What is a conversion funnel? Instrument a three-step onboarding funnel with events.

What's really being asked

This question checks if you understand that a conversion funnel is an ordered sequence of user actions analyzed to find drop-off points, and whether you can move from abstract definitions to concrete instrumentation. Interviewers want to see that you think in events rather than pages, and that you understand how product analytics tools like Mixpanel count unique users across a time-bound sequence.

The full answer

First, a crisp definition of a funnel as a strict ordered sequence of events that measures what share of users progress from step to step. Second, three explicit onboarding steps chosen for a real product flow such as account signup, profile completion, and creating a first project. Third, exact event names you would fire at each step, for example user_signed_up when the account is created, profile_completed when the required fields are saved, and first_project_created when the initial project is persisted. Fourth, the technical requirement that each event must carry a distinct user_id and timestamp so the analytics system can attribute the sequence to the same person. Fifth, a brief note on conversion windows, meaning you only count users who move from step one to step three within a defined period such as seven days.

The mistakes people make

Describing funnel steps as URLs or page views rather than discrete product actions is a major red flag because modern product analytics uses an event-based model. Another weak pattern is giving vague event names like step_one or page_loaded instead of semantically meaningful names like account_created. Forgetting to mention user identity or timestamps also signals shallow instrumentation experience. Finally, confusing total event volume with unique user counts will hurt your credibility since funnels deduplicate per user.

What usually comes next

The interviewer may ask how you would handle out-of-order events, such as a user completing step three before step two. They might also ask how you would segment the funnel by acquisition source or device type, or how you would choose the conversion window length. Another common thread is how you would act on the data, for example by triggering an in-app prompt when a user stalls at the profile completion stage for more than ten minutes.

A concrete example

Consider a SaaS onboarding flow where research shows 40 to 60 percent of signups never return after day one. You instrument step one with user_signed_up fired after email verification, step two with onboarding_profile_completed fired after the user saves industry and role fields, and step three with first_integration_connected fired after an API key is validated. In Mixpanel you build a funnel from these three events, set a seven-day conversion window, and discover that 80 percent complete profile setup but only 30 percent connect an integration. That gap tells the growth team to add a contextual tooltip at the integration screen, turning funnel data into a targeted fix.