Data modeling, a concept that has been around for a long time, is gaining more traction with the rise of analytics engineering. The popular data transformation tool, dbt, has redefined what a data model is by allowing users to write modular code and test their models within the documentation itself. This article discusses how to organize source data, define sources, reference sources, document sources, create staging models, follow best practices for writing data models, and properly test them using dbt. By following these guidelines, analytics engineers can ensure that their data models are efficient, maintainable, and easily understood by others on the team.