Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Note
Features in this article are powered by the standard harness, which uses the billing options described in Licensing for agents powered by the standard harness. Learn how to access standard features in Access standard agents and agent flows.
In addition to the native analytics features within Copilot Studio, you can send telemetry data to Application Insights. Copilot Studio supports telemetry at different scopes, depending on whether you need diagnostics for a single agent or centralized observability in OpenTelemetry-aligned format across agents in an environment.
Important
Application Insights is a feature of Azure Monitor, an extensible Application Performance Management (APM) tool that you can use to monitor your live applications. It requires a subscription to Microsoft Azure.
Telemetry scopes
You can collect Copilot Studio telemetry in Application Insights at two scopes: Agent-level telemetry and Environment-level telemetry (preview). The choice determines the monitoring boundary, ownership model, telemetry shape, and the level of observability available for agentic scenarios.
Agent-level telemetry is configured for an individual Copilot Studio agent. The maker or development team responsible for that agent owns it. This export option doesn't include any OpenTelemetry-aligned spans.
Environment-level telemetry is configured centrally for a Copilot Studio environment and is typically owned by administrators, platform teams, or Center of Excellence teams. It provides centralized observability across Copilot Studio agents in the environment using a trace or span-based model aligned with the OpenTelemetry semantic conventions for GenAI, the emerging industry-standard for observing modern distributed and agentic systems. This model makes it better suited for monitoring modern agentic workloads, especially agent invocations, tool execution, outputs, and dependencies across agents.
Scope comparison
The following table compares agent-level and environment-level telemetry in Application Insights:
| Area | Agent-level telemetry | Environment-level telemetry |
|---|---|---|
| Configuration scope | Configured separately for each agent | Configured once for the environment |
| Configuration owner | Maker or developer responsible for the agent | Administrator, platform team, tenant admin, or Center of Excellence team |
| Privilege level | Requires access to the specific agent configuration | Requires tenant-level Power Platform administrator or Dynamics 365 administrator privileges, plus an administrator role in the Dataverse environment |
| Primary audience | Makers and development teams that need agent-specific diagnostics | Administrators and platform teams responsible for governance, operations, and centralized observability |
| Operating model | Decentralized diagnostics per agent or solution team | Centralized observability across agents in the environment |
| Telemetry model | Event-based telemetry | Trace or span-based telemetry aligned with OpenTelemetry concepts |
| Best suited for | Message activity, topic events, custom events, and single-agent troubleshooting | Agent invocations, tool execution, outputs, dependencies, and cross-agent monitoring |
| Primary Application Insights tables | customEvents |
dependencies |
| OpenTelemetry alignment | No | Yes |