Uber’s Ride Booking Dropped by 20% | PM Interview: Root Cause Analysis
You are a PM at Uber and you observed an average 20% drop in Cab bookings in the last 1 month.
Root cause analysis is a systematic process for identifying the underlying causes of problems or events. It is a critical skill for product managers, as it enables them to dissect complex issues, understand why failures occur, and develop strategies to prevent future incidents.
In the competitive landscape of product management the ability to swiftly and accurately identify the underlying causes of issues affecting product performance is not just desirable — it’s imperative.
This is where the expertise in root cause analysis (RCA) comes into play, standing as a cornerstone skill for any aspiring product manager.
This skill set is invaluable in a product management role, where decisions are not only made to address immediate concerns but also to forge long-term strategic directions for product development and user experience enhancement.
When a product manager effectively uses RCA, they become a pivotal force in transforming challenges into stepping stones for innovation and growth. Hence, interviewers often pose RCA-related questions to simulate real-world scenarios where the candidate’s problem-solving capabilities can be observed in action.
Also Read: Amazon’s Add to cart dropped by 14% | Root cause analysis
How should PM go about this Analysis?
Before delving into this issue, a structured approach to address the problem is needed.
- Firstly, the PM provides a concise overview of the product and service in question, establishing a shared understanding.
- Following this, the PM should pose clarifying questions, ensuring both the candidate and interviewer are aligned in comprehending the issue accurately.
- Subsequently, the PM candidate should dissect the problem, examining it through four distinct factors:
a. Basic Systems,
b. External Factors,
c. Internal Factors, and
d. Factors Beyond Our Control.
4. Then based on the data collected the candidate should summarise the findings and try to pinpoint the factor that caused this issue.
5. Lastly, the PM candidate should offer potential solutions to address the issue and insights on how similar situations might be preempted in the future.”
I have created a Template explaining the structure that I use during the PM interview. If you wish to get the template for free. Share your email id in the comments below and write: “RCA template”.
Problem Statement: Imagine you are a product manager for Uber and you see Uber ride booking dropped by 20 %, Evaluate the reason for the drop and the immediate action you would take in the case.
Uber is a multinational company best known for its ride-hailing service, which allows customers to book a ride from a nearby driver using a smartphone app. In addition to ride-hailing, Uber offers several other services:
1. Uber Eats: A food delivery service that allows users to order food from local restaurants through the Uber app.
2. Uber Freights: A service that matches carriers with shippers.
3. Advanced Technology Group: Focuses on autonomous vehicles and other advanced technology development.
4. Uber for Business: Provides business travel, food delivery, and commuter solutions.
The company went public in 2019 and is listed on the New York Stock Exchange.
Candidate: OK from what I understand Uber ride Booking has declined by 20% in the last 1 month.
Interviewer: Yes we have seen an average drop of 20% over the last 1 month on Uber ride booking.
Candidate: Ok this is an average drop seen over 1 month, right?
Interviewer: Yes
Candidate: Have we checked out the Metrics tracking tool if there is any leak in the tracking of events that might have caused this issue?
Interviewer: Yes we checked, Our metrics calculation system is working fine. No issue here.
Candidate: is this a cyclic trend that is seen periodically?
Interviewer: No.
Candidate: Is this issue reported on any specific platform like Android or iOS?
Interviewer: Drop is seen on both the device
Candidate: Is this drop seen on any particular geographic?
Interviewer: This drop is seen in India.
Candidate: OK, Uber in India is operational in specific metro cities, is the drop seen in any specific geographic?
Interviewer: Yes, this drop is seen in metro cities like Mumbai, Bangalore, Delhi, and Chennai.
Candidate: Any specific area where this drop has been seen? Can we pinpoint a location in these metro cities? Since ride bookings are more prominent from Bus stands, train stations, Airports, Malls, etc. So is the drop seen in any of the specific locations here?
Interviewer: We have seen a very significant drop in Airports. The drop in Airport cab booking is causing our metrics to go down.
Candidate: OK, Is there any specific time of the day when this drop is seen? Like in the daytime, Afternoon, or at night?
Interviewer: We see the drop specifically during the Night time from 1 AM to 4 AM.
Candidate: What do we mean by cab booking? Does it mean Cab request or successful cab booking?
Interviewer: Cab booking means how many users successfully availed the Uber cab. That is they booked the cab, started the ride and completed it successfully.
Candidate: Do we have data on the number of Uber cab requests done during the same time i.e. 1 AM to 4 AM?
Interviewer: Yes, we do have that data.
Candidate: Is there a drop in average cab requests done? Or is the cab request down as compared to 1 month back during the same time?
Interviewer: No the cab request metrics seem healthy, no decrease in that. It has increased over time.
Candidate: Do we have data on the number of cabs available in Mumbai during the nighttime? Has this number decreased in the last 1 months?
Interviewer: yes we have the data. The number of cabs available during the same period of 1 Am to 4 AM has increased by 1 month back.
Candidate: Is the data same for the number of cabs available in the Airport during the time of 1 AM to 4 AM? Is their an increase in number of cabs?
Interviewer: Yes we have that data, their is no change in the number of cabs available near the airport.
Candidate: So in the last 6 months the number of Uber cabs available near the Airport has not changed?
Interviewer: Yes
Candidate: Do we have data on the number of Rides accepted by Uber drivers in these metro cities? Do we see any increase or decrease in that?
