Measuring Non-Linear User Journeys: Rethinking Funnels Metrics in A/B Testing

InDrive users create an order, receive bids from drivers, choose a suitable one, waits for the driver to arrive, and then starts and completes the trip. In some tests, conversions at the stages that precede the implemented changes can change in a statistically significant way. In this article, we explain how we investigated these behavioral patterns and, based on them, introduced new metrics that helped make test results more interpretable.