Support for organizing Analytics Product through Domains, Projects and Workspaces

For large scale Analytics Product deployment that requires Separation of Duties and Application Lifecycle Management (ALM) support using Deployment Pipelines the team has to work with nine workspaces and three deployment pipelines. This idea seeks to implement a concept of a Product / Project to help organize large solutions in Power BI Service as well as in metadata feeds such as Scanner API.

Microsoft best practices suggest separating Dataflows, Datasets and Reports into three separate workspaces for separation of duties concerns.

When we apply this to Dev/Test/Prod ALM this leads to nine workspaces at minimum and three deployment pipelines (Dataflow, Dataset, Reports).

This idea helps organize everything into a hierarchy starting with Domain. Each Domain can have 0..N Products/Projects. Each Product that uses a Deployment Pipeline then groups the Workspaces into Dev / Test / Prod categories. This will enable development teams to quickly search and navigate through 10s or 100s of workspaces based on the Product and ALM phase.

Adding this hierarchy into the metadata allows Admins to be able to classify and categorize all the content. In Power BI service, menus can help end-users and developers to find what they need quickly. If a developer is focused on the Dev stage for a specific Project these filters can help them get a focused view of their work.