Funnel Analysis is an analytical practice studying the changes over a course of events or steps in a user journey. It frequently follows the flow inside an application towards achieving a specific goal. For example the purchasing of an item on an ecommerce site: the funnel analysis will show how many users go through the full flow of steps to complete a purchase and how many stop (drop off) at which step (aka fallout rates).
The Funnel Analysis is an effective way to analyse conversion and fallout in applications. It is often used as a way to measure performance of similar products, e.g. web shops.
This practice is fed with data from A/B Testing, Multi-vari Testing, and can be one of the final steps for running experiments which started with the Design of Experiments. Feature Toggles can greatly benefit the analysis as they can be used to tweak the application based on live feed of data and show if certain changes would influence the outcome.
The analysis is the final step before decision-making. It aims to produce actionable insight on how the team can further improve the performance of the product and the impact of past changes.