Docs AI search
Add an AI support agent to your knowledge base to help customers find answers faster.
For customers, finding the information they need within documentation to do your job can be a time-consuming task.
Overview
Even with an extensive knowledge base, customers often struggle to locate the information they require. Information may be scattered or not presented in an easily understandable format. As a result, they turn to customer support and must wait for assistance.
This delays progress and disrupts focus. Customers experience hold-ups until they receive answers, while customer support teams are overwhelmed with repetitive inquiries, diverting their attention from more critical tasks.
Enter the Outverse support agent, a tool designed to serve as a personal assistant for each customer, significantly reducing the time spent searching for essential information. With its ability to sift through thousands of documents in seconds, our support agent delivers the required information swiftly, enabling customers to self-serve support, faster.
To provide our search and chat services, we use foundational large language models (LLMs) and artificial intelligence (AI) services from OpenAI.
Our providers will never use your data to train AI or LLM models. See the link above for details of each OpenAI's policies.
Your AI Agent will use your Outverse documents as context in order to answer customer questions. The AI agent can find relevant information from documents in your knowledge base, provide summaries of key points from multiple pages, and even tackle more detailed questions about the content.
The agent is only able to reference published documents – so any unpublished documents will not be used as context for the agent.
Asking questions
When customers have a question or issue, they will be able to launch the AI agent to see if the answer is already available in your knowledge base.
To use the AI agent, click on the click on the Search or Ask AI bar in the top right of your workspace and enter your question into the text box using natural language.
Customers can ask follow up questions, about the same issue or a separate one, within the same conversation.
User feedback
Users can mark each Agent answer as helpful or unhelpful via the Thumb up / Thumb down icons below each answer. This information is used to improve the Ai Agent's performance over time.
Agent conversation history
Admins are able to view a full history of AI agent conversations by clicking on the AI conversation history in the left-hand Team Toolbar.
Within the Conversation History interface, admins can see:
Full conversation history
Agent answers that were marked as Helpful / Unhelpful
Conversation identifier
Previous Agent interactions / conversations by the same user