Document Analyst best practices
This guide helps you understand the best practices for using the Document Analyst feature effectively, ensuring accurate, contextual insights when analyzing your documents.
The Document Analyst feature enables an advanced agentic mode, allowing the agent to analyze uploaded documents directly within a chat session. This lets end-users provide documents for comparison, analysis, or validation against the agentβs knowledge base.
Compare documents against your knowledge base
The Document Analyst works best when it needs to compare a provided document against data in the agentβs knowledge base. Keep in mind:
- The document will only be analyzed within the chat session.
- Analysis is based on matching the documentβs content to what is stored in the agentβs knowledge base.
- Before uploading, confirm the agentβs knowledge base includes all reference materials or datasets you want to compare against.
- Only one PDF can be analyzed at a time; if you upload another, the agent processes only the most recent file.
- For lengthy or multi-part documents, split content into focused PDFs to improve analysis precision.
Example use cases
π Competitor pricing comparison: Upload a competitorβs pricing page and ask the agent to identify differences in pricing tiers, feature sets, or discount policies compared to your offerings.
βοΈ Compliance verification: Provide a regulatory guideline PDF and your draft policy document. Ask the agent to highlight any deviations from required standards.
π©βπΌ Resume matching: Upload a job description and ask the agent to compare it against candidate resumes stored in the knowledge base, flagging the best matches.
π Sales deck validation: Supply a sales pitch deck and ask the agent to verify that it accurately reflects your latest product capabilities.
βοΈ Proposal drafting: Upload an RFP and instruct the agent to draft a proposal outline or full response using your existing service descriptions.
π§βπ« Academic literature analysis: Provide a research paper and ask the agent to compare its findings against core literature in the knowledge base, noting agreements or gaps.
For more information, see:
Updated 5 days ago