Setting Up Your First Knowledge Base

Your AI agents are only as good as the knowledge they have access to. A well-structured knowledge base is the foundation of accurate, helpful customer interactions. Here’s how to set one up.

What Is a Knowledge Base?

A knowledge base is a collection of documents, FAQs, and information that your AI agents use to answer customer questions. When a customer asks something, the agent searches this knowledge base and uses the most relevant information to craft its response.

This approach — called Retrieval-Augmented Generation (RAG) — ensures responses are accurate and specific to your business, not generic AI-generated text.

Step 1: Gather Your Documents

Start by collecting the documents your support team already uses:

• Product documentation — manuals, specs, feature guides
• FAQ documents — common questions and approved answers
• Policy documents — return policies, terms of service, SLAs
• Training materials — scripts and guidelines your human agents follow

Step 2: Upload to the Platform

Navigate to the Knowledge Base section in your dashboard and upload your documents. We support:

• PDF files
• Word documents (.docx)
• Plain text (.txt)
• Markdown (.md)

The system automatically processes each document:

1. Parsing — Extracts text content from the document
2. Chunking — Splits content into semantically meaningful sections
3. Embedding — Creates vector representations for fast retrieval
4. Indexing — Stores everything for instant search

Step 3: Organize with Tags

Use tags to organize your knowledge base by topic, product, or department. This helps agents find the most relevant information faster and lets you assign specific knowledge sets to specific agents.

For example:

• Tag billing-related documents with billing
• Tag product docs with product-a or product-b
• Tag policies with policy

Step 4: Test Your Knowledge Base

Before going live, test your knowledge base using the built-in chat interface:

1. Open the Brain AI assistant
2. Ask questions your customers commonly ask
3. Verify the responses are accurate and well-sourced
4. Check that source citations point to the right documents

Step 5: Iterate and Improve

Your knowledge base is a living resource. As you review conversations and identify gaps:

• Add new documents when agents can’t find answers
• Update existing documents when information changes
• Remove outdated content to prevent incorrect responses
• Review analytics to see which documents are most frequently accessed

Best Practices

• Be specific: Detailed documents produce better answers than vague ones
• Keep it current: Outdated information leads to incorrect responses
• Use clear language: Write documents as if explaining to a new team member
• Structure content: Use headings, lists, and clear sections
• Include examples: Sample questions and answers help agents understand context

What’s Next?

Once your knowledge base is set up, you’re ready to configure your first AI agent and start handling customer conversations. The knowledge base will continue to improve as you add more content and refine based on real interactions.

Need help getting started? Contact our team for a guided setup session.