Generative AI pulls information from your help center and connectedknowledge sourcesto provide answers to your users. Content is split into smaller chunks and stored in a database for generative AI, so it can find the best matching content to generate replies.

What's my plan?
All Suites Team, Growth, Professional, Enterprise, or Enterprise Plus
Support Team, Professional, or Enterprise

Summary: ◀▼

Generative AI uses chunked knowledge content to provide accurate answers. To improve AI responses, ensure your content is clear, focused, complete, and self-contained. Use familiar wording, define terms, avoid vague pronouns, and structure content with headings, lists, and clear instructions. Include text for images and avoid complex tables. These practices help AI deliver better, faster support answers from your knowledge base.

Generative AI pulls information from your help center and connected knowledge sources to provide answers to your users. Content is split into smaller chunks and stored in a database for generative AI, so it can find the best matching content to generate replies.

Generative AI works best when content is clear, concise, and complete. If you’re using generative AI features, you can optimize your articles to improve the quality and accuracy of generative responses.

Tip: If you're on Professional or Enterprise, try the knowledge copilot (EAP) to get a clear view into AI readability of your content.

This article contains the following topics:

  • Understanding how knowledge sources are used by generative AI
  • Best practices for optimizing content for generative AI

Understanding how knowledge sources are used by generative AI

For generative AI to use a knowledge source, the text is ingested and split into smaller chunks. Chunking is the first step in retrieval-augmented Generation (RAG). RAG lets a large language model (LLM) use information from connected knowledge sources.

Here's an overview of the process:
  1. Knowledge content is ingested and split into chunks.
  2. Each chunk is stored in a database with a numerical vector, representing the semantic meaning of the chunk.
  3. When a user message is sent, the meaning of the message is compared with the meaning of the chunks in the stored vectors.
  4. Generative AI uses the article title and best matching chunks to generate the reply.

Text chunks created from your ingested content are the source of generative AI, so if the content is clear and complete, the results will be better.

Best practices for optimizing content for generative AI

Generative AI works best when it draws from content that is clear, concise, and complete. Good preparation helps generative AI provide faster and more accurate responses.

Most of these guidelines also apply to good writing, in general, but you might find you want to adjust some of the guidelines to make them work for your human audience too.

Best practices for writing content for AI use

Use these best practices for writing content:

  • Ensure content is focused on the topic. Completely cover a topic and avoid covering multiple, unrelated topics.
  • Provide complete answers. Directly answer a user’s question, don't imply an answer, and provide any context the user needs.
  • Create self-contained content. Include all the information the user needs in the topic and avoid linking out.
  • Keep wording clear and concise. Keep paragraphs short and focused, with sentences that are direct and to the point.
  • Eliminate redundancies. Remove any duplicate articles or conflicting information.
  • Define terminology. Explain terms in full and spell out acronyms on first use.
  • Use wording your users are familiar with. Use wording that your users’ use, and that is used in your product or service.
  • Avoid vague pronouns, such as “it” or “they.” Repeat the noun instead of using a pronoun that relies on earlier context.

Best practices for structuring and formatting content for AI use

Use these best practices for structuring and formatting content:
  • Structure content with headings. Use clear headings and subheadings to break content into logical sections.
  • Use structured lists. Use bullet points for facts or tips and use numbered lists for sequential tasks.
  • Avoid nested instructions. If multiple solutions exist, present each as a separate instruction rather than sub-steps within a broader step.
  • Include complete instructions. List all the steps needed to answer the question or solve the issue.
  • Repeat the question or topic from the title. This ensures that context isn’t lost during chunking, and avoids separating questions from their answers.
  • Add written text for pictures. Include text explaining any images, videos, or diagrams. Generative AI interprets text only.
  • Avoid using tables. In general, it’s harder for an LLM to understand information in tables than in sentences. If you need to use a table, try to keep it simple, with minimal text.
Powered by Zendesk