The lessons go through the snapshot process, look at a snapshot issue, and then add a snapshot to a project.
What kind of exercises do the lessons include?
Practice is built around tasks inside a dbt project: defining tests on a model, finding bad data, adding a source and a snapshot, and running the project with dbt build. A closing exercise brings the testing and production steps together.


There are no reviews yet.