AI Monitoring
After you deploy AI Assistant, you can use AI Monitoring to get visibility into how it is used and performs. It lets you understand who uses AI Assistant, which tasks it supports, where it succeeds or struggles, and how that ties to adoption, cost, and business impact.
AI Monitoring is organized around three areas:
- Adoption and engagement in the Usage dashboard
- Quality and performance in Conversation Replay and feedback
- Cost and consumption in the Credits dashboard
Together they help you operate AI Assistant with clear data instead of guesswork.
Usage dashboard
The Usage dashboard is the central place for adoption and engagement across your organization. It connects high-level trends to deeper analysis. From any metric in the dashboard, you can drill into individual conversations to see what drives the numbers.
What you can review
- Prompts over time: Overall usage growth and adoption trends.
- Active users and conversations: How broadly the assistant is adopted across users.
- Average prompts per conversation: Depth of engagement and whether interactions are substantive.
- Feedback (thumbs up / down): Early signals on quality and user satisfaction.
Conversation Replay and feedback
Conversation Replay shows actual AI interactions, not only aggregates. You reach it from the Usage dashboard when you need to move from trends to specific exchanges.
What you can review
- Prompt and response history: The user’s question and AI Assistant replies, as delivered.
- Feedback signals: Which interactions users marked positively or negatively, to prioritize improvements.
- Patterns over time: Recurring themes in strong and weak experiences across many conversations.
This view helps you go beyond roll-ups to see what works, where users hit friction, and how to improve response quality over time. Usage trends answer what is happening. Conversation replay helps explain why.
Credits dashboard
Once you understand usage and quality, the Credits dashboard answers how that activity translates into consumption and cost.
What you can review
- Total credits consumed: Overall AI Assistant usage in credit terms.
- Daily and monthly trends: Spikes, dips, and changes in usage patterns.
- Credits remaining: What is left in your allocation so you can plan before limits or overages.
Use this view to forecast usage, align with budget, and allocate credits across teams more predictably.