In an advanced AI agent, system replies don’t answer customer queries, but instead deliver specific types of messages at different points in a conversation. For example, there’s a system reply for the AI agent’s first message to a customer in a conversation, a system reply for collecting feedback at the end of a conversation, and more.
About system replies
System replies are included within the AI agent by default. Some have prewritten text that you can edit as needed. Include them in your conversation flow to provide a better experience for common scenarios.
- Collect BSAT reply: Collects feedback from your customers to measure how your AI agent performs. If configured to be sent at the end of each conversation, it asks the customer to rate their experience.
- Default reply: Used when the AI agent doesn’t understand the customer’s query and cannot produce a generative reply. For expression-based agents, it’s used when the confidence threshold for intent recognition is not met.
- Escalation reply: Used to configure escalations to a human agent.
- Failed escalation reply: Informs the customer there are no human agents available for escalating their request.
- Technical error reply: Tells customers there is a technical issue. Sent when there is an error, either from the CRM or the advanced AI agent. For example, when an unexpected error occurs when connecting to an external API.
- Generative replies reply: When generative replies are activated and no matching use case or intent is recognized, the AI agent replies with an AI-generated message. Depending on when your AI agent was created, this system reply might be called “uGPT reply.”
- Unsupported language reply: Used when the user’s language is not recognized. For more information, see Configuring language detection for advanced AI agents.
- Welcome reply: Greets customers when they reach out and sets the tone for the conversation. See Adding and removing the welcome reply for advanced AI agents.
Viewing all system replies
You can view all system replies on the Use cases page.
- In AI agents - Advanced, in the top-right corner, use the AI agent drop-down field to select the AI agent you want to view use cases for.
- In the left sidebar, select Content > Use cases.
- Click the Categories filter, select System replies, and click Apply.
Editing system replies
Even though some system replies are preconfigured for you, you can always edit a system reply to better suit your needs.
- In the list of system replies, click the system reply you want to edit.
- Select the reply for the language you want to edit.
- Click Edit dialogue.
The dialogue builder opens.
- Update the system reply as needed.
For help, see Using the dialogue builder to create conversation flows for advanced AI agents.