Asemic

Appearance

Metrics Library Documentation ​

Overview ​

Welcome to our comprehensive Metrics Library. This library provides a standardized approach to tracking, measuring, and analyzing user behavior, business performance, and technical metrics across different types of applications and platforms.

Purpose ​

The Metrics Library serves several key purposes:

  • Establish consistent definitions for common metrics
  • Provide standardized calculation methods
  • Enable cross-platform and cross-product analysis
  • Support data-driven decision making
  • Ensure metric reliability and accuracy

Library Structure ​

1. Core Components ​

Events ​

Events are the fundamental building blocks of our metrics system. Each event represents a specific user action or system occurrence that we want to track.

yaml
table_name: analytics.event_name
tags: [event_type1, event_type2]
columns:
  _id: 
    data_type: STRING
  _user:
    data_type: STRING
    tags: [entity_id_column]
  _time:
    data_type: DATETIME
    tags: [event_timestamp_column]
  # Additional columns...

Properties ​

Properties are derived values calculated from events. They can be:

  • Event Properties (direct measurements)
  • Computed Properties (calculated values)
  • Sliding Window Properties (rolling calculations)
  • Lifetime Properties (cumulative values)

KPIs ​

Key Performance Indicators (KPIs) are high-level user metrics that provide insights into business performance and user behavior.

2. Categories ​

Our library includes metrics for:

  • Standard Activity
  • Monetization
  • User Acquisition
  • Engagement
  • Performance
  • Custom Business Metrics

The following industries are covered with more depth:

  • e-Commerce
  • Mobile Gaming
  • iGaming

Usage Guidelines ​

1. Selecting Metrics ​

When choosing metrics:

  1. Start with standard metrics
  2. Add core metrics for your domain
  3. Add specific metrics for your use case

2. Best Practices ​

Naming Conventions ​

  • Use snake_case for property names
  • Use descriptive, consistent names
  • Include units in metric names when applicable
  • Add clear labels for visualization

Calculations ​

  • Use SAFE_DIVIDE for division operations
  • Include default values for properties
  • Consider data latency in window calculations
  • Document complex calculations

Start exploring our metrics library by selecting your relevant domain and use case from the categories above.