Custom Metrics

What are Custom Metrics?

Custom metrics let you define calculations based on your event data. Instead of manually querying events, you can create reusable metrics that automatically calculate values like total revenue, average session duration, conversion rates, and more.

Metric Types

Count

Count the number of events matching your criteria.

Example: Total number of signups

• •Type: Count
• •Event Names: user_signup

Sum

Sum a numeric property across events.

Example: Total revenue

• •Type: Sum
• •Event Names: purchase_completed
• •Property Path: metadata.amount

Average

Calculate the average value of a numeric property.

Example: Average order value

• •Type: Average
• •Event Names: purchase_completed
• •Property Path: metadata.amount

Unique

Count unique values of a property (e.g., unique users, unique sessions).

Example: Daily active users

• •Type: Unique
• •Event Names: (all events)
• •Property Path: userId

Conversion Rate

Calculate the percentage of users who complete a goal event after a starting event.

Example: Signup to purchase conversion

• •Type: Conversion Rate
• •Numerator Events: purchase_completed
• •Denominator Events: user_signup

Ratio

Calculate a custom ratio between two event counts or sums.

Example: Revenue per user

• •Type: Ratio
• •Numerator: Sum of purchase_completed (metadata.amount)
• •Denominator: Unique count of userId

Creating a Metric

Navigate to Dashboard → Metrics → Create Metric and configure:

• •Name: A descriptive name for your metric
• •Description: Optional explanation of what this metric measures
• •Metric Type: Choose from count, sum, average, unique, conversion rate, or ratio
• •Event Names: Which events to include in the calculation
• •Property Path: For sum/average/unique, which property to aggregate
• •Filters: Optional conditions to filter events (e.g., metadata.plan = "pro")
• •Display Format: How to display the value (number, currency, percentage)

Using Metrics via API

Once created, query metrics programmatically:

curl -X GET "https://serla.dev/api/metrics/metric_123/calculate" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "startDate": "2025-01-01T00:00:00Z",
    "endDate": "2025-01-31T23:59:59Z",
    "timeSeries": true,
    "interval": "day"
  }'

Best Practices

• •Use descriptive names: "Monthly Recurring Revenue" instead of "MRR Metric"
• •Set appropriate display formats: Use currency format for revenue, percentage for rates
• •Add filters for segmentation: Create separate metrics for different user segments or plans
• •Use time series: Enable time series data to visualize trends on dashboards
• •Set goals: Define target values to track performance against objectives

Common Metric Examples

Monthly Recurring Revenue (MRR)

Sum of metadata.amount where eventName = "subscription_created" or "subscription_renewed"

• •Type: Sum
• •Event Names: subscription_created, subscription_renewed
• •Property Path: metadata.amount
• •Filters: metadata.billing_cycle = "monthly"

Churn Rate

Ratio: subscription_canceled / active_subscriptions

• •Type: Ratio
• •Numerator: Count of subscription_canceled
• •Denominator: Count of subscription_active

Trial to Paid Conversion

Conversion Rate: subscription_created (numerator) / trial_started (denominator)

• •Type: Conversion Rate
• •Numerator Events: subscription_created
• •Denominator Events: trial_started