Purchase Probability Research Articles

Shrinking the Upper Confidence Bound: A Dynamic Product Selection Problem for Urban Warehouses

The recent rising popularity of ultrafast delivery services on retail platforms fuels the increasing use of urban warehouses, whose proximity to customers makes fast deliveries viable. The space limit in urban warehouses poses a problem for such online retailers: the number of stock keeping units (SKUs) they carry is no longer “the more, the better,” yet it can still be significantly large, reaching hundreds or thousands in a product category. In this paper, we study algorithms for dynamically selecting a large number of products (i.e., SKUs) with top customer purchase probabilities on the fly, from an ocean of potential products to offer on retailers’ ultrafast delivery platforms. We distill the product selection problem into a semibandit model with linear generalization. There are in total N arms corresponding to N products, each with a feature vector of dimension d. The player pulls K arms in each period and observes the bandit feedback from each of the pulled arms. We focus on the setting where K is much greater than the number of total time periods T or the dimension of product features d. We first analyze a standard Upper Confidence Bound (UCB) algorithm and show its regret bound can be expressed as the sum of a T-independent part and a T-dependent part, which we refer to as “fixed cost” and “variable cost,” respectively. To reduce the fixed cost for large K values, we propose a novel online learning algorithm, which iteratively shrinks the upper confidence bounds within each period, and show its fixed cost is reduced by a factor of d. Moreover, we test the algorithms on an industrial data set from Alibaba Group. Experimental results show that our new algorithm reduces the total regret of the standard UCB algorithm by at least 10%. This paper was accepted by J. George Shanthikumar, big data analytics.

Does Authenticity of Influencers Matter? Examining the Impact on Purchase Decisions

Social media has brought to light the role of authenticity in influencing decision-making, but little work has examined specific authenticity cues that might impact purchase. In other words, does authenticity influence consumer actions? In this paper, we examine an e-commerce platform that leverages influencers to sell beauty products, and examine how two authenticity cues – explicit transparency (#ad) and implicit authenticity (photo originality) – impact purchase decisions. We draw from expectation-confirmation theory and suggest that the impact that perceived authenticity has on purchase decisions and post-purchase satisfaction depends on the type of authenticity cue and is shaped by the expectation of the consumer. We further argue that expectations vary depending on the consumer experience with an influencer and therefore explore the moderation effect of customer loyalty. Our findings suggest that consumers have different reactions depending on the authenticity cues – their purchase probability decreases with transparency but increases with photo originality. We see that with greater loyalty, consumers are more negatively impacted by the #ad tag and are less positively influenced by the originality of the photo. On the other hand, purchase depth—loyalty to one influencer—mitigates the negative effect of the #ad tag and heightens the positive impact of photo originality. We suggest that experience with the influencer provides evidence that the higher the expectation for authenticity (depth) the higher the negative impact of in-authenticity. We also discuss limitations and implications of our work.

When Advertising meets Assortment Planning: Joint Advertising and Assortment Optimization under Multinomial Logit Model

Although assortment optimization has been extensively studied in the literature, there is lack of studies on exploring the impact of advertising on the output of assortment planning. In this article, we fill this gap by considering a novel joint advertising budget strategy and assortment optimization problem, where the platforms must jointly decide (1) the optimal advertising effort allocated to each product and (2) an assortment of products displayed to consumers. We assume that the purchase behavior of a consumer is governed by the classic multinomial logit choice model, where each product is associated with a preference weight and the purchase probability of a product in an assortment is proportional to its preference weight. To capture the advertising effect in the context of assortment planning, we assume that one can increase the preference weight of a product through advertising that product, and the degree of improvement is decided by a response function in terms of the amount of advertising efforts allocated to that product. Given a budget constraint on advertising, our objective is to find a solution, which is composed of an advertising strategy and an assortment of products, to maximize the expected revenue. We analyze the structural properties of this problem and derive effective solutions under different settings. If there is no capacity constraint on the number of products displayed to consumers, we show that the revenue-ordered assortments still maintain optimality, and we leverage this result to derive an optimal solution. For the capacity constrained case, it is hard to solve the optimization problem directly, we overcome this challenge by proposing an efficient iterative scheme to find the optimal solution.

