Product Manager Metrics & Interview Cases

What you'll build

What you will be able to do

Each lab ends with a concrete result that you build yourself and the checker confirms, cell by cell.

• Compute Fennel's north-star metric from its input metrics for two quarters and find the input that is quietly getting worse.
• Turn a month of daily active users into average DAU, DAU/MAU stickiness and active days per user, compare weekends with weekdays and blend the platforms correctly.
• Compare the iOS and Android onboarding funnels step by step, find the step where Android leaks and size the prize of fixing it.
• Turn two install cohorts into day-N retention curves, measure the lift of the new onboarding in points and in %, and chart both curves on a true time axis.
• Measure how many users adopted the shopping-list feature, how deeply adopters use it, how many are power users and how adoption differs by plan.
• Trial-to-paid conversion fell. You split the drop into a mix effect (more trials from a weak channel) and a rate effect (channels converting worse) and name the main driver.
• Size the yearly revenue of a new feature as a chain of assumptions in a low, base and high case, compare it with the build cost and find the break-even conversion lift.
• Score eight backlog items with RICE, rank them, return the winner with INDEX/MATCH and check whether the top three fit into the team's capacity.