This is a bit of a pie-in-the-sky idea, but I think it would be hugely helpful.
Have you ever had a rule that you needed to edit, and found that the only way to get the desired outcome is to add a new task to the logic tree? Has that rule ever had a LOT of actions? In the current system, adding tasks inevitably results in a new “end of the line” output. Since Actions are tied to specific output tasks, adding a task means redoing the Actions associated with the old output entirely.
What if there were a better way? What if, for instance, we had the ability to create a new type of Task - “insert” - which could be used to insert new field output directly into an existing task? That would be awesome! So long, having to rewrite a bunch of actions, hello early lunch break!
I’m aware that there are a TON of fiddly details that would need to get hashed out for this to work properly. You can’t go around just cramming data in to incompatible tasks, how do we know which records get which output values in the inserted data, etc etc. I’m sure smarter people than myself can talk through those limitations and maybe, possibly, find a way to solve for them. There’s also a decent chance that my idea here is fundamentally flawed and I just have to suck it up and rewrite these actions whenever changes come up.
Would be an immense time saver! You might be doing this but@chelbusch had advised me to always add a transformation task as the last task in a data set for this very reason. I know it doesn’t address the use case of bringing an additional data set in, but it can help avoid having to rebuild actions when you just need to generate data with a case statement or another kind of transformation.
+1000%