product-analytics

Event tracking requires defining clear naming conventions using object_action format, specifying required and optional properties, and excluding PII from all events. This framework ensures consistent data collection, reduces downstream analytics errors, and maintains privacy standards across your implementation.

A North Star metric aligns teams around a single outcome tied to activation, retention, revenue, or referral goals. The AARRR framework breaks this into Acquisition, Activation, Retention, Revenue, and Referral inputs, allowing you to track which actions drive your primary metric and identify bottlenecks.

Rigorous experimentation requires defining event schemas, tracking variant assignments, and calculating sample size and significance thresholds before launch. This prevents false positives, ensures reproducible results, and lets you confidently validate feature impacts on product metrics.

Yes. Define cohorts by segmenting users based on event properties—such as signup method or plan type—then track their engagement over time using retention curves. This identifies which user segments remain active and which need intervention.

Include required properties for context (user ID, timestamp, event name) and optional properties for analysis (plan, source, variant). Exclude PII, enforce consistent naming, and document each property's type and allowed values to reduce data quality issues and privacy violations.

Document a single source-of-truth tracking plan that specifies events, properties, naming conventions, and privacy rules. Share it with engineering, analytics, and product teams to ensure consistent implementation across web, mobile, and server-side platforms.