How do I route errors to an appropriate response
We are in the process of testing the digital assistant and for the most part it is OK. When we get errors though the BOT seems to route the utterance to the expenses skill every time rather than just advising that it doesn't know the answer.
When a question is asked, and the BOT isn't able to route to a valid skill and intent it invariably routes to the expenses skill and provides an invalid response. It seems to do this because the expenses skill achieves the highest score even though the score is nowhere near enough for the skill to be the valid response. As an example, if we enter "org chart" the BOT routes to the expenses skill when it should be unable to resolve that utterance. When I look at the flow expenses has scored 40+ percent and the other skills less than