When using a stored procedure, it cannot automatically detect the output columns the same way as it can with a table or view, so you have you enter them manually and they must match.

For example, if your SP returns a row from a table with columns A, B, and C

Then in the "with result set column" you would enter A, B, or C depending on which value you want.

That means you need to be aware of the exact names of the outputs so you can match them up correctly in order to get the results you want.

Additionally, unlike a Table or View, Stored Procedure require more access rights beyond "read" for execution in MS SQL Server, so it could also depend on what account you used for the data source.

For example, an account that has read-only access to the database won't be able to execute a stored procedure unless you provide additional rights to do so.

The reason MS SQL Server does this is that a Stored Procedure can contain/obscure more complex activities like updates or deletes, that shouldn't be available to read-only users.