Hi all. So I saw the recent post that @ConsultingGuru posted about a few consulting case studies and sample answers. There’s some overlap with PM interview questions so I was wondering if PM case studies questions can be posted as well. Thanks!
Definitely. Here is one sample question and answer (obviously these are open ended so there can be many good responses).
Question: There is a data point that indicates that there are more Uber drop-offs at the airport than pick-ups from the airport. Why is this the case and what would you do within the product to change that?
Answer:
Let’s look at what the problem could be. It is either a demand-side problem (a smaller number of passengers are requesting Uber trips) or a supply-side problem (a smaller number of Uber cars are available for pick-ups). On the supply side we could be looking at airport fees that the driver may have to cover if they leave without a passenger, having a very long wait time for a passenger, having very complicated parking or pick up zones at the airport, and the fact that the airport is far away from the city center where the driver will have to go next. However, notice that some of these problems are largely related to a demand-side problem and will be ameliorated by addressing the demand-side problems.
On the demand-side, there are a few things to consider. When customers go to the airport, being on time is a priority over most other factors like cost and comfort. However, they are more likely to prioritize cost over other factors when travelling from the airport since they don’t need to make sure they arrive somewhere by a certain time. There are a few hypotheses that could be causing the relatively lower demand:
- This is an indicator that Uber prices are likely higher for pick-ups than for drop-offs. Thus for pick-ups customers go for alternatives such as taxis, public transport, family/friends, hotel shuttles etc.
- The incentive of comfort, brand recognition, reliability etc. that Uber offers over other alternatives are not enough to compensate for the higher costs for pick-ups.
- Waiting time and the time to locate the Uber is a severe downside that decentivizes customers to use Uber when travelling from the airport.
- Arrival flights are often at more reasonable times than departure flights (e.g. very early in the morning) and so customers are more willing to ask family and friends to pick them up in the latter case.
- Travelling abroad means less availability of phone data and cell service so Uber is not an option.
The first regarding price is the most straightforward to tackle and the most likely explanation. Consider that the higher price for pick-ups is likely due to the lower Uber availability, (= number of Uber cars/number of users requesting), which leads to surge pricing. I hypothesize that this value is much lower at the airport than at the homes of the customers where they are travelling to the airport from. We can conduct a two-sample hypothesis test of these ratio at airports and compared to locations near the respective airports customers could be coming from.
Another approach is to conduct a two-sample hypothesis test of the request ratio (= no of users requesting/no of users opening the app). If there is a significant difference here than that is also evidence of much higher pricing for airport pick ups.
We can also do the same tests for waiting time (including waiting for the Uber and finding the Uber car).
Solutions:
As Uber pool and go users are more price sensitive, they are the target demographic for these price-based solutions:
- Allow pre-booking. Let customers know they can pre-book their pick-up cars by inputting their flight number as well. Drivers will be notified if they flight is delayed and plan accordingly. This would avoid the higher prices they see after they land.
- For each airport, have a schedule of arriving flights and approximate number of passengers who will be arriving. Based on these values, alert a set number of nearby Uber drivers to go to the airport that is enough to not trigger surge pricing.
- Offer discounts and coupons for future rides or Uber eats.
Measure the efficacy of these solutions by measuring the number of pick-ups at set airports relative to drop-offs over the near term horizon to see if there’s a statistically significant change. Also measure how many Uber cars are leaving the airport without passengers/total Uber cars leaving airports.
1 Like