Retail Data Research Articles

Did the COVID-19 pandemic permanently impact e-commerce in the US market?

The pandemic compelled many individuals, initially hesitant about online shopping, to overcome their reservations, acquire essential skills, and transition to online retail. This provided a natural experiment to assess whether the barriers to online shopping and the comfort of traditional in-store habits have hindered a broader shift to e-commerce. This paper uses the US retail data to analyze e-commerce activities before, during, and after the pandemic to determine whether the pandemic has permanently altered the pattern of the activities to determine whether the pandemic has permanently altered the pattern of the activities by utilizing structural break detection tools. Additionally, we carry out a forecasting exercise for post-pandemic based on pre-pandemic data to confirm our findings. Results suggest that while e-commerce activities surged during lockdown, they have predominantly reverted to pre-pandemic patterns. Our findings caution both investors and companies against overreaction in the face of exuberant changes in the market to avoid painful corrections afterward.

Urban spatial configuration and interactions with retail activities: An approach based on contact

This paper explores the relationship between urban spatial configuration and retail location. We present the Contact model to quantify the capacity of diverse urban configurations to foster economic interactions. Employing a network approach, the model simulates potential interactions between residents and retail activities as they move through the city. The model computes the number of shortest paths, weighted by resident density, in a network loaded with retail activities. The method allows for ranking different urban networks based on their potential to stimulate or constrain interactions between inhabitants and retail establishments. The weighted Contact model is tested across various metric radii in a sample of 35 US cities, utilizing population data from US Census Bureau (c2020) and retail data from the ReferenceUSA database (NAICS, c2020). To validate the model, we correlate results with total retail sales data for each city. Our findings reveal that cities with higher weighted Contact values exhibit greater retail sales volume, particularly within smaller radii up to 1000 m. The differences and relevance of this measurement decreases at >1000 m radius showing that the Contact model is relevant at the local level. We conclude that the urban spatial configuration remains to play an important role in the urban economic environment, holding implications for business and service planning, as well as urban planning and design.

Predictive Analytics for Financial Risk Management in Dynamic Markets

effective risk management is an indispensable requirement for improving the flow of transactions in dynamic financial markets. To this end, this study presents an applied predictive analytics methodology, that integrate gradient boosting algorithm to model the risk behavior in dynamic markets. This study, based on predictive analytics in monetary and financial systems, faces an urgent need for robust models that can overcome the uncertainties inherent in dynamic markets. Holistic experimentations on public case study of U.S retail data demonstrate the predictive power of the proposed approach of the state-of-the-art techniques across different performance metrics. This in turn highlights the nuanced interaction between variables and delivering intuitions into crucial risk determining factor.

Augmenting Retail Data With Open Data for Smarter Tourism Destinations

Augmenting Retail Data With Open Data for Smarter Tourism Destinations

TSUNAMI - an explainable PPM approach for customer churn prediction in evolving retail data environments

TSUNAMI - an explainable PPM approach for customer churn prediction in evolving retail data environments

Abstract PR003: An examination of tobacco and alcohol outlet density in Boston to identify priority areas for cancer prevention efforts using environmental justice mapping

