Pricing in Ride Sharing Platforms: Static vs Dynamic Strategies

Abstract

We study pricing strategies in two-sided ride-sharing platforms that facilitate transactions between drivers and customers. In the setting we consider, the platform announces the price of service and customers/drivers react to this price. We model the two-sided market using a queuing framework where customers and drivers arrive into two separate queues and wait to be matched by the platform. The arrival process is stochastic and the rate of arrival depends on the instantaneous price set by the platform. On the completion of service, the platform keeps a fraction of the price paid by the customer as its commission and gives the rest of it to the worker who served this customer. Since the arrival processes as well as the platform’s commission depends on the price per transaction set by the platform, the platform’s revenue is a function of the pricing strategy. Our goal is to characterize pricing strategies which maximize the platform’s revenue.We focus on two classes of pricing strategies, namely, static and dynamic. Static strategies are those where the price per transaction is independent of the number of customers/drivers waiting in the queue. Dynamic strategies allow for the price per transaction to be a function of the number of customers/drivers queued in the system. We characterize and compare the platform’s revenue under optimal static and dynamic pricing strategies.