Investigate why conversion fell after a Driver App launch at global scale

Problem Statement Description

Product context: Uber is a mobility and delivery platform; its products include rides, Uber Eats, grocery and retail delivery, freight, driver and courier tools, and marketplace pricing.

Uber has rolled out a new version of the Driver App globally, and shortly after launch, a key conversion metric has fallen. The drop is visible at marketplace scale and may affect the end-to-end trip experience for riders, including business travelers who depend on fast, reliable, and predictable rides for airport transfers, meetings, and work travel.

Your task is to investigate the decline as a root-cause analysis problem. The situation may involve driver-side behavior, app performance, dispatch and matching flows, pricing or ETA changes, regional rollout differences, instrumentation issues, or downstream effects on rider booking and trip completion. You should frame the anomaly clearly before jumping to explanations.

The investigation should consider:

- What “conversion” means in this context, including the funnel step, numerator, denominator, and whether it is driver-side, rider-side, or marketplace-level.

- How to validate that the decline is real, including instrumentation changes, event logging, app-version tagging, time-zone effects, and metric definition changes.

- Segmentation by geography, app version, platform, driver tenure, ride type, airport/business-travel-heavy routes, time of day, and marketplace maturity.

- The driver workflow after launch, including login, going online, accepting trips, navigation, pickup, cancellation, and trip completion.

- Marketplace effects such as driver availability, acceptance rate, cancellation rate, ETA, surge, reliability, and rider drop-off during booking.

- Competing hypotheses across product bugs, UX friction, latency, localization, training gaps, policy changes, operational rollout issues, or external market factors.

- Evidence needed to prioritize hypotheses, isolate causality, and distinguish launch-related impact from seasonality or competitor/local market dynamics.

- Immediate mitigation, communication, monitoring, and prevention mechanisms appropriate for a global launch affecting marketplace liquidity and trust.

The goal is to show how you would structure a rigorous RCA for a high-impact Uber launch issue: define the anomaly, isolate where the funnel broke, identify the most likely root causes using data and operational signals, and outline how the team would contain customer impact while building confidence in the long-term fix.