TSTianshu Sun

Tianshu Sun

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Tianshu Sun has a total of 42 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Participation governance in consumer and AI co-creation: Whether to intervene and how to design it

    by Mengxia Zhang, Brian Rongqing Han, Tianshu Sun, Lan Luo, Joseph M. Golden

    Artificial intelligence reshapes digital mass customization by altering the structure of consumer co-creation. When consumers start from a draft generated by artificial intelligence rather than a blank canvas, outcomes are jointly determined by two stages: whether consumers initiate co-creation and, conditional on participation, how much refinement effort they exert. This separation makes participation a governable design margin—an issue that is largely muted in traditional blank-canvas customization, where participation and effort are tightly bundled. Partnering with a leading United States online retailer of personalized photo products, we study whether and how firm interventions at the participation stage shape downstream purchase outcomes. We develop a two-stage model in which, relative to a baseline with organic participation, a nudge lowers start-up frictions and expands voluntary entry, whereas a mandate serves as a strong boundary-case intervention that guarantees entry but can weaken how participation translates into purchase, generating a reach–efficiency trade-off. We examine these predictions in two randomized experiments. In Study 1, a light-touch nudge increases participation by 5.98 percentage points and raises revenue and order frequency. The evidence is consistent with gains arising primarily through expanded entry into consumer co-creation with artificial intelligence. In Study 2, mandates increase participation relative to nudges, but we find no clear evidence of additional gains in purchase or revisit intentions. Instead, mandates increase low-value participation, in which consumers exert minimal effort and report low purchase intention. Together, these findings show that in mass customization enabled by artificial intelligence, participation governance is a consequential design choice: firms must consider not only how many consumers initiate co-creation, but also whether the participation they induce creates downstream value.

  • Sequential Search Transformer: A Deep Structural Econometric Model

    by Yicheng Song, Tianshu Sun

    Modeling and leveraging consumers’ dynamic search behaviors presents significant business opportunities. Although deep learning methods excel at processing vast consumer data for predictive tasks, their opaque nature limits interpretability and fails to explicitly model consumer decision making. In contrast, economic theory suggests that consumers follow a sequential search strategy, evaluating alternatives until they find the best match for their preferences. To bridge this gap, we propose the sequential search transformer (SST), a deep structural econometric model that integrates deep learning with sequential search theory to model search and purchase decisions. SST unifies these two approaches into an end-to-end trainable model, improving both predictive accuracy and policy evaluation capabilities. Unlike conventional deep learning models, SST explicitly models consumer decision making, and unlike existing sequential search models, it enables consumer behavior modeling across sessions and sequentially resolves utility uncertainty for searched items. We provide a theoretical analysis of the identification strategy for the SST model and show that all parameters can be identified under the proposed framework. Then, we apply SST to a data set with detailed clickstream data collected from a U.S. e-commerce website. Empirical evaluations show that SST outperforms state-of-the-art deep learning and structural models in predicting consumer searches and purchases. Moreover, policy experiments demonstrate SST’s effectiveness in optimizing product recommendations and new product promotion strategies, ultimately enhancing consumers experience and driving revenue growth. This paper was accepted by Hemant Bhargava, information systems. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2024.04540 .

  • The Value of Last-Mile Delivery in Online Retail

    by Zhikun Lu, Ruomeng Cui, Tianshu Sun, Lixia Wu

    Last-mile delivery, the most expensive stage of the shipment process, has become increasingly important in online retail, raising the strategic question of whether firms should outsource it to customers via pickup stations or offer home delivery. In this paper, we examine the economic value of last-mile delivery to inform this high-stakes decision. Partnering with Cainiao, Alibaba’s logistics platform, we exploit a quasi-experiment where home delivery service was sequentially rolled out to pickup stations in 2021. A staggered difference-in-differences design reveals that home delivery significantly increases sales and customer spending on Alibaba’s retail platform. Because last-mile delivery is labor-intensive and capacity constrained, effectively allocating delivery resources is crucial. To address this challenge, we propose a novel operations-aware targeting framework that integrates causal machine learning with constrained optimization to identify and prioritize the most valuable customers. The framework incorporates both capacity and fairness constraints to maximize sales while maintaining equitable service access. We further extend it with routing optimization, enabling value-aware delivery planning that jointly considers sales uplift and spatial efficiency. We demonstrate that the resulting targeting and routing policies significantly improve revenue performance. Taken together, we show that last-mile delivery is not merely a cost center but also a revenue driver. By adopting tailored logistics strategies that balance customer needs with resource constraints, online retailers can unlock substantial value and win the last mile of e-commerce. This paper was accepted by Jeannette Song, operations management. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.03991 .


