For workflows, see QA and analytics. For export, see Call Data.
Page overview
- Conversation tabs: Switch between Live, Ended, and All to focus on active or completed calls.
- Conversation turns: Shows agent and user utterances in chronological order. Live calls update automatically as new turns arrive. Each turn may display matched Knowledge topics where relevant.
- Caller identity: Displays the user email address if available (for Webchat), or phone number if applicable. Personally identifiable details are shown only to users with PII-level access.
- Environment and metadata: Includes information such as variant name, timezone, timestamp, environment (e.g. Sandbox or Production), and version ID. Active calls display a live duration indicator beside the environment label.
- Matched topics: When topics from the Knowledge area are referenced or triggered, they are shown as tags beneath the corresponding utterance.
- Sources: When the agent retrieves content from Connected Knowledge sources, a Sources tag appears beneath the turn. Click a source name to open an inline preview panel showing the retrieved content. Enable this layer from the Diagnosis dropdown.
- Diagnosis panel: Toggle to view system-level debugging tools and any structured function logs.
- Category tagging: Use the dropdown at the bottom of the screen to assign categories for reporting or follow-up.
- Conversation status: The end of the conversation is visually marked (e.g. “Ended conversation”). Live calls show an “In progress” badge until they close automatically.
Key functionality
- Browse and filter: Navigate and filter through past or live conversations using metadata, time filters, duration, tags, or flows.
- Monitor live calls: Observe ongoing conversations in real time without affecting the call. Live visibility depends on your permission level.
- Drill into specific calls: View a complete transcript, inspect decision logic (if exposed), and assess how the agent handled the interaction.
- Assign categories: Manually label or tag calls for downstream use (e.g. QA review, escalation, training).
- Export data: Export transcripts and metadata for analysis, compliance, or reporting workflows.
PolyScore
Eligible voice conversations display a PolyScore badge at the top of the transcript — a 0–10 quality rating generated automatically by AI. The badge is color-coded (green for 7–10, amber for 4–6, red for 0–3) and expands to show per-dimension breakdowns for Agent Quality and Task Success. Use PolyScore in Conversation Review to:- Prioritize review — sort or filter the conversations table by PolyScore to surface the lowest-quality calls first
- Understand failure modes — expand the dimension breakdown to see whether a low score reflects poor conversation flow, an unresolved task, or caller frustration
- Spot-check high scores — confirm that well-scored calls followed the correct process before using them as benchmarks
Default filters
Conversations defaults to showing Live environment calls within a recent date range. If you are working in Sandbox or have not yet deployed to production, create a saved view with your preferred filters (e.g., Sandbox environment, last 7 days) and set it as your default.Data retention
Related pages
PolyScore
Understand how conversations are scored and what each dimension measures.
Diagnosis
Inspect function calls, flows, latency, and topic citations for a specific turn.
Annotations
Flag transcription errors and missing topics to drive continuous improvement.