Adding Application Insights Telemetry to Power Platform Custom Connectors

January 07, 2026

One of the most powerful features of Power Platform custom connectors is the ability to add custom code through C# scripts. Today, I want to share a template script that demonstrates how to integrate Azure Application Insights telemetry into your custom connectors for enhanced monitoring and debugging.

Why Add Telemetry?

When building enterprise-grade custom connectors, you need visibility into:

• Request patterns and volume
• Error rates and failure scenarios
• Performance bottlenecks
• User behavior and API usage

While Power Platform provides basic logging through this.Context.Logger, integrating Application Insights gives you:

• Rich querying with KQL (Kusto Query Language)
• Dashboards and alerts for proactive monitoring
• Distributed tracing across your integration landscape
• Long-term retention of telemetry data

The Template Script

I’ve created a comprehensive template that shows how to combine both this.Context.Logger (built-in logging) and Application Insights telemetry in a custom connector script. You can find the complete code in my SharingIsCaring repository.

Key Features

The script demonstrates:

1. Dual Logging Strategy
   • Basic logging with this.Context.Logger for immediate visibility
   • Enhanced telemetry with Application Insights for deep analysis
2. Correlation Tracking
   • Every request gets a unique correlation ID
   • Track requests across logs and distributed systems
3. Error Handling with Context
   • Comprehensive error capture with stack traces
   • Errors logged to both systems for reliability
4. Performance Metrics
   • Request duration tracking
   • Operation timing for bottleneck identification

Configuration

The script uses a single configuration point - your Application Insights connection string:

private const string APP_INSIGHTS_CONNECTION_STRING = 
    "InstrumentationKey=YOUR-KEY;IngestionEndpoint=https://REGION.in.applicationinsights.azure.com/";

Leave it empty to disable telemetry entirely, allowing the connector to function normally while you’re in development.

How It Works

The script intercepts requests and:

1. Generates a correlation ID for tracking
2. Logs basic info using Context.Logger
3. Sends rich events to Application Insights asynchronously
4. Processes your custom logic (the TODO section)
5. Tracks completion with timing metrics
6. Handles errors gracefully with full context

Here’s the event flow:

RequestReceived → OperationProcessed → RequestCompleted
                ↓ (on error)
            RequestError

Sample Event Logging

The template shows how to log custom events with structured properties:

await LogToAppInsights("OperationProcessed", new {
    CorrelationId = correlationId,
    Operation = operationName,
    HasPayload = payload.Count > 0
});

These events appear in Application Insights where you can query them with KQL:

customEvents
| where name == "OperationProcessed"
| extend operation = tostring(customDimensions.Operation)
| summarize count() by operation

Best Practices

1. Fire and Forget - The script sends telemetry asynchronously to avoid slowing down your connector
2. Suppressed Errors - Telemetry failures don’t impact the main request flow
3. Privacy Conscious - Body previews are truncated to avoid logging sensitive data
4. Configurable - Easy to enable/disable by setting the connection string

Getting Started

1. Create an Application Insights resource in Azure
2. Copy the connection string from the Overview blade
3. Update the APP_INSIGHTS_CONNECTION_STRING constant
4. Add your custom logic in the TODO section
5. Deploy to your custom connector

Next Steps

This template provides a solid foundation for production custom connectors. You can extend it with:

• Dependency tracking for external API calls
• Custom metrics for business KPIs
• Sampling strategies for high-volume connectors
• Security filtering to redact sensitive data

Check out the full script on GitHub to see all the implementation details.

Have questions or improvements? Feel free to open an issue or submit a PR in the repository!