Factors influencing online and in-store purchase of telecommunications services and equipment in Croatia

The purpose of this paper is to examine differences in factors affecting the probability of online and in-store purchase of telecommunications equipment in the Republic of Croatia. Online questionnaire survey conducted in September 2019 gathered responses that identified respondents’: method of telecommunications services and equipment purchase, their information gathering tools, factors affecting respondents’ telecommunications services and equipment purchasing decisions in general, and online and in-store purchasing decisions. 125 gathered responses were analysed using a binomial logistic regression. The results state that general factors incite respondents to purchase telecommunications equipment and services instore. Factors inciting online purchases are enabling factors and cause respondents to purchase telecommunications services and equipment online. Responses were further divided into subsamples depending on gender and education level. Results are consistent in tested subsamples. In the sample of male respondents, the results are consistent, while the female subsample shows does not exhibit statistically significant results. Furthermore, subsamples that allowed education level differences showed the results are supported for the subsample of respondents with higher education, and are not supported for respondents with secondary education.

Correction to: Which Privacy Policy Works, Privacy Assurance or Personalization Declaration? An Investigation of Privacy Policies and Privacy Concerns

This study focuses on two specific privacy policies, namely privacy assurance and personalization declaration. Specifically, we investigate how these distinct privacy policies affect customers’ privacy concerns and subsequent purchase responses. We have developed a conceptual model that addresses the independent effects of privacy assurance and personalization declaration, as well as the mechanism (i.e., privacy concerns) of these effects. Our model is grounded in motivation theory and supported by a field experiment and a controlled experiment. Our study demonstrates that privacy assurance that claims security protection negatively affects customers’ purchase probability and purchase amount. By contrast, personalization declaration that addresses personalized benefits positively affects customers’ purchase probability and purchase amount. Privacy concerns significantly mediate the negative effects of privacy assurance on purchase responses and the positive effects of personalization declaration on purchase responses. Overall, our findings inform managers of how to deploy privacy policies to reduce customers’ privacy concerns and boost purchase responses.

Effect of air pollution on household insurance purchases. Evidence from China household finance survey data.

In recent years, the health and economic effects of air pollution have attracted considerable attention, and health and insurance services have been closely related to residents’ welfare. However, there are few studies on the influence of pollution on household purchases of insurance. Using data from the 2013 and 2015 China Household Finance Surveys, this study investigates the effect of air pollution on insurance purchases using Logit and Poisson regression models. It is found that air pollution significantly increases the probability of household insurance purchases and the level of premium expenditure, although the impact of air pollution on insurance purchases shows a degree of heterogeneity. Health insurance is more sensitive to air pollution than life insurance and other types of insurance. In areas where NO2 and O3 are the main types of pollutants, air pollution has a greater impact on household insurance purchases.

Logistics Performance, Ratings, and its impact on Customer Purchasing Behavior and Sales in E-commerce Platforms

Problem Definition: We examine the impact of logistics performance metrics such as delivery time, and customer’s requested delivery speed on logistics service ratings and third-party sellers’ sales on an e-commerce platform. Academic/Practical Relevance: While e-commerce retailers like Amazon have recently invested heavily in their logistics networks to provide faster delivery to customers, there is scant academic literature that tests and quantifies the premise that convenient and fast delivery will drive sales. In this paper, we provide empirical evidence on whether this relationship holds in practice by analyzing a mechanism that connects delivery performance to sales through logistics ratings. Prior academic work on online ratings in ecommerce platforms has mostly analyzed customers’ response to product functional performance and biases that exist within. Our study contributes to this stream of literature by examining customer experience from a service quality perspective by analyzing logistics service performance, logistics ratings, and its impact on customer purchase probability and sales. Methodology: Using an extensive data set of more than 15 million customer orders on the Tmall platform and Cainiao network (logistics arm of Alibaba), we employ the Heckman ordered regression model to explain the variation in customers’ rating of logistics performance, as well as the likelihood of customers posting a logistics rating. Next, we develop a generic customer choice model that links the customer’s likelihood of making a purchase to the logistics ratings provided by prior customers. We implement a two-step estimation of the choice model to quantify the impact of logistics ratings on customer purchase probability and third-party seller sales. Results: We surprisingly find that even customers with no promise on delivery speed are likely to post lower logistics ratings for delivery times longer than two days. Although these customers are not promised an explicit delivery deadline, they seem to have a mental threshold of two days and expect deliveries to be made within that time. Similarly, we find that priority customers (those with two-day and one-day promise speed) provide lower logistics ratings for delivery times longer than their anticipated delivery date. We estimate that reducing the delivery time of all three-day delivered orders on this platform (which makeup approximately 35% of the total orders) to two days would improve the average daily third-party seller sales by 13.3% on this platform. The impact of delivery time performance on sales is more significant for sellers with a higher percentage of three-day delivered orders and a higher spend per order. Managerial Implications: Our study emphasizes that delivery performance and logistics ratings, which measure service quality, are essential drivers of the customer purchase decision on e-commerce platforms. Further, by quantifying the impact of delivery time performance on sales, our study also provides a framework for online retailers to assess if the increase in sales due to improved logistics performance can offset the increase in additional infrastructure costs required for faster deliveries. Our study’s insights are relevant to third-party sellers and e-commerce platform managers who aim to improve long-term online customer traffic and sales.

