Mixpanel Product Analytics Certification: Product Analytics

Product Analytics

Product Analytics is the core Mixpanel domain. The exam tests whether you can design and interpret an event-based model: events, properties, user profiles, cohorts, groups, and the reports built on them (Insights, Funnels, Retention, Flows, Impact).

Core rules

• An event is a named action at a time, tied to a distinct_id; properties describe that event. Keep names stable and descriptive, such as Checkout Completed.
• User profiles are separate from events. Write durable profile properties with $set and $set_once; do not bury user traits only in event properties.
• Identity is the join key. identify links anonymous activity to a known user; duplicate or changing distinct_id values split users and break funnels.
• Group analytics attaches events to accounts or teams via group keys, so B2B metrics are not distorted by counting individuals.
• A tracking plan and Lexicon governance keep event and property definitions consistent; renaming is possible, but schema drift is the real risk.
• Funnels need ordered steps and a conversion window; retention needs a cohort definition and return event; Flows shows paths, not causation.

Worked example

You want signup-to-paid conversion. Track Signup Started, Signup Completed, and Purchase Completed, with plan on purchase and signup_source on start. Build a funnel in that order, break down by signup_source, and set a 7-day conversion window. If Signup Completed fires before identify, anonymous and known IDs may split; fix by identifying at account creation and using alias only for genuine pre/post-login merges.

Traps

Putting user traits only on events; high-cardinality properties such as email or timestamp in breakdowns; changing event names without a migration plan; using pageviews as a proxy for product actions; and treating correlation in Flows as causation.