> ## Documentation Index > Fetch the complete documentation index at: https://koreai.mintlify.app/llms.txt > Use this file to discover all available pages before exploring further. # Agent AI Analytics Overview (Beta) [Back to Analytics Overview](/ai-for-service/analytics/overview) The Overview section gives supervisors and administrators a date and channel-filtered view of how agents interact with the Agentic Copilot suggestions, search results, and conversation summaries. Use Overview to: * Monitor suggestion adoption rates. * Identify override patterns. * Measure summary generation performance. ## Access Overview 1. Sign in to **AI for Service**. 2. Go to **Agent AI** > **Analytics**. 3. Select **Agent AI** > **Overview**. 4. Filter data based on the following table: | Filter | Description | | ---------- | ------------------------------------------------------------------------- | | Date range | Filters all metrics by the selected time period. Defaults to **Today**. | | Channels | Filters data by channel—Chat, Voice, Email. Defaults to **All Channels**. | image ## Session and suggestion metrics The top of the Overview section shows four summary metrics for the selected date range. Each metric card includes a delta indicator (↑ or ↓) that compares the current period against the previous equivalent period. | Metric | Description | | ----------------------- | -------------------------------------------------------------------------------------------------- | | Agent AI Sessions | Total Agent AI widget sessions—My bot, search, and Agentic Copilot—across the selected time range. | | Total Suggestions | Total Agentic Copilot suggestions shown to agents across all sessions. | | Used Suggestions | Total suggestions used—copied or sent—across all sessions. | | Suggestions Per Session | Average number of suggestions per session. | ## Agent AI suggestions acceptance rate The **Agent AI Suggestions Acceptance Rate** section breaks down how agents interact with the Agentic Copilot content. It contains two sub-sections: * **AI Suggestions**: Displays the Agentic Copilot suggestion usage metric. * **Search Results**: Displays the search result usage metric. ### AI suggestions The **AI Suggestions** panel displays how agents handle the AI-generated suggestions. A bar chart compares the Copied, Sent, and Overridden values. | Metric | Description | | ----------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Total Suggestions | Total AI-generated suggestions presented to agents. | | Copied | Suggestions agents copied. Displays as a count and percentage of responses. | | Sent | Suggestions agents sent directly to customers. Displays as a count and percentage of responses. | | Overridden | The number of custom requests agents sent to the Agentic app when the AI suggestions didn't meet their expectations. Displays as a count and percentage of responses. | | Avg Time | Average time the system takes to generate each AI suggestion. | #### **View All** Select **View All** to open the dialog lists per-conversation suggestion data in a table with the following columns: | Column | Description | | --------------------------- | ------------------------------------------------------------------------------------------- | | Conversation ID | The unique identifier for the conversation. | | Involved App | The Agentic App associated with the conversation. | | Total Number of Suggestions | The total count of AI-generated suggestions presented to the agent during the conversation. | | Copied | The number of suggestions the agent copied during the conversation. | | Sent | The number of suggestions the agent sent directly to the customer. | | Overridden | The number of times the agent sent a custom request instead of using the AI suggestion. | | Date & Time | The date and time the conversation occurred. | Overview-View-All Select a **Conversation ID** to view the Conversation Log. It shows the conversation summary, transcripts of customer-agent interactions, agent feedback on widget performance and AI suggestions, and conversation identifiers. Refer to [Conversation Log](/ai-for-service/analytics/contact-center/agentai-conversation-logs#conversation-log) for more details. ### Search results The **Search Results** panel displays how many search responses agents received, copied, and sent to customers from the **Search** tab. | Metric | Description | | ------------- | ----------------------------------------------------------------------------------------- | | Total Results | Total search results the Search tab displays to agents. | | Copied | Search results agents copied. Displays as a count and percentage of responses. | | Sent | Search results agents sent to customers. Displays as a count and percentage of responses. | ## Closing summaries The **Closing Summaries** panel displays how agents handled AI-generated end-of-conversation summaries. A bar chart tracks Summaries Generated and Summaries Submitted over the selected date range. | Metric | Description | | ---------------------------- | ------------------------------------------------------------------ | | Summaries Generated | Total end-of-conversation summaries the Agent AI widget generated. | | Summaries Submitted | Summaries agents submitted. | | Avg. Summary Generation Time | Average time, in seconds, to generate a summary. | ## Mid-conversation summaries The **Mid Conversation Summaries** panel displays metrics for intermittent summaries agents generate during active conversations. A bar chart tracks Total Summaries and Total Conversations over the selected time range. | Metric | Description | | ----------------------------------------- | -------------------------------------------------------------------------- | | Total Summaries | Total intermittent summaries agents generated. | | Total Conversations | Conversations in which agents generated at least one intermittent summary. | | Avg. Summaries Generated Per Conversation | Average intermittent summaries agents generated per conversation. |