How to Set Up Deployment Pipeline with Different Data Sources for Dev and Prod

Quick Answer

Configuring parameter rules in the deployment pipeline so that each stage maps to its own value is the answer because this is the mechanism that actually lets Development and Production point to different data sources without touching the report file itself as it moves between stages. That capability starts upstream, though: it depends on the report first being built with dataset parameters defined in Power BI Desktop for the data source connection details, rather than those details being hard-coded into the queries. Once the connection is parameterized, the deployment pipeline can then override that parameter's value independently at each stage, so the same underlying file can pull from a development database while in the Development stage and automatically switch to the production database once deployed to Production, with no manual editing required at each step. Skipping either half of this breaks the setup: without dataset parameters, the pipeline has nothing to override, and without configuring pipeline-level parameter rules, a parameterized report still uses the same fixed value at every stage. The broader takeaway for any deployment pipeline scenario is to look for this two-part structure -- first make the varying piece a genuine parameter at the report or dataset level, then configure the environment-specific values for that parameter at the pipeline level -- rather than expecting either step alone to be sufficient.