Abstract Background. Inequitable access to health-harming substances in the built environment can create and exacerbate race and SES-based disparities in cancer incidence and mortality. Availability and density of tobacco and alcohol outlets is associated with increased experimentation, initiation, and regular use of alcohol and tobacco products. Prior studies examining the sale of tobacco and alcohol indicate that these outlets tend to be concentrated in communities of color and low-income areas. Massachusetts has highlighted environmental justice (EJ), one’s right to be protected from hazards and live in clean, health-promoting environments, as a priority issue and identified block group populations who are at most risk of environmental burdens. The objective of this study was to (a) examine the distribution of tobacco and alcohol outlets in environmental justice block groups in Boston, and (b) compare outlet density in more deprived neighborhoods to more privileged neighborhoods. Methods. Publicly available licensing lists for tobacco and alcohol retailers with geocoded data were obtained from the City of Boston’s open data hub. Lists were de-duplicated so that retailers that sell both tobacco and alcohol products were only represented once. The environmental justice block group layer (2020) was downloaded from Mass.gov. EJ block groups meet one or more of the following criteria: ‘high minority’ (>40% non-white population), ‘low-income’, or ‘high English isolation.’ ArcGIS Pro software was used to spatially join retailer data to Boston block groups and calculate tobacco and alcohol outlet density (the number of retailers per 1,000 people). Sociodemographic and cancer mortality data from the Health of Boston Report was used to classify neighborhoods as ‘more privileged’ (Back Bay, Beacon Hill, North End, West End) or ‘more deprived’ (Dorchester, Hyde Park, Mattapan, Roslindale, Roxbury). Results. Of the 560 Boston block groups analyzed, 458 (81.8%) were considered environmental justice block groups with 60% of EJ block groups classified as ‘high minority’. There were 931 retailers that sell tobacco or alcohol products for off-premises consumption in Boston, with 86% of these retailers (n=811) located in EJ block groups. On average, tobacco and alcohol outlet density was greater in EJ block groups compared to non-EJ block groups (2.09 vs 1.25 retailers per 1,000 residents). Nearly half (48%) of environmental justice block groups in Boston are located in the most deprived neighborhoods compared to only 3.7% in the most privileged neighborhoods. Tobacco outlet density is significantly higher in deprived neighborhoods compared to privileged neighborhoods (1.25 vs 0.98, p < 0.01). However, alcohol density is lower in more deprived block groups compared to more privileged ones (0.35 vs 0.74, p < 0.01). Conclusion. Environmental justice mapping can be used as a tool to identify areas where cancer prevention efforts may be enhanced by modifying the built environment to reduce clusters of alcohol and tobacco outlets. Citation Format: Lindsay Kephart. An examination of tobacco and alcohol outlet density in Boston to identify priority areas for cancer prevention efforts using environmental justice mapping [abstract]. In: Proceedings of the 16th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2023 Sep 29-Oct 2;Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2023;32(12 Suppl):Abstract nr PR003.

Reducing carbon footprint and promoting resource sustainability in the retail industry through the prevention of dead stocks

In the retail industry, majority of the traditional inventory management methods commonly focus on minimizing business costs but fail to consider implications on environmental sustainability. Existing inventory classification methods have failed to consider the interaction among the different performance indicators for a timely recognition of the potential dead stocks and dead stocks in the inventory. This study aims to identify and effectively classify potential dead stocks and dead stocks through an enhanced inventory classification method that integrates the remaining product shelf life as an indicator for timely recognition of dead stocks. The study also aims to provide guidelines in setting up decision rules needed in objectively classifying inventory based on each performance indicator through an automated algorithm with decision rules. Using data sets taken from open databases that provide real-world data, results show that the automated algorithm can effectively detect the presence of dead stocks and distinguish such inventory from potential dead stocks in the system which existing inventory classification methods are unable to do. One-Way ANOVA Tests performed showed that the proposed classification method can consistently classify inventory across different retail data sets while Monte Carlo simulation was used to simulate the amount of inventory that becomes dead stocks in a multi-period inventory management system. Results show that 17% of these potential dead stocks, if undetected, become dead stocks in the following period. Over a ton of carbon emissions is prevented with lesser dead stock waste disposal. The value of this study lies in the fact that classifying inventory to detect the presence of potential dead stocks is an essential first step in solving the dead stock problem not just for business profitability but also for sustainable resource consumption and a sustainable environment.

BMI and the Food Retail Environment in Melbourne, Australia: Associations and Temporal Trends.

Research into the link between food environments and health is scarce. Research in this field has progressed, and new comprehensive methods (i.e., incorporating all food retail outlets) for classifying food retail environments have been developed and are yet to be examined alongside measures of obesity. In this study, we examine the association and temporal trends between the food environment and BMI of a repeated cross-sectional sample of the adult population between 2008 and 2016. Methods: Food retail data for 264 postal areas of Greater Melbourne was collected for the years 2008, 2012, 2014, and 2016, and a container-based approach was used to estimate accessibility to supermarkets, healthy and unhealthy outlets. Data on BMI for postal areas was obtained from the Victorian Population Health Survey (n = 47,245). We estimated the association between the food environment and BMI using linear mixed models. Results indicated that BMI increased as accessibility to healthy outlets decreased by up to -0.69 kg/m2 (95%CI: -0.95, -0.44). BMI was lower with high and moderate access to supermarkets compared to low access by -0.33 kg/m2 (-0.63, -0.04) and -0.32 kg/m2 (-0.56, -0.07), and with high access to unhealthy outlets compared to low access (-0.38 kg/m2: -0.64, -0.12) and moderate access (-0.54 kg/m2: -0.78, -0.30). Conclusion: Our results show that increasing access and availability to a diverse range of food outlets, particularly healthy food outlets, should be an important consideration for efforts to support good health. This research provides evidence that Australia needs to follow suit with other countries that have adopted policies giving local governments the power to encourage healthier food environments.

