The Three-Sided Market of On-Demand Delivery

Abstract

On-demand food and grocery delivery services are subject to cross-side interactions among customers, suppliers, and crowdsourced drivers in a three-sided market. Each customer places an order from a supplier via an on-demand delivery platform, and the order is delivered by an ad-hoc driver, who serves as system catalyst connecting suppliers to customers by acting as the delivery mechanism. This paper investigates the interplay of cross-side interactions, and studies the commissions paid by suppliers and customers and the wage offered to drivers. All three players are price-sensitive, but the customers are also time-sensitive as they avoid long delivery times. All players are heterogeneous in their valuation of the service, and their numbers from each side is endogenously dependent on the price, wage, and commission of the platform. We use continuum approximation of the vehicle routing problem to derive customer waiting times. We validate the results of the analytical model with numerical experiments on more realistic cases. The main insights are the following. First, the platform requests lower commissions and offers a higher wage when maximizing social welfare compared to profit. Second, sensitivity analysis on the market parameters shows that, for example, the driver wage is insensitive to the population of customers and suppliers. Third, the direction of change in the optimal commissions with respect to market parameters is the same in both social welfare and profit maximization. Fourth, the platform subsidizes customers when the market is over-supplied, and the suppliers when the market is under-supplied. Fifth, the supplier surplus is only marginally affected by the product price due to the counter-commissioning reaction of the platform.

Keywords: On-Demand Delivery, Sharing Economy, Multi-sided Markets, Continuum Approximation