Autonomous Digital Assistant to Human Agent seamless Handoff

In earlier articles, I discussed Autonomous Digital Assistantprovisioning a Digital Assistant, building skills and making it multi-lingual. In this post, I would like to take the discussion forward to address certain scenarios where there is a need for Human Intervention when the Bot cannot handle the conversation and instead redirect the chat to a human agent.

Chatbots can augment the call centre interactions by taking away the load from the agents in answering the most frequently asked questions and automating transactions, however, there are always complex queries where the Bot cannot understand, and the customer would like to speak to an agent. Here is where the killer combination of Bot and Humans will drive the customer engagement more efficiently. You can make the chatbot available where customers are always engaged, i.e. messaging platforms like Facebook, Whatsapp, Sack, in addition, you can also allow your customer chat to a person as and when required taking away the frustration when their problem doesn’t get solved when interacting with the chatbot.

Let’s go through the steps involved to configure the Bot designed in Oracle Digital Assitant seamlessly handoff to a live Agent in Oracle Service Cloud

Step 1: Configure the Agent Integration with your chatbot.

Out of the Box, ADA (Autonomous Digital Assistant) supports Oracle Service Cloud as the live agent.

Provide the Domain Name, Hostname, Site Name and the Admin user set up in service cloud for the agent interaction. Enable the switch for Interaction.

The general pattern for Oracle service cloud Agent Web UI URL:

Here is how the configuration should appear like :

Sample Configuration Screen with Service Cloud Details.

Step 2: Detect the transfer action in the dialogue to redirect the chat to a human agent when the bot cannot handle or when the customer asks for it specifically.

You can easily transfer visitors to your customer support team using the AgentInitiation system component. Lead your clients to your human experts at any point in the dialogue flow and multiple times. Pass the Entities and the conversation flow to the transfer action so that the live agents can have a complete view of the chat flow and respond accordingly.

Create an Intent to detect the user input for speaking to an Agent and edit your BotML to capture the intent in the dialogue flow.

We could also map the unresolvedIntent to initiate the Agent handoff.

Add the Agen Initiation and Agent Conversation System Components to the dialogue flow.

Added a System.output message so that the customer is aware that the Bot is about to connect to a live agent.

Make sure you provide the agentChannel information based on the Step 1. (name of the Integration)

Observe that in the AgentConversation template there is an option to pass the Entity flows. This enables the Agent to respond just like the Bot if the customer is asking for information that Bot could have responded, however the bot couldn’t detect the intent appropriately.

That’s is all is required for the Agent Hand off.

Tip : Make sure Bot Conversation Switch is enabled for the bot chat history to be passed on to the live agent.

Here is the video of the Chatbot to live agent interaction in action :

Thank you for taking the time to read this article.



6 thoughts on “Autonomous Digital Assistant to Human Agent seamless Handoff”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: