I have a call where the conversation progressed from one state to a next state, then returned to the previous state even though there is no path for that to happen. I would like to understand what I’ve done to allow this to happen.
Team looked into exactly what happened at the 2:05 mark.
The flow going from not_local → not_local_assistance → back to not_local is actually expected platform behavior, not a misconfiguration on your end. Here’s what happened: when the conversation transitioned from not_local to not_local_assistance, the caller (Stacy) was still actively speaking — she was in the middle of saying “Because we are using them.” The platform detected that the user was talking right at the moment of the transition, so it automatically reverted back to not_local to ensure her speech was processed in the correct context (the node she was actually responding to). Once she finished speaking, the flow moved forward to not_local_assistance again normally.
This is a built-in feature designed to handle cases where a caller’s speech overlaps with a node transition — it prevents the system from “skipping ahead” while the user is still providing input for the current node. The threshold is about 1 second: if the user speaks within 1 second of a transition, the system will revert.
Thanks for the explanation! My only comment to this would be that what the user said clearly triggered a transition. Going back caused an issue instead of solving one. Do you have any recommendations on what I can do to prevent this in the future?
It is a mechanism that reverts the transition if the user is still speaking. If a transition occurs but the user speaks again within 1 second, the system reverts to ensure the full context is captured, then transitions to the correct node with complete context.