Indexed search is how Moveworks answers employee questions from content that has been ingested and stored in advance. When an employee asks a question, Moveworks retrieves the most relevant pre-indexed content and returns a summarized answer with citations.
For the three-step search flow that powers both indexed and live search (plugin selection → retrieval and ranking → summarization), see How Search Works.
Indexed search respects source system permissions at query time. Employees only see results for content they have access to in the source system. For files specifically, Moveworks continuously syncs file permissions alongside content so that access always reflects the current state of the source.
For more, see Permissions.
Do I still need to format articles for snippetization?
Yes. Snippetization is still part of how Moveworks ingests and breaks down articles. Summarization quality depends on the number and quality of snippets produced. Articles optimized for snippetization will produce better results than unstructured ones. See Writing AI-Ready KB Articles for the guide on structuring content for optimal results.
How do ServiceNow Knowledge Blocks affect search quality?
Knowledge Blocks produce well-formatted snippets, which in turn produce better summarizations. Snippets are also used for rendering citations, so well-structured blocks improve the full search experience.
Does Moveworks use meta tags?
Meta tags are used as one signal in the relevance calculation. Relevance is primarily determined by the article title and body matched against the query. Tags that closely overlap with query entities (e.g., Outlook for a query about resetting an Outlook password) receive a relevance boost; generic tags have less impact.
What file types are supported for file indexed search?
Moveworks supports .pdf, .docx, .txt, and .pptx files from repositories like SharePoint Online, Google Drive, and Box.