Granular Billing for Azure Data Factory

This is super cool, but only if this included Databricks charges for notebook jobs started from the pipeline (using Job Clusters). Very often, ADF pipelines are actually leveraging Azure Databricks for ETL heavy-lifting - and this is where the bulk of the costs comes from.

Also, like others said, very often pipelines are actually generic and parametrized, one pipeline serves different table. It would be best if some kind of tagging could be inferred from parameters. Not neccesarily all parameters every time. More like create a pipeline-level setting allowing a developer to dynamically assign its billing tags based on pipeline parameter (that the developer chooses to leverage for that purpose)