Hyper-heuristic for integrated due-window scheduling and vehicle routing problem for perishable products considering production quality

ABSTRACT In today’s competitive environment, industrial units are seeking to reduce costs and increase customer numbers; if their business involves perishable products, achieving these goals is even more important. By integrating scheduling and vehicle routing problems for perishable products, this study tries to minimize costs and maximize customers’ purchase probability. In the scheduling stage, a flexible flowshop scheduling problem considering production quality is studied. After completion of the last job, the distribution stage begins and each product must be delivered in its time-window. This problem is formulated as mixed-integer linear programming and solved by the GAMS solver. Owing to the NP hardness of this problem, a hyper-heuristic is proposed to solve large-size instances. In the proposed algorithm, the acceptance of the solution is based on the Monte Carlo criterion. Finally, the numerical results and analysis demonstrate the proposed algorithm’s superiority on some criteria compared to the non-dominated sorting genetic algorithm-II (NSGA-II).

Which Privacy Policy Works, Privacy Assurance or Personalization Declaration? An Investigation of Privacy Policies and Privacy Concerns

This study focuses on two specific privacy policies, namely privacy assurance and personalization declaration. Specifically, we investigate how these distinct privacy policies affect customers’ privacy concerns and subsequent purchase responses. We have developed a conceptual model that addresses the independent effects of privacy assurance and personalization declaration, as well as the mechanism (i.e., privacy concerns) of these effects. Our model is grounded in motivation theory and supported by a field experiment and a controlled experiment. Our study demonstrates that privacy assurance that claims security protection negatively affects customers’ purchase probability and purchase amount. By contrast, personalization declaration that addresses personalized benefits positively affects customers’ purchase probability and purchase amount. Privacy concerns significantly mediate the negative effects of privacy assurance on purchase responses and the positive effects of personalization declaration on purchase responses. Overall, our findings inform managers of how to deploy privacy policies to reduce customers’ privacy concerns and boost purchase responses.

Impact of Consumer Loss Aversion on Operations in the Context of Remanufacturing

Loss aversion is an important psychological characteristic that has become well supported in finance and marketing. This paper introduces consumer loss aversion into the game model with two substitutable products in a remanufacturing system. The loss-averse consumers gain utility from comparing the price and value dimensions with the superior product. Our model highlights alternative products competition, product selections, and consumption psychology in a remanufacturing environment. The results show that consumer loss aversion has important implications on key decisions in the remanufacturing system. When the manufacturer faces loss-averse consumers, there is a tendency to charge higher prices for two types of products. At the same time, the sales prices increase with the degree of loss aversion. Further analysis reveals that if the manufacturer manipulates consumers with a high probability of purchase, the feature of loss aversion usually benefits him. Our theoretical analyses are proven by numerical computation and some useful conclusions are summarized in this paper.

