Find the Step That's Losing You Customers In Minutes

How it works

Every user's first arrival at each step is extracted from the event log. The canonical order of the steps is inferred from how long after a user's first event each step is typically reached (median), so the flow order comes from the data rather than from row order or alphabet. Progression is then treated as monotonic — a user who reached step 4 is counted at steps 1-3 too, which repairs logs with missing intermediate events. From that, the analysis computes users remaining at each step, step-over-step conversion, cumulative conversion from the entry step, absolute users lost per step, and the median elapsed time between consecutive steps.

Use it when you have an event log of people moving through a defined flow — signup, checkout, onboarding, application, support escalation — and want to know which step is costing you the most.

Not for measuring why a step fails (run an experiment), not for flows where users legitimately take different paths in different orders, and not the meta-analysis publication-bias chart that shares the word funnel.

What data do you need?

An event log of a signup flow — one row each time a user reaches a step:

Minimum 60 rows · Best with 3-8 steps, 500-10,000 users

What's in the report?

Standard-library analysis: where in your signup, checkout, or onboarding flow you lose people. Upload an event log (one row per user per step) and get the funnel shape, step-by-step conversion and drop-off counts, the biggest leak in absolute users AND the worst-converting step (they are usually different), how long people take to move between steps, and the same funnel broken down by segment. This is the conversion/marketing funnel — user progression through a flow — not the meta-analysis publication-bias chart of the same name.