Checking out conversations with red labels
Two cases where human attention is needed the most: when the conversation goes triggers the fallback mechanism, and when the bot detects a negative user sentiment.
The fallback mechanism activates when the chatbot is not confident enough while identifying the intent and the corresponding response — the chatbot then alerts human officers to answer the queries on its behalf. The chatbot also uses artificial intelligence to run sentiment analysis. It essentially tells the general sentiment of the customer, through his/her use of words — we are not taking the risk of letting a chatbot handle conversations with angry customers.
Among all conversations that your chatbot has been handling, our chat agent dashboard is designed to get humans to handle the mentioned two cases first. The dashboard puts a red label on the customer name for the attention.
Taking over your chatbot’s conversation, and passing it back
The rule is, the conversations with a red label should be handled first. We are not leaving the customers without answers, or leaving angry customers with no service provided to them.
Click on the “Takeover” button on the upper right-hand side, and the chatbot operation will be paused for that particular conversation — the conversation is all yours. The chatbot will not respond to any queries that the customer sends. You can type in your answers and continue conversing with the customer on the bot’s behalf. You may also choose to let the customers know that they are talking to a human right now, not a chatbot.
While you handle the conversation, the chatbot uses machine learning to learn how you respond to that particular intent. It is now ready to respond to a similar intent in a future encounter.
After you are done, click “Resume” to let the bot continue its operations for that conversation.