Analytics

Monitor Agent Studio usage with conversation and message metrics.

This is a beta feature according to Algolia’s Terms of Service (“Beta Services”).

Agent Studio analytics help you understand how users interact with your agents. Track conversation volume, token usage, and search activity to monitor performance and optimize costs.

Access analytics

1. Go to the Algolia dashboard and select your Algolia application.
2. On the left sidebar, select Generative AI.
3. Go to Agent Studio > Analytics.

Filters

Use filters to refine the data displayed in the analytics dashboard.

Time period

Select a date range to analyze metrics for a specific period. This helps you identify trends, compare performance across time frames, or investigate specific events.

Agent

Choose to view metrics for:

• All agents: aggregated data across all your agents
• Specific agent: data for a single agent

Metrics

The analytics dashboard displays the following metrics:

Metric	Description
Total conversations	Number of conversations users had with your agents
Input tokens	Tokens sent to the large language model (LLM), including prompts and context
Output tokens	Tokens generated by the LLM in responses
Search calls	Number of times your agents called the Algolia Search tool
Unique authenticated users	Distinct users who interacted with your agents

Token usage

Input tokens represent the data sent to the LLM, including:

• System prompts and agent instructions
• User messages
• Tool outputs (such as search results)
• Conversation history

Output tokens represent the responses generated by the model. Monitoring token usage helps you understand costs and optimize your agent’s prompts.

Search calls

This metric tracks how often your agents use the Algolia Search tool. A high number of search calls indicates users are asking questions that require data retrieval from your indices.