Diagnose Why Fewer Uber Ride Requests End in Confirmed Trips, Then Fix It
Company: Uber
Role: Product Manager
Category: Analytics & Experimentation
Difficulty: medium
Interview Round: Other
You are a product manager at Uber. The ride confirmation rate, broadly how often a rider's request turns into a confirmed ride, has dipped, and you own both finding out why and getting it fixed. Explain how you would investigate the drop and what you would do once you know the cause.
```hint Locate the leak before guessing why
Lay out the steps between a rider tapping request and the ride counting as confirmed, then find which step and which slice of riders, drivers or app versions accounts for the drop before you list causes.
```
### Clarifying Questions
- Exactly what counts as confirmed: a driver accepting the request, or the trip actually starting at pickup? Are rider or driver cancellations after acceptance counted as failures?
- How large is the drop, did it appear suddenly or build up gradually, and how long has it lasted?
- Is it concentrated in particular cities, ride products, rider or driver app platforms, or times of day, or does it appear everywhere?
- Did anything ship or change around the time it started: an app release, a pricing or dispatch change, a driver incentive change, or an outside event?
### What a Strong Answer Covers
- A precise definition of the metric and of the request-to-trip funnel behind it, with the numerator and denominator checked separately
- Ruling out data and instrumentation problems before treating the drop as real
- Segmentation that pins the drop to a funnel step and to the slices that explain it
- Hypotheses spanning rider demand, driver supply and behavior, product and technical changes, and outside factors, tested in order of likelihood and cost to check
- A fix matched to the confirmed cause, rolled out with a way to verify recovery and with guardrails for riders, drivers and marketplace health
### Follow-up Questions
- Segmentation shows the drop only among drivers still running an older version of the driver app. What do you do in the first day, and what do you change so it does not happen again?
- Drivers are accepting fewer offers while the number of drivers online is flat. How do you tell whether drivers are turning down the trips' economics or not seeing the offers properly?
- After your fix ships, the rate recovers only halfway. How do you decide whether there is a second cause or the remaining gap is seasonal?
Overview: A metric-diagnosis question in which an Uber product manager must investigate a dip in the ride confirmation rate and fix it. It tests defining the request-to-trip funnel, ruling out data problems, segmenting the drop, ordering hypotheses across riders, drivers, releases and outside factors, and verifying a fix with guardrails.