FCFelipe Caro

Felipe Caro

*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


Felipe Caro has a total of 32 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Webrooming in Retail: Evidence from a Field Experiment

    by Felipe Caro, Victor Martinez-De-Albeniz, Pablo Sarmiento, Felipe Vega

    Webrooming, the practice of searching for products online before purchasing in a physical store, is widespread yet we do not know whether it causally increases in-store spending or merely reflects consumers' underlying purchase intentions. We study the causal effect of webrooming using a randomized field experiment with a leading do-it-yourself (DIY) retailer in Chile. The experiment exposed 1.3 million online shoppers to alternative product recommendation policies, with more than 97,000 identifiable customers tracked across channels. Using randomized recommendation assignment as an instrument for online browsing, we estimate that each additional product browsed inflates in-store spending in that category by about 200 times compared to the baseline. The effect peaks in the following week and becomes statistically insignificant by the third week. We find no statistically significant reduction in concurrent online sales. Heterogeneity analyses suggest stronger responses among professional customers, new online customers, and categories with lower online purchase shares. These findings establish that online product exploration is a valuable driver of subsequent offline revenue and contradict the conventional emphasis on minimizing the path to online checkout. Put differently, for functional products such as DIY, stimulating unplanned product discovery clicks can be profitable, not because of additional online sales, but rather increased offline purchases.

  • Open Access

    Do You Have My Size In Stock? Assortment and Inventory Optimization Under the Consider-Fit-Then-Choose Choice Model

    by Yi-Chun Akchen, Kaan Cakiroglu, Felipe Caro, Jacob Feldman

    In apparel retail and other applications, when a customer's preferred size is unavailable, demand may shift to nearby sizes. This substitution creates new assortment and inventory optimization challenges because product availability---and, consequently, demand---becomes coupled across sizes. To address these challenges, we introduce the consider-fit-then-choose (CFTC) model, a simple yet rich framework for capturing size-dependent choice behavior. The model applies to settings in which each product may be offered in multiple sizes and the set of available sizes directly affects both customer preferences and consideration sets through a notion of fit, measured by the distance between an available size and the customer's ideal size. Under this model, we study both assortment optimization and the show-all inventory selection problem, in which the retailer chooses initial inventory levels before the selling horizon and subsequently offers every in-stock product to each arriving customer. We first show that assortment optimization under the CFTC model is NP-hard and develop a polynomial-time approximation scheme (PTAS) when customers are willing to deviate by at most $O(1)$ sizes from their ideal size. Combined with the recent black-box framework of Fu et al. (2026), this result yields a nearly 0.272-approximation for the corresponding show-all inventory selection problem. We then ask whether stronger guarantees can be obtained by exploiting the specific choice dynamics of the CFTC model. Our approach to both the fluid and stochastic versions of the inventory problem begins by stocking only every other size, thereby decoupling demand across the stocked sizes. Under this restriction, the CFTC model reduces to a special class of mixed multinomial logit models, which we prove satisfies the convex chain decomposition (CCD) property introduced by Goyal et al. (2023). This property allows us to translate solutions to a classical choice-based linear program into feasible show-all inventory policies. Building on this structure, for the fluid problem we develop a polynomial-time (1/2-ε)-approximation algorithm under adjacent-size substitution and a mild condition on the preference weights. For the stochastic problem, we establish an asymptotic 1/2-approximation guarantee using a new coupling argument that connects the stochastic inventory process to its fluid counterpart. Numerical experiments on instances calibrated using footwear data demonstrate that our inventory solutions achieve small optimality gaps across a broad range of substitution patterns and problem settings.

