Skip to main content
Conversation analysis has two distinct parts: the instructions you save for a review and the results stored with a conversation. Keep both tied to the configuration version used for that record.
Evaluation criteria and data-collection configuration can be saved. Conversation details can return existing analysis. The analysis and evaluation run routes currently read the conversation without computing new results.
You need agent edit access to author settings, or conversation read access to inspect existing results.
Recording pending for Review conversation analysis settings. No playable video is available yet.

Recording pending: Review conversation analysis settings. Follow the written guide below.

1

Define the review

Write success criteria and data fields that answer a concrete question. Preserve existing analysis_items references when reviewing a migrated configuration.
2

Save a reviewable configuration

Use the agent’s Analysis settings or the API draft workflow. Read the saved settings back through versions and drafts. Configuration acceptance does not run analysis.
3

Inspect existing evidence

Read conversation details. Check analysis, its criteria results, collected values, and rationales against the transcript. Treat absent results as unmeasured.
Screenshot pending for analysis field configuration. This illustration is not a product screenshot.

Screenshot pending: Analysis configuration with synthetic evaluation criteria and collection fields. Distinguish saved fields from execution.

No. The current route returns an existing conversation rather than starting computation.
Topic discovery, sentiment analysis, and source attribution are local preview controls. They do not establish saved or running analysis jobs.