Segmenting two-sided markets

Abstract

Recent years have witnessed the rise of many successful e-commerce marketplace platforms like the Amazon marketplace, AirBnB, Uber/Lyft, and Upwork, where a central platform mediates economic transactions between buyers and sellers. A common feature of many of these two-sided marketplaces is that the platform has full control over search and discovery, but prices are determined by the buyers and sellers. We summarize our results from [Banerjee et al. 2017] where, motivated by this application domain, we study the algorithmic aspects of market segmentation via directed discovery in two-sided markets with endogenous prices. We consider a model where an online platform knows each buyer/seller's characteristics, and associated demand/supply elasticities. Moreover, the platform can use discovery mechanisms (search, recommendation, etc.) to control which buyers/sellers are visible to each other. We develop efficient algorithms for throughput (i.e. volume of trade) and welfare maximization with provable guarantees under a variety of assumptions on the demand and supply functions.

Optimal Search Segmentation Mechanisms for Online Platform Markets

“…Our work is related to the burgeoning literature that studies online platform marketplaces of using control levels other than pricing to influence the market outcomes [5,6,8,25]. Kanoria and Saban designed a framework to facilitate the search for buyers and sellers on matching platforms, and found that simple restrictions on what buyers/sellers can access would boost social welfare [25].…”

“…Our result, displaying only a subset of products to buyers can increase the equilibrium revenue, extends the findings in these two pieces of work to the context of revenue optimization. Banerjee et al studied how the platform should control which sellers and buyers are visible to each other, and provided polynomial-time approximation algorithms to optimize social welfare and throughput [6]. In their model, supply and demand are associated with public distributions.…”

“…The platform can also use other types of control, such as commissions/subscriptions fees [8], to influence the outcomes of markets. The rich control options for online platforms have led to an increasing discussion about the design of online marketplaces with different optimization objectives [5,6,25,30].…”

“…To calculate the social welfare and revenue, we first need to specify demand and supply in online marketplaces. Much of prior work simply represent the demand/supply curves with non-increasing/non-decreasing distributions [6,8]. Instead, we consider a demand and supply function derived from a basic market setting in which each seller has one unit of product to offer, and each buyer demands at most one unit of product chosen from the products displayed to her 3 .…”

“…Our work in this paper is related to work on the design of markets for networked platforms [6,7,27,32]. We present a detailed discussion of related work towards the end of the paper.…”