XLXueming Luo

Xueming Luo

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xueming Luo has a total of 57 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Artificial Intelligence, Emotional Labor, and Service Operations

    by Zheng Fang, Yuqian Chang, Xueming Luo, Qingsheng Wu, Jaakko Aspara

    Problem definition: Emotional labor is increasingly demanded in service operations, placing tremendous psychological strain on human employees and posing challenges to scalability and sustainability. Our study scrutinizes whether artificial intelligence (AI) service bots may tackle these challenges by examining how and when AI’s engagement in emotional labor enhances economic performance in service operations. Methodology/results: We provide causal evidence from a pair of randomized field experiments conducted in partnership with a firm for loan collection service. Results suggest that, compared with human employees, undisclosed AI service agents display the required emotions (both positive and negative) more accurately. However, AI’s advantage of higher emotion display accuracy does not always guarantee better economic performance. For AI to collect more payments from borrowers than human workers, the displayed emotion must be contextually appropriate. Specifically, AI substantially outperforms human workers in debt collection by 49%–94% when the emotion display instructions are suitable for the collection task (i.e., displaying positive emotions to borrowers with minor delinquency but negative emotions to borrowers with repeated delays). However, when the displayed emotions are unsuitable, AI backfires and performs worse than human employees because of its unwavering adherence to inappropriate emotional display instructions. Further, AI’s performance advantages over human agents are amplified when the suitable emotions involve negative (versus positive) valence. We also leverage the machine learning method causal forest to explore heterogeneous treatment effects across customer segments. Managerial implications: Our research suggests that operations managers should deploy AI to reduce frontline employee emotional burnout, develop explicit emotional labor guidelines for AI, and use negative-emotion AI strategically to boost compliance and efficiency. It is also important to identify emotion-intense operational tasks, target AI when it has clear advantages, and set up continuous monitoring and quality control for AI emotional performance. Funding: Z. Fang acknowledges support received from the National Natural Science Foundation of China [Grant 71925003] and the Double First-Class Initiative of Sichuan University. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2024.1459 .

  • AI orchestrator: How recommendation algorithms shape complementor strategy and market equality

    by Xiaowei (Elliott) Zhang, Siliang (Jack) Tong, Xueming Luo, Zhijie Lin, Jing Li

    Research Summary: This study investigates how AI recommendation algorithms shape complementor strategies and market equality on digital platforms. Using two quasi‐natural experiments from a food‐sharing platform, we examine impacts of sequential algorithmic upgrades: from a location‐based baseline to popularity‐based (PopRec) and then personalization‐based (PersRec). Analyses of over 1.7 million observations reveal that PopRec drives complementors to concentrate on a few offerings, while PersRec encourages new product introduction; yet these strategic shifts come at the expense of one another. Furthermore, PopRec reduces revenues of superstars but boosts revenues for long‐tail complementors, enhancing market equality. Conversely, PersRec exacerbates market inequality by asymmetrically benefiting superstars. By bridging platform orchestration and AI frontiers, this study demonstrates the strategic potential of AI in platform management while underscoring the importance of algorithmic accountability. Managerial Summary: Digital platforms increasingly deploy AI recommendation algorithms to manage ecosystem performance. This study reveals that these algorithms also function as effective orchestration mechanisms, incentivizing and shaping complementor offering strategies at scale. We find that popularity‐based algorithms incentivize complementors to specialize and concentrate on a few core product offerings, while personalization‐based algorithms foster broader new product introduction. Our results demonstrate the vital strategic value of AI for managing platform ecosystems. Crucially, both algorithms present inherent trade‐offs for complementor strategies and trigger unforeseen market dynamics. Platform owners must meticulously design and implement algorithmic systems, balancing ecosystem generativity with operational control and considering far‐reaching market inequality implications.


  • Can Providing Algorithmic Performance Information Facilitate Humans’ Inventory Ordering Behaviors?

    by Yingda Lu, Xueming Luo, Liqiang Huang, Danni Wang

    Over recent years, companies have been increasingly adopting algorithmic decision systems (ADS) to replace humans. In this paper, we focus on how ADS facilitates human managers’ decision making rather than replacing humans altogether in the context of inventory ordering decisions, specifically on the effect of providing ADS performance information to human managers. Using a pair of field experiments, our results suggest that providing ADS performance information can improve their inventory ordering decisions. Interestingly, providing both positive and negative ADS performance information enhances human managers’ inventory ordering decisions, with the latter proving even more effective. Our analyses reveal that compliance and deliberation are two mechanisms through which ADS performance transparency influences inventory ordering decisions. Furthermore, we explore the heterogeneous effect and find that disclosing ADS performance information is more helpful for human managers overseeing products with lower sales and higher uncertainty, thus solving the pain points where human managers rely on algorithmic recommendations the most. Our results also show that ADS performance information serves as an equalizer that benefits low-performing managers more. These results demonstrate the importance of algorithm performance transparency in the human adoption of ADS and shed light on the managerial implications of using ADS within corporations.


