6 Comments
User's avatar
Julio Lara's avatar

Thanks! Really great post. One question that came to mind though, how do the dbt modeling layers (staging → intermediate → marts) compare to the medallion architecture layers (bronze → silver → gold)? Are they just different naming conventions for similar concepts, or do they serve distinct purposes in practice?

Expand full comment
Pipeline to Insights's avatar

Hi Julio, thank you so much for your comment and feedback. Great question .I believe that while the Medallion architecture and dbt modelling layers share some conceptual similarities, they actually serve distinct purposes and operate at different levels of the data stack.

Medallion focuses on the data platform and storage layers, while dbt modelling layers are all about SQL transformations within dbt. They’re not mutually exclusive. They complement each other well. You can use Medallion to structure your lakehouse and dbt to organise and transform your data within those layers. That's what I think but happy to hear your thoughts :)

Expand full comment
Giuseppe Santoro 🚢's avatar

That's a great article. I've tried to learn about DBT for a s while now.

I find the official docs too low level and other articles expect you to know the basics already.

Expand full comment
Pipeline to Insights's avatar

Thank you so much for your feedback 💐🙏😊

Expand full comment
Donald Parish's avatar

The "marts" here look more like reports to me than Fact or Dimension tables?

Expand full comment
Pipeline to Insights's avatar

Hi Donald, you are definitely right! We wanted to keep the example as simple as possible while showing dbt's functionalities with some direct outputs for this first post.

For the later posts we are preparing example's with real data models.

Expand full comment