Despite a general consensus that use of information technology (IT) is an important link between IT investments and performance, the extant literature provides only a limited explanation as to when the use of IT lifts performance. We posit that the impact of knowledge management systems (KMS) usage is contingent on users' alternative sources of knowledge as well as their specific task environments. We investigate under what conditions repository KMS use leads to higher performance outcomes in a retail grocery context. We use a unique longitudinal dataset composed of objective measures of KMS use and sales performance of 273 managers over 146 weeks collected from a retail grocery chain. We obtain two main results. First, we find a diminishing impact of KMS use for managers who also use other sources of codified knowledge, namely physical or computerized alternative knowledge sources, whereas a complementary relationship seems to exist between KMS use and social sources of knowledge. Second, KMS use produces higher benefits for managers whose task environments require a greater volume of information and knowledge, but smaller benefits for those managers whose task environments demand rapidly changing information and knowledge. Our work contributes to both the IT business value and the KM literature by studying the contingent impact of IT usage while broadening the theoretical scope of the situated knowledge performance framework with a critical empirical test based on fine-grained objective and longitudinal data.
The objective of this work is to examine various psychological forces underlying the behavior of people’s online gambling, an increasingly popular form of entertainment in the gaming industry. Drawing on extant theories, we first developed a model of how cumulative outcomes, recent outcomes, and prior use affect online gambling behavior differently. We empirically tested the model using longitudinal panel data collected over eight months from 22,304 actual users of a gambling website. The results of a multilevel panel data analysis strongly supported our hypotheses. First, consistent with gambling theory, individuals' online gambling was found to increase with any increase in a cumulative net gain or cumulative net loss. Second, as the availability heuristic prescribes, a recent loss reduced online gambling, whereas a recent gain increased it. Third, consistent with the literature on repeated behavior, regular use and extended use moderated the relationship between current and subsequent gambling. Taken together, the present study clarifies how people react differently to immediate and cumulative outcomes and also how regular use and extended use facilitate routine behavior in the context of online gambling. In general, our findings suggest that the three perspectives, i.e., gambling theory, the availability heuristic, and repeated behavior, should be taken into account to understand online gambling, which is in essence a series of risk-taking attempts with the potential of eventually becoming routine behavior. This study is expected to offer valuable insights into other types of online games that could engage people in risking real or cyber money and, at the same time, could be easily enmeshed with everyday life (e.g., fantasy sports, online virtual worlds).
Despite growing interest in the economic and policy aspects of information security, little academic research has used field data to examine the development process of a security countermeasure provider. In this paper, we empirically examine the learning process a security software developer undergoes in resolving a malware problem. Using the data collected from a leading antivirus software company in Asia, we study the differential effects of experience on the malware resolution process. Our findings reveal that general knowledge from cross-family experience has greater impact than specific knowledge from within-family experience on performance in the malware resolution process. We also examine the factors that drive the differential effects of prior experience. Interestingly, our data show that cross-family experience is more effective than within-family experience in malware resolution when malware targets the general public than when a specific victim is targeted. Similar results—for example, the higher (lower) effect of cross-family (within-family) experience— were observed in the presence of information sharing among software vendors or during a disruption caused by a catastrophe. Our study contributes to a better understanding of the specific expertise required for security countermeasure providers to be able to respond under varying conditions to fast-evolving malware.
To improve operational efficiencies while providing state of the art healthcare services, hospitals rely on information technology enabled physician referral systems (IT-PRS). This study examines learning curves in an IT-PRS setting to determine whether agents achieve performance improvements from cumulative experience at different rates and how information technologies transform the learning dynamics in this setting. We present a hierarchical Bayes model that accounts for different agent skills (domain and system) and estimate learning rates for three types of referral requests: emergency (EM), nonemergency (NE), and nonemergency out of network (NO). Furthermore, the model accounts for learning spillovers among the three referral request types and the impact of system upgrade on learning rates. We estimate this model using data from more than 80,000 referral requests to a large IT-PRS. We find that: (1) The IT-PRS exhibits a learning rate of 4.5% for EM referrals, 7.2% for NE referrals, and 12.3% for NO referrals. This is slower than the learning rate of manufacturing (on average 20%) and more comparable to other service settings (on average, 8%). (2) Domain and system experts are found to exhibit significantly different learning behaviors. (3) Significant and varying learning spillovers among the three referral request types are also observed. (4) The performance of domain experts is affected more adversely in comparison to system experts immediately after system upgrade. (5) Finally, the learning rate change subsequent to system upgrade is also higher for system experts in comparison to domain experts. Overall, system upgrades are found to have a long-term positive impact on the performance of all agents. This study contributes to the development of theoretically grounded understanding of learning behaviors of domain and system experts in an IT-enabled critical healthcare service setting.
Although companies have spent a great deal of money to adopt CRM (customer relationship management) technologies, many have not seen satisfactory returns on their CRM implementations. We study optimal CRM implementation strategies and the impact of CRM investments on profitability. For our analysis, we classify CRM technologies into two broad categories: targeting-related and support-related technologies. While targeting CRM improves the success rate of distinguishing between nonloyal and loyal customers, support CRM increases the probability of retaining the loyalty of existing customers. We also consider the costs of implementing each CRM type separately as well as both types simultaneously. We show that the optimal CRM implementation strategy depends on the initial mass of loyal customers and diseconomies of scale in simultaneous implementation. We also find that the two types of CRM technologies are substitutive rather than complementary in generating revenue. We discuss why it is difficult to avoid overinvestments in CRM when the nature of the investments is misunderstood. We study the optimal CRM implementation scope and the impact of different types of CRM on customers. We develop a model that not only considers both the revenue and costs sides but is also helpful in determining the deployment of right CRM technology in the right scope.