  • From Stopping to Shopping: A Field Experiment on Free Return and Free Shipping Retargeting Policies in Online Retail Operations

    by Xueming Luo, Le Wang, Qingsheng Wu, Takeshi Moriguchi

    Problem definition: Online retailers face the challenge of converting website traffic into actual purchases and profits. Specifically, the study examines how effective different forms of retargeting strategies are for engaging consumers who visit online stores but leave without making a purchase. Methodology/results: A field experiment was conducted to evaluate the impact of retargeting strategies offering free shipping (FSR) and free returns (FRR) on customer purchasing behavior and profitability. The key findings reveal that simple product reminders do not significantly boost purchases. In contrast, FSR and FRR implemented late in the shopping process led to significant increases in both purchase rates (18.6% and 31.6%) and targeting profitability (28.2% and 37.8%), respectively. Additional analyses with advanced machine learning methods of causal forest algorithms identified significant heterogeneity of the FSR and FRR effects. FRR was particularly effective among younger customers with less experience on the platform and when the abandoned product was of higher value. In contrast, FSR was more effective among older customers with longer tenure and when the product price was lower. Managerial implications: The findings benefit academics, managers, and decision makers by emphasizing the need for personalized retargeting strategies that address customers’ perceived risks, particularly in the later stages of the shopping process. The research highlights the importance of tailoring retargeting strategies to specific customer profiles to maximize effectiveness and retail profitability. Funding: L. Wang acknowledges the support from the City University of Hong Kong [Grants 7005959 and 7006152]. Supplemental Material: The online appendices are available at https://doi.org/10.1287/msom.2024.0779 .


  • Deep Reinforcement Learning for Sequential Targeting

    by Wen Wang, Beibei Li, Xueming Luo, Xiaoyi Wang

    Deep reinforcement learning (DRL) has opened up many unprecedented opportunities in revolutionizing the digital marketing field. In this study, we designed a DRL-based personalized targeting strategy in a sequential setting. We show that the strategy is able to address three important challenges of sequential targeting: (1) forward looking (balancing between a firm’s current revenue and future revenues), (2) earning while learning (maximizing profits while continuously learning through exploration-exploitation), and (3) scalability (coping with a high-dimensional state and policy space). We illustrate this through a novel design of a DRL-based artificial intelligence (AI) agent. To better adapt DRL to complex consumer behavior dimensions, we proposed a quantization-based uncertainty learning heuristic for efficient exploration-exploitation. Our policy evaluation results through simulation suggest that the proposed DRL agent generates 26.75% more long-term revenues than can the non-DRL approaches on average and learns 76.92% faster than the second fastest model among all benchmarks. Further, in order to better understand the potential underlying mechanisms, we conducted multiple interpretability analyses to explain the patterns of learned optimal policy at both the individual and population levels. Our findings provide important managerial-relevant and theory-consistent insights. For instance, consecutive price promotions at the beginning can capture price-sensitive consumers’ immediate attention, whereas carefully spaced nonpromotional “cooldown” periods between price promotions can allow consumers to adjust their reference points. Additionally, consideration of future revenues is necessary from a long-term horizon, but weighing the future too much can also dampen revenues. In addition, analyses of heterogeneous treatment effects suggest that the optimal promotion sequence pattern highly varies across the consumer engagement stages. Overall, our study results demonstrate DRL’s potential to optimize these strategies’ combination to maximize long-term revenues. This paper was accepted by Kartik Hosanagar, information systems. Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2022.4621 .


  • Rejoinder: Heterogeneous Impact of Brands’ Support for Black Lives Matter on Consumer Responses

    by Yang Wang, Marco Shaojun Qin, Xueming Luo, Yu (Eric) Kou

    We discuss three main lines of comments on our paper: internal validity of the research given consumers’ potential misunderstanding of Black Lives Matter (BLM), challenges to the causal inference strategy, and interpretations of our results. We address them with additional robustness checks and reaffirm our view of the original findings as offering a panel of evidence with a consistent storyline: brands’ support for BLM on social media induces, on average, a sizable negative effect on consumer responses. Whereas the commentary focuses on the average effect of BLM treatment, we highlight that the negative effect is highly heterogenous and can be mitigated when brands’ BLM support is directed at Democratic consumers and delivered in a context where their BLM support can be interpreted as more authentic (e.g., no bandwagon, self-promotion, or cheap talk; consistent history of prosociality; and aligned brand prosocial mission). Furthermore, we want to emphasize the nascent nature of research on BLM and social justice movements for marketing, and our work represents a first empirical deep dive into the nature of consumer responses to the branded support of the BLM movement on social media. It is our hope that this research and the accompanying commentaries can inspire a wealth of new marketing perspectives on this critically important topic.


