Is data sharing consistently good with AI service? Impact of data sharing and service strategies in competitive channels

Abstract

AI services based on data sharing have recently become essential for online platforms and retailers. This study examines an online platform that offers AI services to itself and its associated retailer, which competes by offering substitute products. The retailer can share data with the platform or develop alternative service channels. Our research explores the dynamic interaction between the platform’s decision to offer AI services, the retailer’s data-sharing choices, and service channel selection. Our findings reveal that data sharing can significantly enhance AI service quality and boost market demand for the platform’s AI service. However, it also creates a power imbalance between platforms and retailers. For platforms, gaining access to retailer data enhances service quality, increases demand, and boosts revenue. For retailers, data sharing creates a Synergy Effect that improves both price and demand and finally increases revenue. However, data sharing is not always beneficial, as the Synergy Effect may be inefficient or diminished because of the Triple Squeeze Effect, ultimately resulting in profit losses. We also analyze the role of data processing capability and find that increasing capability does not always lead to improved outcomes, particularly when high costs burden the platform. Our results indicate that choosing the platform’s service is usually more beneficial for the retailer due to the Synergy Effect, while the platform’s service provision strategy needs to be cautious. Our study provides valuable managerial insights, suggesting that both platforms and retailers must carefully evaluate the cost-benefit trade-offs associated with data sharing and data processing investments to ensure profitability and sustainable growth.