Analytics Engineering with SQL and dbt

Chapter 4. Data Transformation with dbt

The primary purpose of dbt is to help you transform the data of your data platforms in an easy and integrated way by simply writing SQL statements. When we place dbt in an ELT workflow, it matches the activities during the transformation stage, providing you with additional components—such as version control, documentation, tests, or automated deployment—that simplify the overall work of a data specialist. Does this remind you of the actual activities of an analytics engineer? Well, that’s because dbt is one of the modern tools that defines what analytics engineers do, giving them the instruments integrated with the platform, which reduces the need to set up additional services to answer specific problems and decreases the overall system complexity.

dbt supports the tasks described for an analytics engineer, empowering them to run the code in their data platform collaboratively for a single source of truth for metrics and business definitions. It promotes central and modular analytics code, leveraging DRY code with Jinja templating language, macros, or packages. In parallel, dbt also provides the security that we typically find in software engineering best practices, such as collaborate on data models, version them, and test and document your queries before safely deploying them to production, with monitoring and visibility.

We’ve provided a thorough introduction to dbt. However, within this chapter, we will delve even deeper into the specifics ...