Bravin AI

Introduction

Measuring your AI agent's performance is essential for continuous improvement and ROI justification. This guide covers the most important metrics and how to use them.

What you'll learn:

Key performance indicators (KPIs)
How to set up tracking
Interpreting analytics data
Running effective A/B tests

Key Metrics to Track

1. Customer Satisfaction (CSAT)

Score: 1-5 stars or thumbs up/down
Target: > 85% positive
Measures overall satisfaction

2. Time to Resolution (TTR)

Average time to solve a customer issue
Target: < 3 minutes for simple queries
Lower is better

3. Containment Rate

% of conversations handled without human intervention
Target: > 70%
Indicates agent effectiveness

4. Conversion Rate

% of visitors who complete desired action
Target: varies by industry (typically 2-5%)
Direct business impact

5. Response Time

Time from message to first response
Target: < 5 seconds
Affects customer satisfaction

Focus on 3-4 key metrics that align with your business goals. Tracking too many can be overwhelming.

Analytics Dashboard

The Bravin AI Dashboard provides real-time insights into your agent's performance.

Access: Navigate to Analytics → Dashboard

Main Sections:

Overview - Key metrics at a glance
Conversations - Volume, topics, sentiment
Performance - Response times, resolution rates
Customer Satisfaction - CSAT scores and feedback
Conversions - Goal completions and revenue attribution

Create custom dashboards for different stakeholders (management, support team, marketing).

Measuring Customer Satisfaction

Setup CSAT Survey:

Go to Settings → Feedback
Enable "Post-Conversation Survey"
Choose survey type (stars, thumbs, NPS)
Set trigger timing (immediately or after 24h)

Best Practices:

Keep it simple (1-2 questions)
Ask "Was this helpful?" at the end
Offer optional comment field
Send follow-up for negative ratings

CSAT Configuration

{
  "csat": {
    "enabled": true,
    "type": "stars",
    "trigger": "conversation_end",
    "question": "How satisfied are you with this conversation?",
    "followUp": {
      "enabled": true,
      "threshold": 3,
      "message": "We're sorry to hear that. Could you tell us more?"
    }
  }
}

Time to Resolution

Tracking TTR:

Automatically calculated for each conversation
View in Analytics → Performance
Filter by topic, channel, time period

Improving TTR:

Train Knowledge Base - More accurate responses
Optimize Response Length - Concise but complete
Use Quick Replies - Common questions ready
Enable Handoff - For complex issues

A very low TTR (< 30 seconds) might indicate superficial responses. Balance speed with quality.

Conversion Tracking

Setup Goal Tracking:

Go to Analytics → Goals
Click Create New Goal
Define goal (e.g., "Book Demo", "Make Purchase")
Set trigger conditions

Goal Types:

URL Match - Visitor reaches thank-you page
Event - Custom event fired (e.g., button click)
Intent - Agent detects purchase intent
Form Submit - Lead capture form completed

Goal Tracking Example

// Track custom conversion event
bw('track', 'conversion', {
  type: 'demo_booked',
  value: 0,
  currency: 'USD',
  metadata: {
    source: 'chatbot',
    plan: 'enterprise'
  }
});

A/B Testing Best Practices

What to Test:

Agent personality/tone
Response templates
Greeting messages
Call-to-action phrasing
Handoff timing

Running a Test:

Go to Settings → A/B Tests
Choose what to test (e.g., greeting message)
Create variants (A: formal, B: casual)
Set traffic split (50/50)
Define success metric (CSAT or conversion)
Run for minimum 2 weeks or 1,000 conversations

Statistical Significance:

Need at least 100 samples per variant
Look for >10% improvement
Use Bravin's built-in significance calculator

Test one variable at a time to isolate what's driving changes.

Optimization Tips

Weekly Optimization Checklist:

Review conversation transcripts (sample 10-20)
Identify common questions without good answers
Update knowledge base with new information
Analyze negative CSAT ratings for patterns
Test new response variations
Monitor containment rate trends
Check for technical errors or slow responses

Monthly Strategic Review:

Compare metrics to previous month
Review A/B test results
Adjust agent personality based on feedback
Update goals and KPIs as business evolves
Calculate ROI and cost savings

Set up automated weekly reports to be emailed to your team. Go to Analytics → Reports → Schedule.

Measure Performance: CSAT, TTR, Conversions

Introduction

Key Metrics to Track

Analytics Dashboard

Measuring Customer Satisfaction

Time to Resolution

Conversion Tracking

A/B Testing Best Practices

Optimization Tips

Was this guide helpful?

Next Steps

Customize Agent Voice

Train Knowledge Base