Summary
The Astra Dashboard gives you a complete view of how your AI agents are performing across your account. Use it to track key performance metrics, measure business impact, compare Text and Voice AI performance, identify trends, and uncover opportunities to improve your AI agents.
Instructions
The Astra Dashboard is the central place for monitoring the overall health and business value of your AI agents. It provides an account-wide view of performance, helping you understand how your AI is handling customer conversations across different channels.
You can use the dashboard to:
Monitor AI performance across all agents
Compare Text Agent and Voice Agent performance
Measure business impact and return on investment (ROI)
Track conversation trends over time
Identify knowledge gaps and escalation patterns
Filter your dashboard view
By default, the dashboard displays combined data from all active AI agents. Use the filters in the top-right corner to select specific agents:
Filter by agent
Click the All Agents dropdown.
Select a specific agent.
The dashboard will update to show data for the selected agent only.
Filter by timeframe
Click the date range dropdown.
Select the reporting period you want to view, such as Last 7 days, Last 30 days, or another available range.
The dashboard will refresh to display data for the selected timeframe.
Review core performance metrics
The Overview section provides a side-by-side comparison of Text Agent and Voice Agent performance.
Key metrics include:
1. Total conversations
The total number of customer conversations handled during the selected period.
2. Resolution rate
The percentage of conversations successfully resolved by the AI without requiring human assistance.
3. Average handle time
The average time required for the AI to complete a conversation, measured in minutes.
4. Drop-off rate
The percentage of users who leave the conversation before receiving a helpful answer or after expressing dissatisfaction with the AI's response.
Measure business impact and availability
The Availability section highlights how your AI supports customers outside normal operating hours and across multiple languages.
1. After-hours coverage
Shows the total number of conversations successfully handled outside your defined business hours.
2. Weekend coverage
Displays the number of conversations handled on Saturdays and Sundays, with a breakdown by day.
3. Languages served
Shows the number of unique languages your AI used to support customers, reducing the need for multilingual support teams.
Analyze account-wide trends
The Performance section helps you understand how customer interactions are progressing across your account.
1. Conversation funnel
Provides a visual breakdown of conversation outcomes, including:
Total conversations
Resolved conversations
Escalated conversations
Dropped conversations
This helps you understand how customers move through the support journey.
2. Conversation volume trend
Displays conversation activity over time, helping you identify increases, decreases, and traffic spikes.
You can view:
Text Agent performance
Voice Agent performance
3. Top FAQ categories
Shows the topics customers ask about most frequently, such as:
Pricing
Account issues
Product information
For each category, you can view the AI's resolution rate to identify areas performing well and topics that may require additional training.
Identify optimization opportunities
The bottom section of the dashboard helps you uncover areas where your AI can be improved.
1. Knowledge gaps
The Knowledge Gaps panel displays the most common questions your AI could not answer.
To improve AI performance:
Review unanswered customer questions.
Click the Book icon next to a query.
Use the shortcut to add the knowledge to your training data.
2. Escalation reasons
The Escalation Reasons panel shows why conversations were transferred to a human agent.
Common reasons may include:
Complex customer requests
Missing information
System errors
Customer frustration
Reviewing escalation trends can help you:
Improve AI training
Refine conversation flows
Optimize guardrails
Strengthen integrations