Application databases are typically optimised for transactional operations, not analytical queries. Directly running complex analytical queries on the application database can slow down the application's performance. By staging the data in a separate data warehouse, you can offload the analytical processing from the application database, ensuring that the application's performance is not affected.

Application databases often store data in a format that is optimised for the application's needs, not for analytics. Tools like dbt can be used to transform and structure the data in a way that is more suitable for analytics. For example, dbt can be used to create a semantic layer that simplifies complex data structures into a more accessible format for Cube to interpret and visualise

If you have data from multiple sources that you want to analyse together, it can be beneficial to stage all this data in a single data warehouse. Tools like dbt can be used to integrate this data, ensuring that it is consistent and can be analysed together

dbt provides a framework to test assumptions about the results generated by a model, which helps provide assurance that your SQL is transforming data in the way you expect, and your source data contains the values you expect. This can be crucial for ensuring the accuracy and reliability of your analytics

As your data volume grows, it can become increasingly difficult to manage and analyse the data directly in the application database. A separate data warehouse can provide more scalability, allowing you to handle larger data volumes more effectively