I don’t know the average but I’m sure it depends on what’s being supported.
In our case, a portion of our support questions can’t possibly be handled by the chatbot because they require an action to be taken. For example, if an account change needs to be made or equipment needs to be returned (which requires a refund and/or an authorization) then the bot can’t do it.
It also depends on the complexity of the technical support. Some questions are straight forward. If the user had just searched the forum with the right terms they would have found the answer. But some users would rather just ask the question and get an answer. In that case, the bot is pretty good at a searching the forum, finding the answer, and presenting it to them. But when the problem requires investigation or multi-variable analysis, the bot tends to give a generic answer that doesn’t add much value.
We expected this though. Building out the knowledge base and improving the prompt is ongoing work. And we anticipate the LLM and the plugin to continually improve as well.