Detecting Customer Trends for Optimal Promotion Targeting

Problem definition: Retailers have become increasingly interested in personalizing their products and services such as promotions. For this, we need new personalized demand models. Unfortunately, social data are not available to many retailers because of cost and privacy issues. Thus, we focus on the problem of detecting customer relationships from transactional data and using them to target promotions to the right customers. Academic/practical relevance: From an academic point of view, this paper solves the novel problem of jointly detecting customer trends and using them for optimal promotion targeting. Notably, we estimate the causal customer-to-customer trend effect solely from transactional data and target promotions for multiple items and time periods. In practice, we provide a new tool for Oracle Retail clients that personalizes promotions. Methodology: We develop a novel customer trend demand model distinguishing between a base purchase probability, capturing factors such as price and seasonality, and a customer trend probability, capturing customer-to-customer trend effects. The estimation procedure is based on regularized bounded variables least squares and instrumental variable methods. The resulting customer trend estimates feed into the dynamic promotion targeting optimization problem, formulated as a nonlinear mixed-integer optimization model. Though it is nondeterministic polynomial-time hard, we propose a greedy algorithm. Results: We prove that our customer-to-customer trend estimates are statistically consistent and that the greedy optimization algorithm is provably good. Having access to Oracle Retail fashion client data, we show that our demand model reduces the weighted-mean absolute percentage error by 11% on average. Also, we provide evidence of the causality of our estimates. Finally, we demonstrate that the optimal policy increases profits by 3%–11%. Managerial implications: The demand model with customer trend and the optimization model for targeted promotions form a decision-support tool for promotion planning. Next to general planning, it also helps to find important customers and target them to generate additional sales. History: This paper has been accepted as part of the 2019 Manufacturing & Service Operations Management Practice-Based Research Competition. Funding: This work was supported by the U.S. National Science Foundation [Grant CMMI-156334]. Funding from the Oracle Corporation through an ERO grant is also gratefully acknowledged. Supplemental Material: The online appendices are available at https://doi.org/10.1287/msom.2020.0893 .

제품옵션의 크기가 구매가능성에 미치는 영향: 제품유형과 해석수준에 따른 정보처리과정을 중심으로

본 연구는 제품유형에 따른 메시지 유형, 시간적 거리감, 사회적 거리감에 따라서 소비자들이 선호하는 제품옵션의 크기에 대한 수준을 밝히는 것이다. 본 연구에서는 3개의 실험을 진행하였고 분석방법에는 t-test, ANOVA, Chi-square test가 사용되었다. 실험 1은 제품유형에 따른 메시지 유형과 제품옵션에 대해서 분석하였고, 실험 2는 제품유형에 따른 시간적 거리감과 제품옵션에 대해서 분석하였으며, 실험 3은 제품유형에 따른 사회적 거리감과 제품옵션에 대해서 분석하였다. 그 결과 모든 실험에서 쾌락적 제품에 대한 상위수준상황이 제시되었을 경우에는 제품옵션의 크기가 작은 것을, 실용적 제품에 대한 하위수준상황이 제시되었을 경우에는 제품옵션의 크기가 큰 것을 선호하였다.

Supplemental Nutrition Assistance Program participation and current restricted food expenditures: implications for policy.

National concerns over food insecurity and obesity have prompted legislation seeking to further restrict Supplemental Nutrition Assistance Program (SNAP) purchases. The objective of this study is to provide insight on the potential impact of proposed purchase restrictions by comparing SNAP participant and income-eligible non-participants' expenditures on current SNAP-restricted foods, that is, hot foods, prepared foods, alcohol, vitamins and meal supplements. Cross-sectional study. Bivariate analysis and multivariable regression analysis with an instrumental variables approach were employed to compare the probability of purchasing and expenditures on current SNAP-restricted foods among SNAP participants and income-eligible non-participants. National Household Food Acquisition and Purchase Survey, 2012-2013. 2513 households, of which 1316 were SNAP participants and 1197 were income-eligible non-participants. Both the share of households purchasing and household expenditures on current SNAP-restricted foods were similar among SNAP participants and income-eligible non-participants. Results provide further empirical evidence that proposed SNAP purchase restrictions on sugar-sweetened beverages, snack foods and luxury foods are unlikely to have a meaningful effect on SNAP household food purchases.

