Strategic Expectation Setting of Delivery Time on Marketplaces

• 成果类型: Article
• 署名作者: Xie, Si; Sharma, Siddhartha; Mehra, Amit; Azizc, Arslan
• 署名单位: University of Texas System; University of Texas Dallas; Indiana University System; IU Kelley School of Business; Indiana University Bloomington
• 刊物名称: INFORMATION SYSTEMS RESEARCH
• ISSN/ISSBN: 1047-7047
• DOI: 10.1287/isre.2021.0497
• 发表日期: 2024
• 页码: 1965-1980
• 关键词: recommender systems; loss aversion; IMPACT; uncertainty

摘要

Delivery speed is an essential component of the service provided by online delivery platforms. Because improving actual delivery speed is expensive, platforms can instead create a perception of faster delivery by showing a conservative estimate of the delivery duration when a customer places an order. We use detailed transaction-level data from a major food delivery platform to examine the effects of setting conservative delivery speed expectations on customers' likelihood of future purchases and restaurant choices. When delivery is slower (faster) than expected, we find that customers are less (more) likely to purchase again from the platform and the focal (same) restaurant they ordered from. A reduction in purchases from the platform is expected to reduce purchases from nonfocal (other) restaurants as well; however, we find no significant impact. This is possibly because of a spillover effect, as customers may switch their purchasing to these restaurants. Our findings thus highlight the effect of setting conservative expected delivery times in a platform setting. Additionally, we provide evidence that customers with a consistent past delivery experience are less responsive to a single instance of positive or negative delivery performance. We further find heterogeneous effects of slower/faster than expected delivery on new versus existing customers and orders with zero versus nonzero delivery charges. These results suggest that the strategy to set conservative expected delivery times can be judiciously implemented for consumer segments for whom future demand is more likely to improve with a better delivery experience. Finally, we examine the trade-off between current and future demand with the setting of a conservative estimated delivery time. We do additional analysis and also conduct an experiment on Amazon Mechanical Turk to show that the gain in future demand is more than the loss in current demand, thus establishing the efficacy of our suggested strategy.