Analytics and Insights Overview

Insights is the analytics and monitoring section for your AI agents. Use it to track agent quality, customer sentiment, operational efficiency, and cost performance across executive dashboards, evaluation pages, voice diagnostics, and analytics pipelines.

Document	What It Covers
Usage Insights	Executive dashboard with KPIs, trend charts, outcome distribution, and ROI. Operational analytics with LLM performance, session/trace exploration, and billing-unit consumption reporting.
Quality Insights	Per-agent quality scorecards with side-by-side comparison, status flagging, and quality trends. Aggregated quality health across five evaluation dimensions with trend analysis and issue flagging.
Customer Insights	Intent distribution, sentiment trajectory, frustration detection, and resolution tracking. End-user ratings and verbatim comments captured from chat sessions.
Voice Analytics	Call quality (MOS), ASR accuracy, end-to-end latency, barge-in, and DTMF fallback metrics for voice-enabled agents.
Agent Transfer	Efficiency, queue performance, and human-agent metrics for conversations escalated to human operators.
Analytics Pipelines	Built-in and custom analytics pipelines with a visual node-based editor for defining your own evaluation logic.
Sessions	Runtime observability for all agentic projects. Monitor conversations, inspect execution traces, troubleshoot failures, and analyze performance and token metrics.

Before You Begin

Confirm the following before working with Insights pages:

You must have at least Viewer-level access to the project.
Agent Performance, Quality Monitor, and Customer Insights require analytics pipelines enabled in Settings. Without pipelines, these pages display a placeholder prompting you to enable them.
Voice Analytics requires at least one voice channel deployment to generate data. Without a voice deployment, the page is empty.
Dashboard KPI cards, such as Quality Score and Avg Sentiment, display a dash (–) until pipelines evaluate sufficient conversation data.

Get Started

Build

Test and Evaluate

Deploy

Analytics and Insights

Administration

References

Analytics and Insights Overview

Before You Begin

​Before You Begin

Before You Begin