When to Broadcast? Inventory Disclosure Policies for Online Sales of Limited Inventory

41 Pages Posted: 27 May 2020 Last revised: 6 Jul 2021

See all articles by Shi Chen

Shi Chen

University of Washington - Foster School of Business

Zibo Liu

Fudan University, School of Management, Department of Information Management and Information Systems

Kamran Moinzadeh

Foster School of Business - University of Washington

Yong Tan

University of Washington - Michael G. Foster School of Business

Date Written: July 3, 2021

Abstract

Online sales of limited inventory such as lightning and flash deals have become popular among e-commerce platforms including Amazon and eBay. Motivated by empirical studies of the impact of inventory information on demand in flash sales (Cui et al. 2019 and Calvo et al. 2020), we study the platform’s best timing of disclosing inventory information to maximize the expected sales in a finite horizon. We analyze the following common policies in practice: “always disclose,” “never disclose,” and the fixed threshold policy. The fixed threshold policy requires the platform to broadcast once the inventory level drops below a preset level. We also propose a time-dependent threshold policy, which requires a comparison between the current time and a time-threshold associated with the current inventory level, and show that the proposed policy is the optimal policy under certain assumptions that are consistent with the empirical findings of the extant literature. For both threshold policies, we devise efficient algorithms to optimize the policy parameters, and we compare them through a numerical study. We find that both threshold policies can significantly outperform the two simple policies of always or never disclose. In particular, when the observational learning (herding) effect dominates the scarcity effect on demand, the fixed threshold policy is near optimal. However, when the scarcity effect dominates, employment of a fixed threshold policy may backfire as the customers may interpret not disclosing as a signal of slow sales. Furthermore, when both effects are not profound, the proposed time-dependent threshold policy can significantly improve the fixed threshold policy, so the platform should employ the proposed policy instead of the fixed-threshold policy. Therefore, our study provides not only effective and efficient algorithms for policy optimization but also guidelines for policy selection.

Keywords: Inventory Management, Information Disclosure, Lightning and Flash Deals, Dynamic Programming, Stochastic Models, Optimization

Suggested Citation

Chen, Shi and Liu, Zibo and Moinzadeh, Kamran and Tan, Yong, When to Broadcast? Inventory Disclosure Policies for Online Sales of Limited Inventory (July 3, 2021). Available at SSRN: https://ssrn.com/abstract=3587816 or http://dx.doi.org/10.2139/ssrn.3587816

Shi Chen (Contact Author)

University of Washington - Foster School of Business ( email )

Michael G. Foster School of Business
University of Washington
Seattle, WA 98195-3200
United States

Zibo Liu

Fudan University, School of Management, Department of Information Management and Information Systems ( email )

Shanghai
China
25011054 (Phone)

Kamran Moinzadeh

Foster School of Business - University of Washington

Foster School of Business
University of Washington
Seattle, WA 98195-3200
United States

Yong Tan

University of Washington - Michael G. Foster School of Business ( email )

Box 353226
Seattle, WA 98195-3226
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
264
Abstract Views
1,099
Rank
210,785
PlumX Metrics