2. What it is
Knowledge Pages are authored, page-type knowledge sources with a dedicated editor. They are best for long-form content where structure and narrative context improve answer quality.
Knowledge Pages
Publish deep product knowledge that AI can reliably retrieve
Use rich-text pages for structured explanations, policies, and procedural guidance where depth and context matter.
Rich-text authoring workflow
Knowledge Pages provide a dedicated rich-text editor for long-form support articles and structured documentation that the AI assistant can retrieve in context. The editor supports inline formatting, headings, and structured layouts so teams can author comprehensive guides without leaving the platform. Save protection ensures that in-progress edits are never accidentally lost.
Inline title and rich-text body editor
Write and format full-length articles with a purpose-built editor that supports headings, lists, and inline formatting.
Create and edit flows with save protection
Unsaved changes are protected with confirmation prompts so teams never lose work during the authoring process.
Designed for full-article knowledge content
The page format is optimized for multi-paragraph explanations, procedures, and guides that require narrative depth.
Indexing lifecycle and visibility
Each page is indexed as a page-type source with chunk counts, retrieval counts, and active/inactive status control visible directly in the list view. Teams can see at a glance which pages are actively being retrieved and how many chunks each page generates. Created and updated timestamps make it easy to identify stale content that needs review.
Page source type with chunked indexing
Pages are automatically split into retrieval-optimized chunks and indexed alongside other knowledge types in the same pipeline.
Status and activity toggles
Toggle pages active or inactive to control whether they participate in retrieval without needing to delete and recreate content.
Created and updated timestamps in list views
Track when each page was authored and last modified so teams can prioritize review of older, potentially outdated content.
Update behavior for evolving docs
When content changes, pages are automatically re-embedded so retrieval stays aligned to the latest published version. Content hashing detects whether the text actually changed, avoiding unnecessary re-embedding when only metadata is updated. This keeps the knowledge base current without manual intervention while minimizing compute overhead for minor edits.
Content hashing to detect changes
A hash comparison determines whether page text actually changed, preventing wasteful re-embedding on metadata-only saves.
Re-embedding on content updates
When content does change, new embeddings are generated automatically so the assistant immediately retrieves the latest version.
Metadata-only updates when text is unchanged
Saves that only affect title or status skip the embedding pipeline entirely, making routine maintenance fast and efficient.
3. When to use it
- Multi-step onboarding or troubleshooting guides.
- Policy explainers that need context and examples.
- Feature documentation where sections matter for retrieval.
4. When not to use it
- Exact fixed responses that should never vary (use Custom Answers).
- Tiny, frequently changing facts (use Snippets).
- External docs that should sync automatically (use Knowledge Sources).
5. Setup steps in app
- Open Knowledge Pages and create a new page.
- Write title and rich-text content in the editor.
- Save page and verify chunk and retrieval metrics in the list.
- Edit content when docs change to trigger re-embedding.
- Toggle inactive if the page should stop influencing retrieval.
6. Best use cases
- Comprehensive setup instructions with fallback paths.
- Feature and pricing explainers with nuanced caveats.
- Internal support playbooks that require full context.
7. Common pitfalls and how to avoid them
- Pitfall: pages that are too broad. Avoid by splitting into focused articles.
- Pitfall: stale guidance. Avoid with release-linked update ownership.
- Pitfall: overlap with snippets/custom answers. Avoid by assigning one source of truth per statement.
8. Success metrics to track
- Page retrieval count by article.
- Gap recurrence for topics covered by pages.
- Resolved conversation rate for page-covered themes.
9. Related capabilities
- Knowledge Sources for synced external content.
- Snippets for concise fast-updated text.
- Custom Answers for deterministic exact wording.
- Knowledge Gaps for identifying missing coverage.
Which option should you use?
Use this matrix to choose the right knowledge option for each support intent.
| Intent | Best option | Why | Link |
|---|---|---|---|
| Exact policy wording | Custom Answers | Deterministic response for high-risk phrasing. | /custom-answers |
| Short operational facts | Snippets | Fast to publish and update. | /snippets |
| Multi-step guides | Knowledge Pages | Better for long-form structure and context. | /knowledge-pages |
| Synced external docs | Knowledge Sources | Best for URL crawl and document ingestion workflows. | /knowledge-sources |
| Baseline product framing | Product Profile | Keeps answers aligned with core positioning. | /product-profile |
| Operational constraints and fallback rules | Assistant Contexts | Adds scoped behavioral context without rewriting docs. | /assistant-contexts |
FAQ
10. Knowledge pages FAQs
When should I use Knowledge Pages instead of Snippets?
Use Knowledge Pages for long, structured explainers. Use Snippets for short factual blocks that change frequently. Learn more
Can I disable a page without deleting it?
Yes. Pages support active/inactive toggles so you can temporarily remove them from retrieval. Learn more
What happens after editing a page?
Content changes trigger re-embedding so retrieval uses the updated version; metadata-only changes avoid full re-embedding. Learn more
Launch a conversation layer your whole GTM team can use.
Unify grounded AI, web and WhatsApp coverage, human handoff, and conversation analytics without stitching together separate tools.