AI Agent Analytics

Measure your AI agent effectiveness

Track every agent action, measure conversions, and continuously improve results with A/B testing and detailed analytics.

Agent Dashboard

Complete picture of your agent's performance: efficiency metrics, cost per conversation, and technical monitoring in real-time. Track conversion rates, average response time, and other key indicators.

Agent Dashboard

Goals Reached

call_tool → HubSpot → meeting_booked
12
15
17
16
21
19
14
MonTueWedThuFriSatSun
Conversion Rate
3.5%
+0.8%
Goal Cost
$2.40
per conversion

Goals & KPIs

Set measurable goals for your agents: calling specific tools, reaching particular graph nodes, successfully completing scenarios. Monitor execution and optimize agent behavior.

Goal NameTypeTarget
Lead CreatedtoolHubSpot → create_lead
Meeting ScheduledtoolCalendar → create_meeting
Demo Completednodedemo_completion_node

Detailed Cost Reports

See exact cost of each call: tokens used, model, duration, operation type. Analyze costs by periods, users, and scenarios. Find expensive operations and optimize spending.

Cost Breakdown

This Week:$136.00
$24.50
Dec 1
$28.20
Dec 2
$26.80
Dec 3
$31.40
Dec 4
$25.10
Dec 5
Avg cost per call:$1.13
Total agent calls:120
Most used model:Claude Sonnet 4.5

A/B Testing

Create multiple versions of one agent and automatically determine which version performs better. Compare prompts, parameters, and logic on real users without risk.

A/B Test Results

Running • 5 days
Variant A (Control)
Conversion3.2%
Avg. Cost$2.80
Sessions1,245
Variant B+28% better
Conversion4.1%
Avg. Cost$2.65
Sessions1,198
Statistical Confidence:
94%

Cross-platform

Use built-in templates for quick start, connect n8n for visual logic building, or develop custom LangGraph graphs with Flutch SDK. Complete freedom of choice.

Flutch built-inFlutch built-in
LangGraph
n8n
Make