  • Frontiers: How Support for Black Lives Matter Impacts Consumer Responses on Social Media

    by Yang Wang, Marco Shaojun Qin, Xueming Luo, Yu (Eric) Kou

    This paper studies the direct and moderated impact of brands’ support for Black Lives Matter (BLM) on consumer responses.


  • Predicting Stages in Omnichannel Path to Purchase: A Deep Learning Model

    by Chenshuo Sun, Panagiotis Adamopoulos, Anindya Ghose, Xueming Luo

    The proliferation of omnichannel practices and emerging technologies opens up new opportunities for companies to collect voluminous data across multiple channels. This study examines whether leveraging omnichannel data can lead to, statistically and economically, significantly better predictions on consumers’ online path-to-purchase journeys, given the intrinsic fluidity in and heterogeneity brought forth by digital transformation of traditional marketing. Using an omnichannel data set that captures consumers’ online behavior in terms of their website browsing trajectories and their offline behavior in terms of physical location trajectories, we predict consumers’ future path-to-purchase journeys based on their historical omnichannel behaviors. Using a state-of-the-art deep-learning algorithm, we find that using omnichannel data can significantly improve our model’s predictive power. This enhanced predictive power benefits various heterogeneous online firms, regardless of their size, offline presence, mobile app availability, or whether they are selling single- or multi-category products. Using an illustrative example of targeted marketing, we further quantify the economic value of the improved predictive power and the value of data.


  • Top Cited Papers
    Open Access

    The Janus face of artificial intelligence feedback: Deployment versus disclosure effects on employee performance

    by Siliang Tong, Nan Jia, Xueming Luo, Zheng Fang

    Companies are increasingly using artificial intelligence (AI) to provide performance feedback to employees, by tracking employee behavior at work, automating performance evaluations, and recommending job improvements. However, this application of AI has provoked much debate. On the one hand, powerful AI data analytics increase the quality of feedback, which may enhance employee productivity (“deployment effect”). On the other hand, employees may develop a negative perception of AI feedback once it is disclosed to them, thus harming their productivity (“disclosure effect”). We examine these two effects theoretically and test them empirically using data from a field experiment. We find strong evidence that both effects coexist, and that the adverse disclosure effect is mitigated by employees' tenure in the firm. These findings offer pivotal implications for management theory, practice, and public policies. Managerial Artificial intelligence (AI) technologies are bound to transform how companies manage employees. We examine the use of AI to generate performance feedback for employees. We demonstrate that AI significantly increases the accuracy and consistency of the analyses of information collected, and the relevance of feedback to each employee. These advantages of AI help employees achieve greater job performance at scale, and thus create value for companies. However, our study also alerts companies to the negative effect of disclosing using AI to employee that results from employees' negative perceptions about the deployment of AI, which offsets the business value created by AI. To alleviate value‐destroying disclosure effect, we suggest that companies be more proactive in communicating with their employees about the objectives, benefits, and scope of AI applications in order to assuage their concerns. Moreover, the result of the allayed negative AI disclosure effect among employees with a longer tenure in the company suggests that companies may consider deploying AI in a tiered instead of a uniform fashion, that is, using AI to provide performance feedback to veteran employees but using human managers to provide performance feedback to novices.


  • The Impact of Platform Protection Insurance on Buyers and Sellers in the Sharing Economy: A Natural Experiment

    by Xueming Luo, Siliang Tong, Zhijie Lin, Cheng Zhang

    The sharing economy has radically reshaped marketing thought and practice, and research has yet to examine whether and how platform-level buyer protection insurance (PPI) affects buyers and sellers in this economy. The authors exploit a natural experiment involving an unexpected system glitch during a PPI launch and estimate difference-in-differences models using over 5.4 million data points from a food sharing platform. Results suggest that PPI significantly increases buyer spending and seller revenue, affirming the benefits of this platform-level insurance in the sharing economy. The authors also uncover multifaceted buyer-side and seller-side responses that enable such benefits. PPI increases buyer spending by boosting product orders and variety-seeking behavior. Furthermore, it enhances seller revenue by increasing customer retention and acquisition. This work contributes to the literature by (1) putting a spotlight on the topic of PPI, a platform governance policy that reduces consumer risks and improves the efficacy of sharing platforms; (2) accounting for how PPI alters buyer and seller behaviors on a platform; (3) addressing what types of buyers and sellers benefit more or less from PPI; and (4) offering guidance for managers to improve platform reputation, marketplace efficiency, and consumer welfare in the context of the sharing economy.