The rider experience with Airport rides is going down. What would you do?

Question

The rider experience with Airport rides is going down. What would you do?

Asked at Google over a year ago

1.1k views

Problem Solving category

& 1 other company

Asked at

& 1 other company

1.1k views

How to answer Problem Solving questions

Interview Guide

Answers (3)

To answer this question, Log in.

Top Problem Solving interview questions

Invite members
Invite by email

Add another

SPD_2017 · Answer 1 · 2022-08-16T17:12:31+0000

Clarify:

Do we have more information when you say "rider experience"? Is the experience measured based on wait time, travel time, vehicle features for airport riders, and accessibility? open for assumptions
Is this drop sudden or over time? - over time
Is this issue observed in a specific city or airport or geography? - Yes, let's say SJC airport.
Is there any information on any latest events happening that can potentially cause impacts such as political change, and a rise in travel? - No
Is there any known competition in the ride-hailing business that is causing premium Uber drivers to leave the platform? - No

User personas:

International traveler - took a long flight to the destination, tired and exhausted, carries large luggage,
Domestic traveler - takes 1-5 hour flight, weekend travel plans, carries light luggage
Work traveler- domestic, travels to attend meetings, mostly in a hurry, carries light luggage

Potential cause:

Technical Issue: App failure, less responsive APIs, bugs that are impacting the way riders haul a ride, and ride match occurs.
External Factors: political or env change resulting in reduced popularity amongst premium drivers for Uber this includes construction at the roads /airport, and rallies that are making it difficult to get to and from the airport.
New Competition - new ride-hauling business taking away Uber drivers that are resulting in a less to perfect rider experience.
New product feature launch - Uber recently launched the "Express Match" feature which is specifically designed for airport ride hauling. This way, uber estimates the demand based on the data analysis, and rides are requested before the rider lands at the airport/reaches the curb. This is expected to reduce the wait time resulting in a better rider experience. However, It does look like there is some glitch in the system that is not letting this express match function as expected, resulting in delays / longer wait times for airport riders.

To summarize, I would like to confirm that based on my analysis and background information provided, it doesn't look like #1,2,3 are the cause behind the rider experience going down for airport riders. However, it does strongly point to the #4 where the latest feature has impacted the wait time at the SJC airport riders.

I would reach out to the development and release team that worked on the express match feature and evaluate roll-back until the issue can be identified.
To avoid such issues in the future, I will work towards establishing a process that includes a thorough regression testing plan to make sure App users are not impacted. This can be done by establishing a simulated ML/AI-backed algorithm that shows a preview of what the experience will look like for any new feature.
I would also work with the communications team to inform all the past riders in the 4-6 weeks history that took rides from SJC airport that we have identified the issue and we are working on it.

Anshuman Panwar · Answer 2 · 2022-08-30T03:34:41+0000

Clarify

Are we a cab aggregator like lyft/uber?

Yes

What is exactly meant by rider experience? Ride ratings, wait time, pricing etc?

General rating including wait time

Is it happening in a particular city or all over?

Assume all major US cities

Since when is it happening?

Around a week

Happening round the clock or particular time of the day? Weekdays? Weekend?

Assume happening all through the week

Is it happening only for airport rides? Both drop and pickup?

Yes for airport rides only, both sides

Internal Factors

Since the drop is sudden lets check some internal factors:

Any major changes in the app (UI etc for driver as well as rider app)?
Change in surge pricing algo?
Major bug, latency, A/B test etc?
Downtime
Sign in issues

External Factors

Assuming we cant trace any internal factors let check for some external factors

Major competitor
Bad PR
Driver strike etc?

User Journey

Let briefly look at user journey to see if we can understand the problem more deeply

Rider:

Uses app to book a cab (can pre-book as well for airport rides)
App shows wait time and approximate fair

Driver:

Driver gets a prompt for a new ride (with approximate distance or destination)
Drivers can accept the ride if they want
They can later cancel it as well

Hypothesis

Hypothesis1 : Drivers get paid less now for longer rides and get less incentive for waiting (Airport rides have higher wait time especially pickups)

To validate my hypothesis I will look at the acceptance rate for long rides among drivers or check the acceptance rate specifically for airport rides. If there is a drop in acceptance rate then it shows that drivers are less willing to accept airport rides
I will also check for cancellation rates for longer rides and compare it with the cancellation rate for shorter rides. I will also compare it to the metric value one week back. If there is a surge in cancellations, it strengthen our hypothesis

In this case I will change the incentive structure to pay drivers some incentive for longer rides and wait time.

Hypothesis 2: We recently introduced an incentive for drivers who complete at least n (say 10) rides in a day.

This will encourage drivers to take shorter rides and less drivers will be available for longer rides
To validate my hypothesis I will look at the average ride distance. If the hypothesis is true, this metric will see a drop
Secondly I will check for a change in acceptance rate for short and long distance rides. The acceptance rate for short rides should go up while those for long rides should go down.

In this case we might want to change the incentive based on not only the number of rides but total distance driven as well.

Sathiya Prakash · Answer 3 · 2022-07-04T14:44:26+0000

I'd start off with some clarifying questions:

- What type of company are we? (Primarily cabs/ pooling /only airport rides etc.)

- Is there a particular geographical area where we are seeing this behavior?

- Will we be able to figure out a time period & any specific time/day of the week where we observe this behavior?

- Are we seeing this particularly in one direction (from/to) or is it for both trips in general?

- Are we measuring the experience after the ride? (If so, through how?)

Based on the responses to the above questions, will get a fair idea of the problem and its scope.

Suggestions:

If we are primarily a ride-hailing company then we can check our metrics on # of airport rides taken, # airport rides cancellations after booking, etc, then we could explore solutions by understanding the drivers' issues on airport rides, incentives to take up airport rides, increasing cancellation % for drivers, etc.

If it's a particular geographical area/time period, then we could explore external factors like (if) the journey experience to/from the airport has changed (roads, blocks, traffic increase, etc.), # cab availability.

If it's a 'TO' airport ride issue - We could check metrics like # wait time, ETA changes, # cab availability at the day/time, etc.

If it's a 'FROM' airport ride issue - We could check metrics like customer waiting time for the cab, # cab availability near the airport proximity at the day/time, and # of cancellations by drivers post-booking.

Beyond this, if we get to know how we are measuring this behavior/metric, we can do more qualitative and quantitative analysis.