Crowdsourcing and pre-sale competition in online platforms

Abstract

This thesis studies two operational management problems on common online platforms. We first consider a creative crowdsourcing platform that provides on-demand creative work such as design, editing and even programming. Typically, a client submits a project requirement together with a budget to the platform, and the platform posts the project information to individual service providers for them to focus on. To ensure the service ability over time, we construct a stochastic sequential allocation model where the platform controls the maximum number of participants for each project. Our results show that under mild conditions the optimal policy follows an inverted-U shape, and the platform can take advantage of the market volatility to judiciously gain more revenue. The work also contributes to some the strategic level decisions. In particular, a platform can pursue market share expansion with marginal revenue loss when there are ample resources; and merging can bring in materialized benefits only when two platforms are suitably different in terms of the resource and demand ratio. We then study the presale competition on common e-commerce platforms, where it serves as a preheating tool for platform-level sales events. Such presale promotion offers a pre-ordering option with a discounted price, and is committed by paying a nonrefundable deposit. We consider a two-period model, where consumers arrive in both periods and decide to pre-order in the first period with uncertain values, or to purchase in the spot period after their uncertainty is resolved. We analyze the effects of presale on competing sellers in an asymmetric market. Our results show that a seller's equilibrium choice is to prevent their pre-ordered consumers from leaving their market, instead of making profits on the non-refundable part. Under common assumptions presale is able to benefit competing sellers by taking advantage of consumers' uncertainty, and mitigates the competitive advantage of the seller with higher popularity level and reduce the profit gap between the two sellers.

Date of Award	2020
Original language	English
Awarding Institution	The Hong Kong University of Science and Technology