  • Geographical Pattern of Online Word of Mouth: How Offline Environment Influences Online Sharing

    by Tianshu Sun, Yanhao “Max” Wei, Joseph Golden

    A key feature of e-commerce is the access to a wide range of geographical areas; e-commerce platforms spend much effort customizing their strategies across geographical areas. Meanwhile, online word of mouth (WOM) is becoming an increasingly important driver for product sales on e-commerce platforms. This study investigates the geographical pattern of online WOM. Specifically, we examine whether and how customers’ local environment influences the generation and direction of online WOM. Leveraging a unique research design, we measure the online WOM at a U.S. national e-commerce platform by tracking the WOM referral decisions of customers. Whereas digital technologies have enabled customers to share without physical boundaries, we find that location still plays significant roles in the generation and direction of online WOM. First, a customer’s offline social environment (e.g., friend visiting frequency, neighbor interactions) significantly explains the generation of online referrals. Second, online referrals are largely bounded locally, and referrals in areas with more local social interactions are more likely to stay local. Third, even when referrals travel far, they are more likely to point to destinations socially similar to the origins. We derive important implications for firms in allocating resources across geographical areas and in using referrals to identify high-potential geographical markets.


  • Unraveling Generative AI from a Human Intelligence Perspective: A Battery of Experiments

    by Wen Wang, Siqi Pei, Tianshu Sun

    This study introduces a novel, human-centered framework for evaluating the holistic intelligence of large language models (LLMs), using behavioral theory and experimental benchmarks drawn from human intelligence. Through extensive online experiments, the framework reveals that GPT-4 outperforms humans in cognitive, emotional, and creative intelligence, but falls short in social intelligence, especially in social interest, self-efficacy, and understanding mental states. Beyond theoretical insight, the study validates this framework by assessing GPT-4’s impact across diverse job roles, finding results consistent with established labor market research. It also offers a reusable tool for firms and policymakers to evaluate LLM intelligence and forecast job-level impacts. This enables informed decisions about where and how to integrate LLMs, match models to specific job requirements, and identify risks in socially intensive roles. The framework provides a foundation for responsible LLM deployment, ensuring alignment with human-centered structures and supporting strategic workforce planning.


  • Commercializing the Package Flow: Cross-Sampling Physical Products Through E-Commerce Warehouses

    by Brian Rongqing Han, Leon Yang Chu, Tianshu Sun, Lixia Wu

    Many e-commerce platforms have established their warehouses to facilitate the storage and delivery of packages. This paper studies a novel business practice—cross-sampling through e-commerce warehouses—that allows physical free samples provided by one (sampling) brand to be distributed with the packages of another unrelated (distributing) brand. In close collaboration with Alibaba, we implement cross-sampling through a large-scale field experiment, in which more than 55,000 free samples are distributed, to empirically examine its effectiveness in driving online sales of the sampling brand. First, we find significant increases in store visits and sales of the sampling brand both in the short term and long term up to 14 months afterward. The long-term effect is driven mainly by acquiring customers who did not purchase from the sampling brand to visit the online stores and eventually make a purchase. Second, contrary to the previous literature on within-category free sampling, we find a significant sales increase in both the sampled item and other items, which further spills over to indirect channels. Therefore, cross-sampling can also help promote the sampling brand as a whole. Finally, we illustrate the potential for personalization for cross-sampling. Cross-sampling is more effective for customers who recently viewed related products or just purchased nonessential products from the distributing brand. By taking into account the interaction among brands, items, and customers, we can further improve the profitability of cross-sampling by targeting the “right” packages. Overall, because cross-sampling is scalable, effective, and flexible, we demonstrate the high potential of a new business practice that combines offline logistics control and online information to generate additional business value for customers, brands, and the platform. This paper was accepted by Victor Martínez-de-Albéniz, operations management. Funding: T. Sun acknowledges research support from CKGSB Research Institute, Center for Digital Transformation, and ABRI Grant. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2020.00902 .