Impact of a physician-targeted letter on opioid prescribing

We study the effect of a physician-targeted nudge letter on opioid prescribing. In May 2017, the Social Insurance Institution of Finland sent a personal information letter to all physicians who had issued a prescription containing at least 100 tablets of paracetamol-codeine combination to a new patient. The aim of the letter was to draw the physicians’ attention to their prescribing practices and to decrease the size of the first codeine prescription. Using individual level register data and a difference-in-differences strategy, we estimate that the letter decreased the average number of tablets purchased by new patients by 12.5 percent and the probability of a first purchase being at least 100 tablets by six percentage points. We also find that these effects were larger among consistent high prescribers. However, we do not find similar effects on other mild or strong opioids.

Assessment of socio-demographic factors, health status and the knowledge on probiotic dairy products

The probiotic dairy products have become the fastest growing in functional foods market in response to the increasing numbers of consumers interested in improving their health. Therefore, data collected from 314 participants in the city of Bursa, Turkey, were analyzed with the aim to assess the socio-demographic characteristics and health status as well as the consumer’s knowledge and awareness of probiotic dairy products in order to gain a better understanding of consumers’ attitudes. It was observed that socio-demographic characteristics of the participants were not significantly associated with the degree of knowledge or awareness and purchasing frequencies of probiotic dairy products. Having a higher educational level, higher income level and being a woman indicated an increased the awareness, knowledge and purchase probability of probiotic dairy products. This study emphasized that consumers should be more informed about the positive relationship between a healthier life and the consumption of probiotic dairy products.

Industrial buying during the coronavirus pandemic: A cross-cultural study

With the onset of the 2020 coronavirus pandemic, industrial suppliers are increasingly challenged to close their open sales opportunities and keep generating business. Against this backdrop, the authors of this study investigate which offerings industrial customers are most likely to purchase as the pandemic progresses. Drawing on positive decision theory and empirically investigating 31,353 sales opportunities across 57 countries, the authors show that the coronavirus pandemic significantly decreases industrial customers' purchase probability, especially for high-priced offerings. In countries with low uncertainty avoidance and strong long-term orientation (e.g., China, India, Singapore), purchase probability is less affected by the pandemic. The coronavirus pandemic even increases purchase probability for offerings with low prices in countries where cultures are simultaneously uncertainty-avoidant and short-term oriented (e.g., Argentina, Brazil, Mexico). This is presumably because customers safeguard their operations in the face of impending supply shortages. Consequently, this helps suppliers focus on the right sales opportunities to secure their business during exogenous global shocks such as the coronavirus pandemic.

Data-driven decision and analytics of collection and delivery point location problems for online retailers

The location of collection and delivery points (CDPs), impacted by online customers’ demand data, plays an important role for online retailers. While previous delivery points optimization researches do not use customer behavior data, we propose new models, integrating with customer behavior data analysis, to optimize collection and delivery points for online retailers. We explore a real customer behavior data and use totally 257,685 users’ records (212,062 records for training set and 45,623 records test set). We first estimate the customer purchase probability by five data mining models. Based on the estimation results, we establish two facility location models to respectively optimize the attended and unattended CDPs locations with the objective of cost minimization. Our numerical experiments make a quantitative analysis of customer service level and location cost. Our results can further help online retailers to decide the suitable CDPs with trading off the consumer service level and the total logistics cost. We make interesting contributions: (i) we analyze real customer behavior data and find that gradient boosting trees algorithm outperform other four algorithms when estimating customers’ purchase probabilities; (ii) We propose a new data-driven method integrating data mining models and facility location models to determine CDP locations for online retailers.

How Transparency into Internal and External Responsibility Initiatives Influences Consumer Choice

Amid growing calls for transparency and social and environmental responsibility, companies are employing different strategies to improve consumer perceptions of their brands. Some pursue internal initiatives that reduce their negative social or environmental impacts through responsible operations practices (such as paying a living wage to workers or engaging in environmentally sustainable manufacturing). Others pursue external responsibility initiatives (such as philanthropy or cause-related marketing). Through two experiments conducted in the field and complementary online experiments, we compare how transparency into these internal and external initiatives affects customer perceptions and sales. We find that transparency into both internal and external responsibility initiatives tends to dominate generic brand marketing in motivating consumer purchases, supporting the view that consumers take companies’ responsibility efforts into account in their decision making. Furthermore, the results provide converging evidence that transparency into a company’s internal responsibility practices can be at least as motivating of consumer sales as transparency into its external responsibility initiatives, incrementally increasing a consumer’s probability of purchase by 6.40% and 45.85% across our two field experiments, conducted in social and environmental domains, respectively. Our results suggest that it may be in the interest of both business and society for managers to prioritize internal responsible operations initiatives to achieve both top- and bottom-line benefits while mitigating social and environmental harms.This paper was accepted by Charles Corbett, operations management.

