Analytics Overview

Track agent performance, conversation quality, user engagement, and content effectiveness across all analytics tools

AGO’s analytics tools help you understand how your agents are performing, what users are asking about, and where to focus improvement efforts. Here’s what’s available and where to find it.

Dashboard

The Dashboard (accessible from the sidebar) has ten tabs, each focused on a different area:

Tab	What It Shows	Doc Page
Overview	Key metrics (threads, users, thread-to-ticket rate, positive eval rate), volume trends, distribution charts	Performance Dashboard
Evaluation	Human/LLM evaluation results, CX Score gauge, dimension averages, score trends, agent comparison, satisfaction	Evaluation Dashboard
Knowledge	Top shared documents, top viewed documents, top sources	Knowledge Analytics
Feedback	Feedback type distribution, issue frequency, recent submissions	Feedback Analytics
Tickets	Ticket volume, conversion rate, priority distribution, top requesters/organizations	Tickets Analytics
Topics	Searchable topic list, daily trends, sub-topic breakdown	Topics Analytics
User Analytics	New user registrations, daily active users, top organizations, top users	User Analytics
Team Pulse	Review activity, reviewer breakdown, KO rate, and tags on poorly rated conversations	—
Follow-ups	Follow-up replies and conversation continuations	Follow-up Replies
MCP	MCP call volume, error rates, calls by method/source/tool, duration by method	MCP Analytics

All tabs share the same header filters: time range, organization, user type, and agent.

Open Filters → Conversation source to include all conversations, only AGO conversations, or one source imported through the API. Historical imports contribute to the dates when the conversations took place. Their quality/CX scores and topics use the existing analysis configuration and the associated AGO agent. Participant profiles created for imports do not count as new AGO registrations. Imported transcripts do not provide tool usage, generation costs, latency, or declared satisfaction; MCP analytics are unavailable when an imported source is selected.

AI response speed is available under Developer → Monitoring → Response times.

Browse Pages

Three list pages let you search, filter, and export individual records:

Page	Sidebar Link	What You Can Do
Conversations	Chat History	Filter by date, agent, CX Score, topics, content search. Export as CSV. Click a row to review and evaluate.
Feedback	Feedbacks	Filter by issue type, admin tags, status, reviewer. Assign reviewers and set status inline. Export as CSV.
Tickets	Tickets	Filter by date, status, priority. Export as CSV.

Quality & Evaluation

These features work together to measure and maintain conversation quality:

Feature	What It Does
Thread Evaluator	Automatically scores conversations daily using an LLM. Generates CX Score, dimension scores, and review flags.
CX Score	Weighted score (0-100) across five dimensions. Filter, export, and track trends.
Topic Analyzer	Automatically categorizes conversations by topic using keyword matching and AI classification.
Conversation Labels	Manage the topics and admin tags used to organize conversations.
Conversation Management	Review individual conversations, set manual evaluations, assign reviewers, manage tags.

Getting Started

If you’re new to AGO analytics:

1. Start with the Dashboard — the Overview tab gives you a quick health check
2. Enable the Thread Evaluator — automated scoring surfaces quality issues you’d otherwise miss
3. Set up topics — categorizing conversations makes analytics much more useful
4. Read the tutorial — Monitor and Improve Agent Performance walks through the full workflow

Guides

Guide	When to Use It
Monitor and Improve Agent Performance	First-time setup and ongoing improvement workflow
Diagnose Quality Issues	CX Score dropped or thread-to-ticket rates are rising
Set Up Topic Analysis	You want to categorize conversations by subject
Export Analytics Data	You need data in a spreadsheet or BI tool