How would you reduce ride cancellations on Uber? [Uber PM Interview Question]

Interviewer: How would you reduce ride cancellations on Uber?


Visit Topic-wise & Company-wise preparation.

Book a slot for a Mock Interview and in-depth analysis

All the links can be found on the homepage.

Shailesh: Okay, before I jump onto the solution, I would like to clarify some points for better understanding. So the ride can be cancelled from both the parties rider and the driver? Which side are we talking about?

Interviewer: We see more cancellations from the driver's side.

Shailesh: From how long have we seen this trend?
Interviewer: You can say from a year, we see these cancellations.

Shailesh: Okay, are we seeing these cancellations from particular geography?
Interviewer: Let’s say we see this trend globally.

Shailesh: And are we seeing these cancellations in a particular time like peak hours, non-peak hours, or midnight?
Interviewer: We are seeing these cancellations almost all day.

Shailesh: Are we seeing these trends for a particular type of people, any specific demographic of the drivers?
Interviewer: All types of drivers

Shailesh: Since Uber has both iOS and Android apps, is any specific app on which we see these cancellations?
Interviewer: No, it’s platform-agnostic; we see it on both the platforms.

Shailesh: Alright, then let’s deep dive into the question. First, let me list the reasons and pain points for drivers who cancel their rides.

  1. It may be the case that the drivers don’t want to have ridden in some particular direction.
  2. It may be the case that the drivers are just leaving for their home and want to end the day.
  3. One of the most important reasons can be that they don’t want shorter rides.
  4. Sometimes the rider takes too long to come to the pickup location.

For these broad reasons, I am going ahead with the possible solutions.

  1. Uber’s goal is to make transportation readily available for all, don’t quote me in words, but it should be their broad objective. In case the driver doesn’t want to go in a particular direction, we can impose a penalty on the driver because frequent cancellations will have a poor user experience for the rider, ultimately impacting Uber’s brand equity. In this case, Uber can allow 1/2 cancellations per day, after which the driver has to bear a penalty for the same. The driver should also prompt that it will affect his rating.
  2. In case the driver wants to go home and end the day, the Uber algorithm can work such that in the next 4–5 minutes of switching off, it can search for the new ride near to the driver’s home if it can find the ride, connect it, if not switch off the day for the driver.
  3. In case the reason is that they don’t want shorter rides, Algorithm can be adjusted in such a way that if the demand is soaring high, then connect the rider with varying distances like one long ride after two short rides, or we can incentivise the driver for completing a particular number of short rides like a bonus for a driver for every five short rides.
  4. If the rider is taking too long to come to the pickup location, the rider should be prompt that the driver has arrived at the location. If he isn’t making up to the location within 5 minutes of the cab's arrival, he will be charged, and his rating will be affected.

We can try these solutions in a particular region for A/B testing and can track the following metrics:

  1. Reduction in % cancellation of the rides from the driver's side
  2. Increase in the number of rides in the 5 minutes of switching off mode
  3. % increase in the shorter rides

Comments

Popular posts from this blog

Guesstimates for Product Interviews, Tips to ask clarification Questions

Design Uber for Kids. [Product Management Interview Question]

System Design Concepts for Product Managers — Part 1