AVERAGE

Calculates the average value of a specified property across all matching events.

Step-by-Step Setup when creating an AVERAGE-based metered feature:

1. Navigate to Features

   • Go to Product Catalog → Features
   • Click “Add Feature”

2. Basic Information

   • Name: “Response Time” (or descriptive name)
   • Type: Select “Metered”

3. Event Configuration

   • Event Name: api.response (must match your event data)
   • Aggregation Function: Average
   • Aggregation Field: response_time_ms (the property to average)

4. Usage Settings

   • Usage Reset: Periodic (for monthly averages) or Cumulative
   • Unit Name: ms (milliseconds)

5. Save Feature

Calculation Example

Event Data

[
  {
    "event_id": "evt_001",
    "event_name": "api.response",
    "external_customer_id": "customer_123",
    "timestamp": "2024-01-15T10:00:00Z",
    "properties": {
      "response_time_ms": 10
    }
  },
  {
    "event_id": "evt_002", 
    "event_name": "api.response",
    "external_customer_id": "customer_123",
    "timestamp": "2024-01-15T10:05:00Z",
    "properties": {
      "response_time_ms": 20
    }
  },
  {
    "event_id": "evt_003",
    "event_name": "api.response",
    "external_customer_id": "customer_123", 
    "timestamp": "2024-01-15T10:10:00Z",
    "properties": {
      "response_time_ms": 30
    }
  },
  {
    "event_id": "evt_004",
    "event_name": "api.response",
    "external_customer_id": "customer_123", 
    "timestamp": "2024-01-15T10:15:00Z",
    "properties": {
      "response_time_ms": 40
    }
  },
  {
    "event_id": "evt_005",
    "event_name": "api.response",
    "external_customer_id": "customer_123", 
    "timestamp": "2024-01-15T10:20:00Z",
    "properties": {
      "response_time_ms": -1
    }
  },
  {
    "event_id": "evt_006",
    "event_name": "api.response",
    "external_customer_id": "customer_123", 
    "timestamp": "2024-01-15T10:25:00Z",
    "properties": {
      "response_time_ms": 0
    }
  }
]

Calculation Process

1. Event Matching: All events with event_name = "api.response"
2. Deduplication: Remove duplicate event IDs using anyLast() (each event has unique ID in this case)
   • evt_001 → 10 ms
   • evt_002 → 20 ms
   • evt_003 → 30 ms
   • evt_004 → 40 ms
   • evt_005 → -1 ms
   • evt_006 → 0 ms
3. Average Calculation: (10 + 20 + 30 + 40 + (-1) + 0) / 6 = 99 / 6 = 16.5 ms

Result: 16.5 ms

Use Cases

Response Time Monitoring

Perfect for: API response times, page load times, query performance

{
  "event_name": "api.response",
  "external_customer_id": "acme_corp",
  "properties": {
    "response_time_ms": 185
  }
}

Performance Metrics

Perfect for: Processing duration, execution time, latency measurements

{
  "event_name": "job.processing",
  "external_customer_id": "user_456",
  "properties": {
    "duration_seconds": 45.2
  }
}

Business Analytics

Perfect for: Order values, session lengths, transaction amounts

{
  "event_name": "order.placed",
  "external_customer_id": "merchant_789",
  "properties": {
    "order_value": 125.50
  }
}

System Monitoring

Perfect for: CPU usage, memory consumption, bandwidth utilization

{
  "event_name": "system.metrics",
  "external_customer_id": "customer_101",
  "properties": {
    "cpu_percent": 65.8
  }
}

File Size Analysis

Perfect for: Average file sizes, document lengths, data volumes

{
  "event_name": "file.processed",
  "external_customer_id": "data_user",
  "properties": {
    "file_size_mb": 12.7
  }
}

When to Use AVERAGE

✅ Use AVERAGE when:

• Need mean values over time periods
• Tracking performance or quality metrics
• Measuring typical behavior or usage patterns
• Billing based on average consumption levels

Next Steps

• Creating a Metered Feature - Complete setup guide
• Sending Events - How to transmit AVERAGE events