Investigate why conversion fell after a Driver App launch

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 recently launched a new version of the Driver App, and soon after the launch, conversion for business travelers has fallen. In this context, conversion may refer to riders successfully moving through the trip funnel, such as opening the app, requesting a ride, being matched with a driver, and completing the trip.

You are the PM investigating whether the Driver App launch caused or contributed to the decline. The issue may involve driver availability, acceptance behavior, ETA accuracy, cancellations, pricing, dispatch, app reliability, or measurement changes. Your task is to structure the root-cause investigation, not to jump directly to a fix.

Business travelers are especially sensitive to reliability, wait time, cancellations, airport and downtown coverage, receipt accuracy, and predictable arrival times. The investigation should separate whether the decline is a true marketplace problem, a segment-specific experience issue, or an instrumentation/reporting artifact.

The experience should consider:

- How to define the conversion metric and identify exactly where the funnel drop occurred

- Pre-launch versus post-launch comparisons across markets, platforms, app versions, time periods, and trip types

- Driver-side signals such as app crashes, login issues, acceptance rates, cancellations, online hours, and dispatch latency

- Rider-side impact for business travelers, including ETAs, surge, failed matches, pickup reliability, and completed trips

- Instrumentation checks to confirm whether event logging, attribution, or dashboard definitions changed during the launch

- Segmentation by geography, airport trips, commute hours, enterprise accounts, driver cohorts, and device types

- Hypotheses that connect Driver App changes to rider conversion outcomes through marketplace liquidity

- Short-term mitigation, longer-term prevention, and how to monitor recovery after any intervention

The goal is to demonstrate a clear RCA approach: frame the anomaly, validate the data, isolate affected cohorts, build and test causal hypotheses, assess business impact, and propose an evidence-driven path to mitigation and prevention.