Hourly Pay Pilot at CityCab
When CityCab, a food-delivery platform, switched 200 riders in one zone from per-delivery pay to a fixed hourly wage for a three-month pilot, the company expected slower deliveries. Instead, average delivery time fell by four minutes. Riders also began accepting longer trips that they had earlier avoided, and the number of orders left unassigned at peak hours dropped. CityCab's operations team concluded that hourly pay improves service and announced a citywide rollout. The finance team objected: the pilot had been run only in the central zone, where order density is highest and waiting time between orders is shortest. In low-density suburbs, an hourly wage could pay riders for idle time, while per-delivery pay transfers that risk to the rider. The pilot also coincided with a fare discount that raised order volume by 18%. CityCab's CEO wants a decision within two weeks.
1. State the operations team's conclusion and identify two features of the pilot that weaken it.
Read the worked answer
Conclusion: switching riders from per-delivery pay to hourly pay improves service, so CityCab should roll it out citywide.
First weakness: the pilot ran only in the central zone, where order density is highest and riders wait least between orders. The result may not transfer to low-density suburbs, where hourly pay could fund idle time rather than faster deliveries.
Second weakness: the pilot coincided with a fare discount that raised order volume by 18%. Higher order volume can itself reduce delivery times and unassigned orders. The improvement cannot be credited to hourly pay without separating the discount effect. The absence of a control group on per-delivery pay makes this worse.
2. Characterise the finance team's objection: is it mainly about internal validity, external validity, or measurement? Explain.
Read the worked answer
It is mainly about external validity: whether a result found in the central zone can be generalised to low-density suburbs. The finance team is not disputing that delivery time fell in the pilot; it is disputing that the same pay policy will produce the same effect in a different operating environment.
There is also an internal-validity problem in the pilot as a whole, because the fare discount confounds the pay change. But the finance team's specific objection about idle time in suburbs is an external-validity concern.
3. Recommend a decision to the CEO. Include one design change for a second pilot and one operational safeguard.
Read the worked answer
Do not order a citywide rollout yet. The evidence supports a service improvement in one high-density zone, but it does not separate hourly pay from the fare discount and does not test low-density operations.
Design change: run a second pilot in at least one central and one low-density zone, randomising riders within each zone to hourly pay or per-delivery pay, and hold fares constant. Run it long enough to cover peak and off-peak weeks, and track idle time, delivery time, unassigned orders and rider earnings.
Operational safeguard: if hourly pay is extended before full results, use zone-specific pay or a minimum-orders guarantee so that low-density suburbs do not pay riders for long idle periods. Phase the rollout by zone and review after eight weeks.