Estimating linear mixed effect models with non-normal random effects through saddlepoint approximation and its application in retail pricing analytics

ABSTRACT Linear Mixed Effects (LME) models are powerful statistical tools that have been employed in many different real-world applications such as retail data analytics, marketing measurement, and medical research. Statistical inference is often conducted via maximum likelihood estimation with Normality assumptions on the random effects. Nevertheless, for many applications in the retail industry, it is often necessary to consider non-Normal distributions on the random effects when considering the unknown parameters' business interpretations. Motivated by this need, a linear mixed effects model with possibly non-Normal distribution is studied in this research. We propose a general estimating framework based on a saddlepoint approximation (SA) of the probability density function of the dependent variable, which leads to constrained nonlinear optimization problems. The classical LME model with Normality assumption can then be viewed as a special case under the proposed general SA framework. Compared with the existing approach, the proposed method enhances the real-world interpretability of the estimates with satisfactory model fits.

Recency, Frequency, Monetary Value, Clustering, and Internal and External Indices for Customer Segmentation from Retail Data

While there are several ways to identify customer behaviors, few extract this value from information already in a database, much less extract relevant characteristics. This paper presents the development of a prototype using the recency, frequency, and monetary attributes for customer segmentation of a retail database. For this purpose, the standard K-means, K-medoids, and MiniBatch K-means were evaluated. The standard K-means clustering algorithm was more appropriate for data clustering than other algorithms as it remained stable until solutions with six clusters. The evaluation of the clusters’ quality was obtained through the internal validation indexes Silhouette, Calinski Harabasz, and Davies Bouldin. When consensus was not obtained, three external validation indexes were applied: global stability, stability per cluster, and segment-level stability across solutions. Six customer segments were obtained, identified by their unique behavior: lost customers, disinterested customers, recent customers, less recent customers, loyal customers, and best customers. Their behavior was evidenced and analyzed, indicating trends and preferences. The proposed method combining recency, frequency, monetary value (RFM), K-means clustering, internal indices, and external indices achieved return rates of 17.50%, indicating acceptable selectivity of the customers.

Unveiling Patterns and Insights in a Retail Dataset: A Data Analytics Approach

Abstract: Data analytics plays a crucial role in extracting valuable information from extensive sets of data, empowering companies to make informed choices based on data. This research paper presents a thorough examination of retail data collection using diverse data analytics approaches. The goal is to discover significant patterns, tendencies, and valuable insights that can enhance business strategies and enhance customer contentment. The paper outlines the characteristics of the data set, explores the employed data analytics techniques, showcases the outcomes, and emphasizes the consequences and potential uses of the discoveries.

Data Warehouse, Detection and Transfer of Anomalies in Retail Data

In this article, we offer some suggestions for anomaly detection on the data received from the source to the Data warehouse. As a result, it is aimed to prevent the entry of dirty and noisy data into the data warehouse. We think that knowing that there is clean and healthy data in the data warehouse will be resistant to anomalies in the processed data used for data science. In order to reach our goal, studies were carried out on the data in the retail sector. We aimed to determine our theoretical thoughts from some topics such as user erroneous login data in the retail and energy industry, abnormal sales over employees during the campaign period, product stock abnormality, and incorrect pricing. When we examined many studies, we saw that they made anomaly detection after estimation. Before taking the data from the source to the data warehouse, we thought that anomaly detection would be more efficient and healthier. Analysis and results were evaluated on the data obtained in the wiseboard retail project of Gtech company.

Agri-food importing firms amid a global health crisis

This paper exploits daily customs transaction data on the universe of Swiss agri-food importing firms to assess the response of firms to a global shock. Estimating a linear model that regresses product-level import margins on daily COVID19 shocks and a host of fixed effects, we find that the pandemic had a substantial trade-reducing effect on imports. The trade effects were driven mainly by a reduction in the number of importing firms (i.e., 63% of the total effect), and much less by the number of products imported and the average import value per product per firm. We explore several sources of heterogeneity and show, among others, that larger and incumbent firms were affected more by the trade adjustments. Our results also reveal that the relative contribution of each import margin to the decline in aggregate imports depends on the level of data aggregation (i.e., daily, weekly or monthly). Finally, we validate and confirm our main findings by testing two mechanisms: (i) third-country supply-side effects using insights from structural gravity models and (ii) changes to consumer demand using consumer mobility, and retailer and consumer scanner data.