Interviewer: The overall number of rides accepted has increased in the last 1 months
Candidate: And the number of Rides accepted near the Airport during the time of 1 Am to 4 AM? Has that increased as well?
Interviewer: No that is decreased actually.
Candidate: Uber has some rivals in India, like Ola, Indrive, and other cab services like Blu Smart, etc. Do we know if the booking has decreased for these cab aggregators during the same time?
Interviewer: No we don’t have data for that.
Candidate: Do we know if these competitors have different pricing at night? Which helps drivers to make more money?
Interviewer: No, we don’t have intel on that.
Candidate: The increased ride requests from passengers, it shows that there is still demand, but there is an issue from the supply side.
I know that Uber pays cab drivers bonuses on completing certain milestones. So have we made any changes to it?
Interviewer: These changes often take place throughout the year. Looking for anything specific?
Candidate: ave we made changes in the last 1 months?
Interviewer: yes we have
Candidate: Have we changed compensation or milestones related to trips from the night?
Intreviewer: 1.5 months back we removed the night-long Long Ride pickup bonus, decreased compensation on the small ride bonus, and removed the Daily 5 ride bonus.
Candidate: What is the nighttime long-ride pickup bonus?
Interviewer: This was a small bonus amount we used to pay the drivers when they would accept the ride which was far from the location of the cab. Now we don’t compensate for the same.
Candidate: So let's say if the driver is far from the airport like 4 KM, the algorithm would show the ride to the Uber driver, and when the driver accepts the ride. An extra amount apart from the ride amount was paid to the driver.
Interviewer: yes right.
Candidate: Is this currently done by competitors?
Interviewer: We currently don’t have data on that.
Candidate: OK, And what are the Small ride bonus and Daily 5 ride bonus?
Interviewer: A small ride bonus is when the Driver accepts 5 small rides, rides which are less than 6km, a bonus was given. The Daily 5 ride bonus was when the driver completes 5 rides in a day a bonus was given now it is moved to 20 rides.
Candidate: OK, I know that the cab has to pay a fee to enter the Airport. Are there any changes to that in the past 6 months?
Interviewer: No changes to airport parking or entry fees.
Candidate: Drivers have a different Uber app experience. Do we know if there are any changes made to that?
Interviewer: No significant changes were made in the Drivers app experience.
Candidate: I know that the Uber app goes into dark mode during nighttime and the UI changes. So any significant changes in the app during night mode?
Interviewer: No changes.
Candidate: Was any issue reported by the cab drivers for the app?
Interviewer: Nothing out of the blue, some glitches which were fixed.
Candidate: issues related to rides acceptance or booking?
Interviewer: No, issues were raised regarding FAQ pages. But those are fixed.
Candidate: Any buzz on social media, against the company? Or issues mentioned on social media?
Interviewer: No, nothing related to app functioning.
Candidate: Any backlash in social media related to Compensation removal?
Interviewer: Yes, we do see those when compensation changes are done. So they are routine.
Candidate: Did we see the same when the above 3 compensation changes were done?
Interviewer: yes.
Candidate: Any special offers or marketing campaigns run by competitors for nighttime rides? Do we know anything about that?
Interviewer: Nothing that we know of.
Candidate: Has there been any incident with the Uber rider or driver in recent times? Like robbery, assault, etc?
Interviewer: no nothing in the last 6–7 months.
Candidate: OK I believe I have enough information to go ahead and create a hypothesis, about what caused the drop in Uber ride booking near Airports during the night in the last 1 month.
What caused the Issue?
Based on the thorough analysis and investigation I was able to narrow down the issue to 1 possible cause, because of which the drop might have occurred.
Change in Compensation type: With the removal of night compensation, It might not be profitable for Uber drivers to go to the Airport and pick up passengers. Drivers need to pay the Aiport fees for taking the cab in. So it might not be economical for them to take ride from Airport.
During late night time, there are fewer passengers who might be traveling to the airport. Most of the passengers whose flights are between 1–4 AM reach the airport around 10–11 PM. So there is the possibility that cabs might not be present around 1–4 AM time.
What will be my next steps?
Now that I have a hypothesis I need to validate the same.
Introduce a new feature which asks users to pay small nominal fees to get a cab at the earliest to the airport. The UI will show the possible waiting time without the nominal fee and the waiting time with the nominal fee.
The user who pays the fee will get a cab at the earliest as compared to the user who does not pay the fee. Drivers who completed a ride nearby will be directly shown a New “Star Ride” available. Which means it has extra monetary benefits to accepting the ride.
This will motivate the drivers to service these passengers and will help fill the demand-supply gap in the airport during the nighttime.
This feature needs to be A/B tested to validate if it is the true issue.
Final thoughts
Here the candidate does a good job of trying to evaluate all the possible cases that might have caused the issues and then systematically digs deeper where he sees the possible cause of the issue.
It is not always possible to follow a given path in RCA-type of questions. But it's important to have clarity and clarity of thought to address these cases.
RCA-type questions never have one right answer. There might be multiple reasons for an issue to happen.
So don’t get fixated on the solution, always try to get hints from the data the interviewer is providing.
Sometimes what you might think is a problem, might just be part of the issue. So always ask for more insights and let the interviewer direct the interview.
The ability to perform a root cause analysis is essential for product managers to ensure they’re not just treating symptoms but are solving the core issues that can lead to improved product performance and user satisfaction.
Interviewers look for candidates who can demonstrate a structured, analytical approach to problem-solving, combined with a user-centric mindset and a data-driven decision-making process.