  • On Size Substitution and Its Role in Assortment and Inventory Planning

    by Yi-Chun Akchen, Felipe Caro

    Problem definition: How should (apparel) retailers manage product sizes? For example, if most customers wearing a given shoe size, such as 9.5, are willing to accept a half-size up or down, is it necessary for a retailer to carry that size at all? Additionally, although identical products in different sizes are treated as distinct stock-keeping units in inventory management, they are often aggregated for assortment and strategic planning. However, there is no theoretical justification for this approach. In this paper, we address the fundamental questions about size management that have remained largely unexplored in the operations literature. Methodology/results: We propose a choice model where each customer forms a consideration set based on the in-stock availability of products of her best-fit size and adjacent sizes. Using a real-world data set from a large footwear retailer, we show that nearly 25% of the unmet demand caused by stockouts spills over to adjacent sizes. We further solve the assortment and inventory optimization problems under the proposed choice model. Our findings demonstrate that the optimal assortment remains unchanged, regardless of the likelihood that customers might purchase adjacent sizes. We utilize this finding and further show that inventory policies that ignore size substitution can be (asymptotically) optimal when the demand rate is high or the selling horizon is long. We also propose a mixed-integer program to determine inventory levels that account for size substitution and achieve higher profits in low-demand settings. Managerial implications: We show that the prevalent size-aggregation approach employed in apparel retail operations is sensible in high-demand settings, such as e-commerce. In contrast, when the expected demand over the selling horizon is low, size substitution can be relevant and should be considered in stocking decisions. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2023.0674 .


  • Top Cited Papers

    Introduction to the Special Issue on the Human-Algorithm Connection

    by Felipe Caro, Jean-Edouard Colliard, Elena Katok, Axel Ockenfels, Nicolas Stier-Moses, Catherine Tucker, D. J. Wu


  • The Value of Online Interactions for Store Execution

    by Felipe Caro, Victor Martínez-De-Albéniz, Borja Apaolaza

    Problem definition: Omnichannel retailers interact with customers both online and offline. So far, they have used the richer information available—gathered from customer interactions across digital and physical channels—to optimize the sales process by designing the right channel and supply chain structures and by personalizing offers, pricing, and promotions. We advance an additional dimension of omnichannel value: retailers can use online clickstreams to better understand customer needs and optimize store layouts to maximize webrooming conversion, which we define as the ratio of sales to webrooming activity. Methodology/results: We develop a model in which in-store purchases depend on the customer’s shopping list and the effort required to locate and reach the products within the store. Category location in the store thus drives the likelihood of a sale. We then apply our model to a large home improvement retailer and find that shoppers’ preferences are revealed by nearby online traffic and hard-to-reach locations lead to lower webrooming conversion. Finally, we optimize category location assignments using our demand model and find that putting higher-interest and higher-price items in the most effective locations can increase revenues by about 2%–5% in comparison with models that ignore online clicks. Managerial implications: We show how using online clickstream information for optimizing offline operations can create significant value. More fundamentally, our results provide a word of caution that in some retailing segments such as home improvement, longer in-store paths might not necessarily be better. Funding: V. Martínez-de-Albéniz acknowledges the financial support provided by the Agencia Estatal de Investigación (AEI) from the Spanish Ministry of Science and Innovation [Grant PID2020-116135GB-I00 MCIN/AEI/10.13039/501100011033]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2024.1290 .


