Root Cause Analysis: Uber Ratings Down
About Uber: Uber Technologies, Inc., is a service provider, that allows users to book a car and a driver to transport them in a way similar to a taxi. It is based in San Francisco with operations in approximately 72 countries and 10,500 cities in 2021.
Problem Statement: The Play store rating has dropped from 4.5 to 3.8 stars in the last 2 weeks. Find out the root cause of the problem.
My understanding of the problem:
- Ratings have reduced from 4.5 to 3.8; the app store ratings are weighted towards more recent ratings to reflect changes and updates that are made on the application
- Since the ratings reflect recent changes and updates, the time frame for consideration for root cause analysis shall be restricted for the last two weeks
Assumptions:
1. Uber app is only on Android
2. Considering Uber India Application
3. Considering the Uber Passenger application where the ratings have dropped and not the Uber Driver application
Let’s dig deeper and try to identify possible reasons that could lead to a reduced rating on the app store:
After going over the external and internal variables, let’s examine each of the aforementioned claims to determine the reason for the downgrade.
External Factors:
1. Regulatory Changes by Government (Theme: Pricing): Rise in GST rates from the existing 5% thereby impacting the cab fares
2. Economic Changes (Theme: Pricing): Increase in petrol/diesel/CNG prices increasing the cab fares
Approach: Since the above two pointers impact on a nationwide basis to all ride-hailing applications, we can check the Play console application for the peers (Ola, Rapido, Meru, etc.) ratings in the last two weeks.
As we notice that the ratings for the peers have not been impacted majorly, we can conclude that regulatory changes and a rise in energy prices have not impacted the application ratings. Although there is a possibility that the competitor absorbed the price rise and user’s did not have any major impact. This will be covered in the next hypothesis.
3. Competitor providing cheaper ride service (Theme: Pricing):
Approach:
Compare the prices for a set of destinations on the peer apps vs. Uber across multiple locations.
The fares on peer applications (Ola, Meru) had a ∆ 2.0% in each city, and hence we can conclude that competitor pricing has not impacted the application ratings.
1. Brand Issues due to an ongoing controversy:
a. Driver-related issues
b. Statements hurting sentiments of a demographic group
Approach:
Check the Play console applications for the following:
· Post this we will start reading the reviews provided on the application, and identify searching for the specific keyword, themes, or topics using highlights
Once identified we will look for similar keywords on the socials media platforms like Twitter using hashtags and trends, google news, LinkedIn
·Similar to this, we found that there was an incident of driver misbehavior 2 days back which was also trending news on LinkedIn and Google that had resulted in a controversy and eventual downgrade in rating.
Internal Factors:
1. Recent hike in cab fares
Approach: As there was a recent hike in the cab fares, we will look if there has been an impact on the number of rides. Since the number of rides booked had a ∆ 5.0% w-o-w; also the fares on peer applications had a ∆ 2.0%, we can conclude that recent fare hikes are not the reason for the drop in ratings.
2. Recent tech update due to which the users are facing issues
Approach: Check with the tech team if there were any application updates on the android platform and simultaneously check the application versions installed on the user device or device information. The entire user journey must be examined to understand where the issues could be.
A tech update was released, however, it was identified that daily active users or number of bookings were in line with the trend. Hence, we could negate that the update released caused the downfall in-app ratings.
User Journey:
3. Withdrawal of certain schemes/discounts
Approach: Internally check with the respective teams if any promotions/discounts were taken back. There were no promotions/discounts withdrawn.
4. Driver supply-side constraint:
a. Drivers not available at a particular location
b. Increased wait time for the user in driver allocation
Approach: Check for driver attrition/count of drivers in the last two weeks, based on different cities and regions. Identify if there was any significant downtrend.
As the driver count or attrition remained as per trend, we could conclude that there was no supply-side constraint.
To identify increased wait time for users in driver allocation, we can look at data points such as screen time after the user presses the “Book” button.
Current data shows no significant surge in screen time for the users. Hence, we could negate the hypothesis.
5. Increased ride cancellations by the driver:
a. Driver does not agree to go to the drop location
b. Driver not accepting the payment mode selected by the user
Approach: Find the trend for cancellation % to the number of daily bookings.
Since we identified an increasing trend in the ride cancellation %, we could attribute it to:
a) Driver does not agree to go to the drop location
b) Driver not accepting the payment mode selected by the user.
We can bifurcate the cancellation % on the basis of payment modes (cash vs. online).
Through the analysis, canceled rides with cash payment were significantly lesser than in online payment mode. Hence, we could identify that the cancellations were directly linked to the payment mode selected by the user.
To further deep down on the issue, we rolled out a survey for the driver partners on the key reasons why online payment modes were canceled more.
Upon receiving the survey response, we understood that the online payment cycle was extended from 3 days to 7 days which was further confirmed by the driver-partner team.
Resolutions for the accepted hypothesis
1. Brand Issues due to an ongoing controversy related to driver mis-behavior:
a. Check on the details for the incident
b. Connect and discuss with the communication and PR teams at Uber, and ask them to suggest company’s communication on this issue. This could be rolled out to Uber’s social media platforms
c. Enable vocational training for driver-partners
2. Increased ride cancellations by the driver:
a. Allow the driver the visibility of the payment mode selected by the user before accepting the ride
b. Allow incentivized methods to promote acceptance of online payments by the driver. For example, a rider accepting online payment gets acceptance priority on the next booking
Disclaimer: All numbers and situations considered are hypothetical and do not represent any real-information. This is prepared only for educational purposes.
