Refresh Single table in a Dataflow

There are some ways in which you can achieve this. For example, you can disable staging to any queries that do not need to run or you can split your Dataflow into other Dataflows (we recommend this approach) so you can take a more modular approach.

By design, you may have Dataflows that would need to inspect multiple queries in order to compute the final result. For example, a query that needs to first perform a specific transformation, and if it yields a specific result then it will continue in a separate fork to create the result of your Dataflow. Computing dependencies is often required and they can only truly be calculated at refresh time, thus selecting a separate scope that could conflict with the one by-design of the Dataflow could yield unexpected results.