Should We Wait Before Outsourcing? Analysis of a Revenue-Generating Blended Contact Center

Problem definition : We consider a revenue-generating call center with inbound and outbound calls, where service and sales activities are blended. For maximizing the call center’s revenue, the call center manager exercises two levels of control: agent reservation for inbound calls and call outsourcing. Given the influence of waits on purchase probability, we investigate the strategy of outsourcing customers who have waited already as opposed to outsourcing customers directly at arrival. Academic/practical relevance : The main novelty of this article arises from the use of a single framework to investigate both the combination of agent reservation with outsourcing decisions, and a waiting time–based outsourcing strategy. The existing literature only considers these two strategies in isolation and is restricted to quantity-based decisions. From a practical viewpoint, our results aim to provide decision support tools that are directly implementable in a call center’s routing software. Methodology : We apply a Markov decision process approach to optimize the manager’s decisions. The particularity of our approach is that we use the experienced waiting time as a decision variable. Results : We prove that the optimal policy for reservation and outsourcing is of threshold type. Our main conclusion is that outsourcing customers after letting them wait in-house generates higher revenue than outsourcing calls at arrival. However, it is also detrimental to service quality. In addition, we identify contexts in which the difference between the two outsourcing strategies is significant. Managerial implications : Contrary to standard call center practices, which either consist of specialized teams for one type of call or only exercise one specific level of decision-making (reservation or outsourcing), we demonstrate the potential of partial outsourcing with partial reservation. Our study shows that the benefits of implementing our results are greatest in small congested call centers.

Selling in the Digital Age

Advances in technology have revolutionized consumer buying behavior. The internet has provided customers with a cheap source of information that can help them make informed buying decisions. However, many salespeople are still relying on the classic model of selling, which is based on the assumption that customers are uninformed, uncertain, and undecided when they meet a salesperson. This model suggests that salespeople, as knowledgeable gatekeepers of information, should guide customers through the sales process. For example, solution selling suggests that the salesperson should diagnose customer needs and then recommend the right products/services to fulfill those needs. The challenger sales model suggests that a seller should actively teach customers, tailor the sales process to them, and take control of the customer conversation In this report, authors address the question: How does employing the classic sales model with customers with greater and lesser preference certainty affect the sales interaction outcomes of (1) purchase probability, (2) sales revenue, and (3) customer satisfaction? In their first study, the authors collected data from 356 individual sales interactions in 15 different stores of a U.S.-based retailer of durable goods over four months. They show that customers with greater preference certainty have a lower purchase probability when the classic sales model is implemented. (i.e., from more than 80% to 60%). Further, the revenue from a sales interaction is reduced by more than 25% when salespeople use the classic sales model on customers with higher levels of preference certainty. Finally, when faced with customers who have a good idea of what they want to buy, salespeople may decrease customer satisfaction by using the classic sales model technique. A second, scenario-based experiment corroborated and replicated the first study, using Amazon’s Mechanical Turk. This study found that, under high levels of preference certainty, challenging the customer results in a lower likelihood to purchase during the shopping visit. Further, these findings were found to be unrelated to price. These findings point to a need to transform the way that salespeople interact with customers. The authors offer two ways that managers can benefit from this research, and help their salespeople move beyond the classic sales model. Training. Sales managers can teach their salespeople about the importance of acknowledging the differences among customers along the decision-making process and train them on the appropriate tactic in each stage. Lead generation. Customer relationship management technology now makes it possible to monitor customers’ research behavior and to index prospects depending on their stage in the process; therefore, this technology can assist salespeople in adopting the appropriate sales tactic. Finally, new sales models should revisit and redefine assumptions of the standard paradigm of selling.