Ask clarifying questions to dissect the problem statement:

Me: To reiterate the problem statement, the number of rides completed by Ola has dropped by 15% overall?
Interviewer: Yes.

Me: Is this drop specific to a certain type of ride, like Ola Autos, Minis, or luxury rides, or is it across all ride types?
Interviewer: The drop is more pronounced in certain categories like Ola Minis and Autos.

Me: Are we considering only completed rides or also canceled rides and no-shows in this drop?
Interviewer: Only completed rides are counted. Canceled rides are not included in the total.

Me: Is the number of rides completed calculated by counting the total number of ride bookings that have been successfully fulfilled by drivers clicking on ‘End Trip’ after reaching the destination?
Interviewer: Yes, but we also need a record of the payment to be successfully processed. Only then would it be considered as the ride was completed.

Me: Have there been any changes to how this number is tracked or recorded recently?
Interviewer: No, the calculation method has remained the same.

Me: Have we confirmed that the analytics tools used to measure ride completions are functioning correctly?
Interviewer: Yes, all the analytics tools are working without any issues.

Me: Has this decrease been observed in a specific region or across all regions where Ola operates?
Interviewer: It’s been observed across several regions but is more prominent in urban areas with high demand.

Me: Is this drop sudden, or has it been happening gradually over time? Over what period has this 15% decrease been observed?
Interviewer: The decrease has been gradual, occurring over the last 2-3 months.

Me: Is this drop consistent across all customer segments, or are there specific user demographics where the drop is more pronounced?
Interviewer: The drop is more significant among frequent users and premium customers who often use the app for their daily commute.

Me: Has there been a similar drop in the number of rides booked too?
Interviewer: No, the number of rides booked has remained fairly stable in comparison.

Identify Possible Causes: 

List out all possible internal and external factors that could have led to this situation. The actual cause could be one of them or a combination of both, in some situations. Use data obtained from the clarifying questions to further narrow down.

Me: Now that I have clarified the problem statement, and since we see that the drop is associated with rides specifically being canceled after they were booked as there is no such decrease in number of rides of booked, I am going to explore a few internal factors that might have led to the decrease, then move on to operational and external factors. Does that work?

Interviewer: Yes, this works. You may proceed with exploring internal factors.

Internal Factors: 

(a) App Performance: 

Me: Has there been any update to the Ola app or tech infrastructure recently? Any bugs or slow performance that could be affecting ride completions?
Interviewer: There have been minor app updates, but no major performance issues have been reported.

Me: Could users or drivers be experiencing app crashes or delays in the ride allocation process?
Interviewer: No significant complaints regarding app crashes or delays have surfaced.

Me: Was there any noticeable trend in average session time per user during this period?
Interviewer: No, the average session time per user has remained more or less similar to the numbers prior to this period.

Me: Were there any changes made to the user journey which might have led to incomplete rides?
Interviewer: There have been minor app updates, but nothing that changes the user journey on the whole.

(b) Payment System: 

Me: Have there been any recent issues with the payment system that could discourage drivers from completing rides through the app, such as delayed payouts or payment disputes?
Interviewer: There haven’t been widespread reports of payment issues, though some isolated cases of payout delays have occurred a few months back.

Me: Were the payout delays usually noted for a specific mode of payment?
Interviewer: Yes, it seems to be mostly occurring in digital payments including Credit/Debit cards and Google Pay.

‍

Operational Changes:

(a) Driver Availability: 

Me: Has there been a reduction in the number of drivers available or actively accepting rides during this period? Has the demand or number of customer requests maybe decreased over this period?
Interviewer: The number of drivers on the platform has remained stable and so has the number of customers on an average.

Me: What do we know about the number of rides which were counted as no-shows or canceled - has there been a considered change in this number?
Interviewer: Yes, this number has dropped quite a bit.

‍

(b) Ride Cancellations: 

Me: Could the drop in completed rides be related to an increase in cancellations by drivers after they accept or even start the ride?
Interviewer: Yes, that seems to be a factor. We’ve observed a rise in drivers canceling after accepting rides.

Me: What is the volume of payment trends broken down by Payment mode? Are the canceled rides more prominent in one of the payment modes?
Interviewer: Yes, most of the canceled rides were to be paid for by GooglePay.

Me: Is it possible that there is perhaps a reason linked to payment issues for such rampant cancellations since the canceled rides had their payment mode as a digital channel?
Interviewer: Yes, that seems possible.

Me: Did the users with these canceled rides place another request for a ride soon after the cancellation?
Interviewer: No, in most situations the users never placed requests close to the cancellation.

Me: Could drivers have started canceling rides since they are trying to bypass the app to avoid facing payout delays. This way they can still keep their customers and urge them to pay the driver directly.
Interviewer: Yes, this seems to be the case. Your analysis of the root cause is correct.

‍

External Factors:

Me: Now that I have an idea what the root cause might be, I am just going to quickly run through some probable external factors that might have additionally contributed to the decrease.
Interviewer: Sure, please proceed.

Me: Are there any competitor ride-sharing platforms offering better incentives or lower commissions for drivers, contributing to cancellations on Ola?
Interviewer: While there is competition, no major shifts in driver preferences have been observed toward other platforms.

Me: Could any external factors, such as local events, strikes, or regulatory changes, have impacted ride completions?
Interviewer: No significant external events or regulatory changes have been reported recently.

‍

Analyze the Causes:

Driver Bypassing the App in High-Demand Areas to Negotiate Direct Payments:

Data Analysis: Investigate ride cancellations in high-demand urban regions and connection with digital payment mode rides. Look for patterns in cancellations after rides are accepted, particularly in areas with more ride requests.

Driver Feedback: Collect feedback from drivers about their motivation to bypass the app. There is a workaround to change payment mode from the customer’s side to Cash but this option exists only before the ride starts - confirm the theory that drivers have been canceling the rides solely due to this reason, as it is the only way they can avoid payout delay.

Plan and Implement Solution(s):

Improve App and Payment System:
Ensure that drivers receive timely payouts and transparent fare breakdowns. 

Driver Retention and Compliance:
Offer incentives for drivers to complete rides through the app, such as bonus payments for completing rides during peak hours or in high-demand areas.

‍

Monitor/Analyze:

Set up real-time monitoring and track the effectiveness of the fix deployed to correct the delayed payout:
-Cancellation Rate: The percentage of rides canceled after they are started.
-Completion Rate: The percentage of ride completed after they are started.
-User Satisfaction: CSAT scores of drivers to be analyzed along with their feedback to assess the effectiveness of the fix and additional incentives.

‍