Product Analytics Fundamentals¶

Product analytics is the discipline of measuring and understanding user behavior within a product to drive business decisions. It combines analytical thinking with data measurement - knowing SQL or Tableau is necessary but insufficient; the core skill is understanding what to measure and why.

Key Facts¶

Product analytics answers: what problem are we solving, who is the audience, how are we solving it, what do we want to achieve, and how do we make money
A product analyst defines key metrics, collects data, analyzes it, runs experiments (A/B tests), and implements findings
The analyst's five-step cycle: Define metrics -> Collect data -> Analyze -> Experiment -> Implement
Analytics should be involved from early design stages, not bolted on after launch
Recommendations must be grounded in numbers, not emotions
"Active user" definition varies by product and must be explicitly defined for each context

Patterns¶

Metrics Pyramid¶

A tool for visualizing processes and finding KPIs. Structure: Business GOAL at top, primary metrics at Level 1, decomposition/secondary metrics at Level 2.

Common errors when selecting KPIs: - Choosing a single indicator (creates blind spots) - Choosing too many indicators (analysis paralysis)

Task formulation must be specific: not "Analyze site traffic" but "Research the impact of monitor resolution on conversion rate and give recommendations."

Vanity vs Actionable Metrics¶

Vanity metrics look good but don't guide decisions: - Total registered users (includes inactive) - Pageviews (without context) - Downloads (no indication of engagement)

Actionable metrics are directly tied to business decisions: - DAU/MAU with precise "active" definition - Funnel conversion rates (step-by-step) - Retention curves (D1/D7/D30) - Feature adoption rate - Revenue per cohort

North Star Metric¶

A single metric that best captures delivered value. Everything measured against impact on this metric. Example: Spotify = "time spent listening per user."

Input metrics (things teams control) drive Output metrics (business outcomes): - Input: feature adoption rate -> Output: retention - Input: onboarding completion rate -> Output: D7 retention - Input: search success rate -> Output: conversion to purchase

Gotchas¶

Product company variables that affect analytics setup: company size/stage, B2B vs B2C, team structure, methodology (Scrum/Agile), tooling, and analytical culture maturity
Marketing analytics and product analytics answer different questions - marketing focuses on channel effectiveness and budget allocation, product focuses on user behavior and feature performance
The metrics pyramid should be reassessed as the product evolves - metrics appropriate for a startup differ from those for a mature product