Medical Cannabis Program Sustainability in the Era of Recreational Cannabis

Since October 2022, a total of 21 states have enacted both medical-use and adult-use cannabis legalization, each with their own unique set of laws, regulations, implementation, structures, and enforcement ("policies"). Unlike adult-use programs, medical-use programs often represent a safer and affordable option for patients with diverse needs; however, current evidence suggests that medical-use program activity decreases after implementation of adult-use retail. The current study compares medical patient registration data and medical- and adult-use retail data from 3 distinct medical- and adult-use states (Colorado, Massachusetts, and Oregon) in the time after adult-use retail implementation in each state. To investigate changes in medical cannabis programs with simultaneous adult-use legalization, correlation and linear regression analyses were used to assess outcome measures: (1) medical-use retail sales; (2) adult-use retail sales; and (3) number of registered medical patients in all fiscal quarters after adult-use retail sales were implemented in each state to September 2022. Adult-use cannabis sales increased significantly over time in all 3 states. However, both medical-use sales and number of medical patients registered in the states increased only in Massachusetts. Results indicate that states' preexisting medical-use programs may undergo critical changes after adult-use cannabis legalization is enacted and implemented. Key policy and program differences, such as regulatory differences in the implementation of adult-use retail sales, may have differential impacts on medical-use programs. For continued patient access, it is critical that future research assess the differences within and between states' medical-use and adult-use programs that permit sustainability of medical-use programs alongside adult-use legalization and implementation.

High-frequency retail data: the interests of the state, enterprises and scientific organizations

The rapid development of technologies for collecting and analyzing big data, including those characterizing trade, is currently taking place. This data, with a high degree of detail, takes into account the whole variety of consumer decisions, which allows them to develop key management proposals on what, where and when to produce and sell. Banks, retail chains, and the state are actively interested in these data. At the same time, individual small and medium-sized enterprises weak use of big data in their activities. The purpose of this study is to highlight the problems and prospects for their application for management purposes, based on an analysis of the current practice of using high-frequency retail data. As a result of the study, the features of the available data of retail companies, payment systems and OFDs, which are manifested in their different structure and limitations for use in the development of management decisions, are highlighted. It is shown that big data characterizing retail trade is available to a narrow circle of people who, as a rule, have their own interests, which are not yet consistent with the idea of open publication of these data, even for scientific purposes. There are very few research publications based on high-frequency fiscal data. Such closeness of data does not create prerequisites for the active development of skills in working with them in most enterprises and organizations, which determines the weak use of microdata for management purposes.

Strategic Visual Merchandising of New and Open-Box Products: Evidence from Experiment and Retail Data

Retailers are increasingly selling returned products as open-box along with their new product offerings. Whereas some retailers position open-box products side-by-side with their new counterparts in the assortment (i.e., side-by-side strategy), other retailers position them separately in a different part of the retail space/website (i.e., separate strategy). We conduct multimethodology research to investigate the effectiveness of these two visual merchandising strategies. First, using a choice-based conjoint design in a controlled behavioral experiment, we identify how the two strategies affect the perceived value of new products. Next, using proprietary data from a retailer that has implemented both strategies across different stores, we empirically test the impact of each strategy on new product sales and returns under real market conditions. Finally, we extend our empirical analysis to prescribe how retailers can identify the right visual merchandising strategy for a given product profile. Overall, our results reveal a novel trade-off between the two visual merchandising strategies. With respect to retail revenue, the separate strategy outperforms the side-by-side strategy as we find that new product sales are higher with the separate strategy. However, with respect to retail cost, the side-by-side strategy outperforms the separate strategy as we find that new product returns are lower with the side-by-side strategy. Our extended analysis reveals that this tradeoff varies by product profile. Using counterfactual analysis, we further demonstrate that customizing the visual merchandising strategy by product mix could increase profit by 3.2% at the focal retailer. This paper was accepted by Jeannette Song, operations management. Supplemental Material: The data files and online appendices are available at https://doi.org/10.1287/mnsc.2023.4786 .