  • To Split or to Merge? How Partitioning Affects Consumption and Engagement with Digital Content

    by Heeseung Andrew Lee, Angela Aerry Choi, Wonseok Oh, Tianshu Sun

    This study reveals how partitioning serialized e-books into many short episodes (small partitioning (SP)) or fewer but lengthier segments (large partitioning (LP)) shapes reading behavior and economic outcomes. Using actual consumption data, we find that SP fosters higher total word consumption but can reduce overall progression through a title. Conversely, LP boosts completion rates, encourages deeper engagement through annotations and reviews, and spurs subsequent purchases. These positive effects are stronger when a book is popular or high in quality. Our results challenge the conventional view that shorter episodes automatically increase user attention, highlighting the need for a tailored approach. Stakeholders can implement LP for well-regarded or lengthy titles to maximize reader satisfaction and stimulate both reviews and further sales. Smaller partitions, however, might be beneficial for shorter works under a pay-per-word scheme or for certain audience preferences. By aligning partitioning strategies with content characteristics, publishers and platform owners can maintain robust consumer engagement, manage possible drawbacks of segmentation, and support sustainable growth. This research provides clear, actionable insights for digital publishing professionals seeking to optimize user experiences and drive profitability. These findings also encourage further inquiry into how partitioning strategies interact with rapidly evolving consumer preferences and reading technologies.


  • Democratizing Data Analytics Products to Drive SME Growth: A Natural Experiment

    by Zhe Yuan, Yitong Wang, Tianshu Sun, Huilan Xu

    Problem definition: Digital platforms provide sellers with the opportunity to leverage data analytics products, granting them convenient access to statistical analyses of their business data. To promote the adoption of data analytics products and facilitate the growth of small and medium enterprises (SMEs), a world-leading E-commerce platform initiated a “Democratizing Data Analytics” (DDA) campaign in May 2021, which gives free data analytics products access to all sellers on the platform. Leveraging a unique panel data set of over 370,000 sellers and a natural experiment, our study seeks to identify the causal effect of data analytics product adoption on seller performance. Methodology/results: This natural experiment led to a sharp increase in SMEs’ adoption of data analytics. Employing look-ahead propensity score matching (LA-PSM) coupled with difference-in-differences methods, we find that sellers who adopted an analytics product experienced a 14.4% increase in the number of transactions and an 18.8% increase in their revenue. We further explore the rich heterogeneity across sellers and find that younger, smaller sellers, and sellers with fewer cumulative transactions, benefit more from data analytics adoption, highlighting the value of data analytics product adoption for SME growth. Finally, we uncovered the intermediate process and mechanisms underlying such growth by examining how SMEs use data analytics and detailed operational decisions they make. After the adoption of data analytics products, sellers (1) spend more on advertising; (2) edit product titles, descriptions, and images more frequently; and (3) carry more new product stock keeping units and more product categories. Managerial implications: Our research indicates SMEs significantly benefit from adopting data analytics, underscoring the pivotal role these tools play in promoting their growth and development. Meanwhile, platforms or digital economies can increase their gross merchandise volume (GMV) by providing data analytics products to their sellers. Funding: This work was supported by funding from National Key Research and Development Project [2024YFB3312900], the National Natural Science Foundation of China [Grants 72141305, 72192803 and 72203202], Fundamental Research Funds for the Central Universities (Z. Yuan), and the CKGSB Research Institute (T. Sun). Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2024.1138 .


  • Top Cited Papers

    How Recommendation Affects Customer Search: A Field Experiment

    by Zhe Yuan, Aj Yuan Chen, Yitong Wang, Tianshu Sun

    The findings of this study have important implications for digital platform designers, managers, and regulators. First, the large-scale field experiment provides valuable insights into the relationship between product recommendation and consumer search under different scenarios. It highlights the importance of understanding consumer demand states and previous interests. Platforms can use these findings to customize product recommendations at an individual level and foster channel complementarity between recommendation and search. Second, the study emphasizes the need to consider channel spillovers. Optimizing recommender systems without considering the impact of channel interactions with search engines may lead to suboptimal results. Platforms should aim for a more coordinated integration of recommendation and search channels, as our conceptual framework illustrates how customers in different demand states can be influenced and served by both systems. Third, the findings offer insights into the potential impact of data regulations on e-commerce platforms. The study demonstrates that data regulations have a greater impact on the recommendation channel compared with the search channel. Platforms should find a balance between recommendation and search when facing stringent data regulations. They may strategically focus on the search channel to gather revealed customer interests, leading to a deeper integration of both channels.