  • Top Cited Papers

    Believing in Analytics: Managers’ Adherence to Price Recommendations from a DSS

    by Felipe Caro, Anna Sáez de Tejada Cuenca

    Problem definition: We study the adherence to the recommendations of a decision support system (DSS) for clearance markdowns at Zara, the Spanish fast fashion retailer. Our focus is on behavioral drivers of the decision to deviate from the recommendation, and the magnitude of the deviation when it occurs. Academic/practical relevance: A major obstacle in the implementation of prescriptive analytics is users’ lack of trust in the tool, which leads to status quo bias. Understanding the behavioral aspects of managers’ usage of these tools, as well as the specific biases that affect managers in revenue management contexts, is paramount for a successful rollout. Methodology: We use data collected by Zara during seven clearance sales campaigns to analyze the drivers of managers’ adherence to the DSS. Results: Adherence to the DSS’s recommendations was higher, and deviations were smaller, when the products were predicted to run out before the end of the campaign, consistent with the fact that inventory and sales were more salient to managers than revenue. When there was a higher number of prices to set, managers of Zara’s own stores were more likely to deviate from the DSS’s recommendations, whereas franchise managers did the opposite and showed a weak tendency to adhere more often instead. Two interventions aimed at shifting salience from inventory and sales to revenue helped increase adherence and overall revenue. Managerial implications: Our findings provide insights on how to increase voluntary adherence that can be used in any context in which a company wants an analytical tool to be adopted organically by its users. We also shed light on two common biases that can affect managers in a revenue management context, namely salience of inventory and sales, and cognitive workload. Supplemental Material: The e-companion is available at https://doi.org/10.1287/msom.2022.1166 .


  • Call for Papers—Management Science Special Issue on the Human-Algorithm Connection

    by Felipe Caro, Jean-Edouard Colliard, Elena Katok, Axel Ockenfels, Nicolas Stier-Moses, Catherine Tucker, D. J. Wu


  • Can Brands Claim Ignorance? Unauthorized Subcontracting in Apparel Supply Chains

    by Felipe Caro, Leonard Lane, Anna Sáez de Tejada Cuenca

    Unauthorized subcontracting—when suppliers outsource part of their production to a third party without the retailer’s consent—has been common practice in the apparel industry and is often tied to noncompliant working conditions. Because retailers are unaware of the third party, the production process becomes obscure and cannot be tracked. In this paper, we present an empirical study of the factors that can lead suppliers to engage in unauthorized subcontracting. We use data provided by a global supply chain manager with more than 30,000 orders, of which 36% were subcontracted without authorization. We find that the frequency of unauthorized subcontracting across factories has a pronounced bimodal distribution. Moreover, the degree of unauthorized subcontracting in the past is highly related to the probability of engaging in unauthorized subcontracting in the future, which suggests that factories behave as if they choose a strategic level of unauthorized subcontracting. At the order level, we find that state dependence (i.e., the status of an order carrying over to the next one) and price pressure are the key drivers of unauthorized subcontracting. Buyer reputation and lead time also play a role. Finally, we show that unauthorized subcontracting can be predicted correctly for more than 80% of the orders in out-of-sample tests and for about 70% of suppliers. This indicates that retailers can use business analytics to predict unauthorized subcontracting and help prevent it.This paper was accepted by Vishal Gaur, operations management.


  • Managing Online Content to Build a Follower Base: Model and Applications

    by Felipe Caro, Victor Martínez-De-Albéniz

    Content providers manage their production by regulating the pace at which content is created and released. They have two types of consumers: existing ones, or followers; and new visitors who may become followers. To maximize their effectiveness, providers must consider the direct, short-term effect of content and also its indirect, long-term effect on the retention and expansion of their follower base. We develop a simple model to study the dynamics of building up a follower base and then combine that model with stochastic dynamic programming to optimize the content provider’s profit. We find that optimal content policies exhibit trajectories whereby both content and followers increase until reaching a steady state at which revenues equal marginal costs. It is therefore preferable to start slow and then accelerate only after successful content has increased the follower base; thus past success makes the content provider work harder. We apply our model to bloggers by analyzing the traffic from new and returning visitors to several blogs. We find that optimized posting activity yields significantly higher profits than does a steady or seasonal content strategy. Our model contributes to the growing literature on data-driven optimization and prescriptive analytics, specifically for content management.


  • Top Cited Papers

    The Future of Retail Operations

    by Felipe Caro, A. Gürhan Kök, Victor Martínez-De-Albéniz

    Retailing consists of all the activities associated with the selling of goods to the final consumer. In this article, we review the research on retail operations published in Manufacturing & Service Operations Research (M&SOM) since 1999. We then discuss the current retail landscape and the new research directions it offers, in which M&SOM can play a prominent role.