Decision support for managing assortments, shelf space, and replenishment in retail

Efficiently managing retail space is critical as the increase in product variety is in conflict with limited shelf space and instore replenishment constraints. This paper develops a general framework for retail space management and presents a decision support model with the related problems within the framework of optimizing assortment, shelf space assignment and replenishment. An integrative approach to these planning problems becomes particularly relevant for fast-moving consumer goods and groceries, where stores are regularly replenished from distribution centers. The planning problem at hand is a multi-product shelf space allocation problem where demand is a composite function of the shelf space allocated and assortment-related demand substitution, and actual replenishment practices from retail are incorporated. The model developed extends existing models of shelf space management by jointly considering space-elastic demand and assortment-based substitution and integrating restocking constraints. For the latter, we consider real-world replenishment processes of retailers that distinguish between period-based and ad-hoc replenishment from the backroom. We develop three solution approaches that are based on efficient pre-processing and a nonlinear binary integer programming formulation of the problem. The computation tests based on retail data show the efficiency of the solution approaches in terms of computation time and solution quality. We reveal the improvement in profit levels that can be achieved from integrating assortments, shelf space planning and replenishment where challenges arise in obtaining feasible solutions with limited shelf space and replenishment constraints. We also use sensitivity analyses to demonstrate the high impact of replenishment constraints on profits and solution structures.

Sales of Electronic Nicotine Delivery Systems (ENDS) and Cigarette Sales in the USA: A Trend Break Analysis

Electronic nicotine delivery systems (ENDS) are a potentially lower-risk tobacco product which could help smokers switch completely away from cigarettes. However, the lack of strong evidence to date of a measurable, population-level effect on reducing smoking has generated skepticism about ENDS’ potential benefits. This study examines whether increased US ENDS sales have been associated with reduced cigarette sales. Retail data on weekly per-capita cigarette and ENDS purchases in the USA during 2014–2019 were obtained from a national sample of brick-and-mortar retail outlets. Trends in cigarette sales were modeled before (2014–2016) ENDS had a substantial market share, and, after adjusting for macroeconomic factors, projected into the post-period (2017–2019). Actual cigarette sales were lower than projected sales (by up to 16% across the post-period), indicating a substantial “cigarette shortfall” in the post-period. To explore whether general (i.e., inclusive of potentially many mechanisms) substitution by ENDS can explain the cigarette shortfall, its association with per-capita ENDS volume sales was examined. Higher ENDS sales were significantly associated with a greater cigarette shortfall: for every additional per-capita ENDS unit, cigarette sales were 1.4 packs-per-capita lower than expected (B = 1.4, p < .0001). Error correction models which account for spurious correlation yielded similar results. These findings support ENDS serving as a substitute for cigarettes (through potentially many mechanisms including cigarette price), causing cigarette consumption to decline. Since ENDS potentially pose lower risk than cigarettes, this general substitution effect suggests that risk-proportionate tobacco regulation could mitigate the tobacco-related health burden.

Strategic Visual Merchandising of New and Open-box Products: Evidence from Experiment and Retail Data

Strategic Visual Merchandising of New and Open-box Products: Evidence from Experiment and Retail Data

Accuracy Differences in Cannabis Retailer Information Ascertained from Webservices and Government-Maintained State Registries Across US States Legalizing the Sale of Cannabis in 2019.

Cannabis retailer locations used to investigate geographic cannabis access are frequently ascertained from two sources: 1) webservices which provide locations of cannabis retailers (e.g., Yelp) or 2) government-maintained registries. Characterizing the operating status and location information accuracy of cannabis retailer data sources on a state-by-state level can inform research examining the health implications of cannabis legalization policies. This study ascertained cannabis retailer name and location from webservices and government-maintained registries for 26 states and the District of Columbia legalizing cannabis sales in 2019. Validation subsamples were created using state-level sequential sampling. Phone surveys were conducted by trained researchers for webservice samples (n=790, November 2019 - May 2020) and government-maintained registry (n=859, February - June 2020) to ascertain information about operating status and location. Accuracy was calculated as the percent agreement among subsample and phone survey data. For operating status and location, webservice derived data was 78% (614/790) and 79% (484/611) accurate, whereas government-maintained registry derived data was 76% (657/859) and 95% (622/655) accurate, respectively. Fifty-nine percent (15/27) of states and the District of Columbia had over 80% accuracy for operating status and 48% (13/27) states had over 80% accuracy for location information with both data sources. However, government-maintained registry derived information was more accurate in 33% (9/27) states for operating status and 41% (11/27) states for location information. Both data sources had similar operating status accuracy. Research using spatial analysis may prefer government-maintained registry derived data due to high location information accuracy, whereas studies looking at broad trends across states may prefer webservice derived. State level COVID-19 restrictions had minimal impact on ascertainment of cannabis retailer operating status and location information via phone survey derived from webservices and government-maintained registries.