Customer Relations
No Comments (from You): Understanding the Interpersonal and Professional Consequences of Disabling Social Media Comments
Michelle Daniels & Freeman Wu
Journal of Marketing, forthcoming
Abstract:
Presumably in an effort to reduce cyberbullying and promote mental health, online influencers often limit viewers' ability to post comments. In this research, we find that influencers incur significant interpersonal and professional repercussions for doing so. Across a Twitter dataset and six experiments utilizing both consequential and hypothetical dependent measures, we find that consumers form more negative impressions of and are less persuaded by influencers who disable social media comments. These outcomes are driven by the perception that the influencer is less receptive to consumer voice (e.g., consumers' thoughts, opinions, and suggestions) and, thus, less sincere. However, we find that this effect is mitigated in situations where consumers feel that it is reasonable for influencers to prioritize self-protection.
Don't Hurry, Be Happy! The Bright Side of Late Product Release
Mushegh Harutyunyan & Chakravarthi Narasimhan
Marketing Science, forthcoming
Abstract:
When a firm releases its product later than its competitor, the firm loses sales because many consumers prefer to buy the competitor's product rather than wait for the firm's product release. Therefore, in the absence of any late-mover advantages, conventional wisdom suggests that competing firms will release their products as soon as possible to avoid losing customers if they were to enter later. However, when the market evolves over time and consumers are forward-looking, we demonstrate that this intuition fails and propose a new explanation for why a firm may strategically release its product later than its competitor. Namely, a firm's late entry can help alleviate price competition due to some consumers' decisions to wait for the firm's product release instead of buying a currently available product. We show that in markets where the growth rate is sufficiently high and differentiation between firms is not too low, the firm's profit gain from alleviated price competition dominates its profit loss from reduced sales, making the firm better off by releasing its product later than the competitor. Surprisingly, when the fraction of consumers who enter the market relatively early increases, the firm may have even greater incentives to release its product late. Finally, we consider markets where firms are differentiated both horizontally and vertically, with one firm having a higher quality (or stronger brand image) than its competitor. We find that high level of vertical differentiation will induce both high- and low-quality firms to rush to the market, releasing their products in the early period. However, when vertical differentiation is moderately high, the high-quality firm may choose to release late, whereas the low-quality firm will prefer to release its product early.
Retail Karma: How Our Shopping Sins Influence Evaluation of Service Failures
Ran Li, Meng Zhang & Pankaj Aggarwal
Journal of Consumer Research, forthcoming
Abstract:
Consumers have an intuitive belief in "karma" which dictates that bad (good) actions lead to bad (good) outcomes. Consequently, consumers perceive a causal connection between their own wrongdoing toward a company and a subsequent service failure that they experience in their interactions with another company. Eight experiments employing different contexts consistently show that consumers who have previously wronged a company (compared to those in a control group) evaluate another unrelated company more positively in response to a service failure by this company. We argue that this more positive evaluation is due to the greater blame consumers assign to themselves as dictated by the "karmic beliefs" held by consumers whereby the subsequent poor service by a different firm is seen as a karmic payback for their own prior transgression. The proposed effect is mitigated when a person's karmic belief is reduced. We also examine a number of alternative explanations (e.g., negative experiences, moral balancing, and immanent justice reasoning) and find that our observed effect is more consistent with a karma-based account.
Online Advertising as Passive Search
Raluca Mihaela Ursu, Andrey Simonov & Eunkyung An
Management Science, forthcoming
Abstract:
Standard search models assume that consumers actively decide on the order, identity, and number of products they search. We document that online, a large fraction of searches happen in a more passive manner, with consumers merely reacting to online advertisements that do not allow them to choose the timing or the identity of products to which they will be exposed. Using a clickstream panel data set capturing full URL addresses of websites that consumers visit, we show how to detect whether a click is ad initiated. We then report that in the apparel category, ad-initiated clicks account for more than half of all website arrivals, are more concentrated early on in the consumer search process, and lead to less in-depth searches and fewer transactions, consistent with the passive nature of these searches. To account for these systematic differences between active and passive searches, we propose and estimate a simple model that accommodates both types of searches. Our results show that incorrectly treating all searches as active inflates the estimated value of brands that advertise frequently. Our model can more accurately recover data patterns, especially for advertising brands. We finish with model extensions and a discussion of the managerial implications.
Can Large Language Models Capture Human Preferences?
Ali Goli & Amandeep Singh
Marketing Science, forthcoming
Abstract:
We explore the viability of large language models (LLMs), specifically OpenAI's GPT-3.5 and GPT-4, in emulating human survey respondents and eliciting preferences, with a focus on intertemporal choices. Leveraging the extensive literature on intertemporal discounting for benchmarking, we examine responses from LLMs across various languages and compare them with human responses, exploring preferences between smaller, sooner and larger, later rewards. Our findings reveal that both generative pretrained transformer (GPT) models demonstrate less patience than humans, with GPT-3.5 exhibiting a lexicographic preference for earlier rewards unlike human decision makers. Although GPT-4 does not display lexicographic preferences, its measured discount rates are still considerably larger than those found in humans. Interestingly, GPT models show greater patience in languages with weak future tense references, such as German and Mandarin, aligning with the existing literature that suggests a correlation between language structure and intertemporal preferences. We demonstrate how prompting GPT to explain its decisions, a procedure we term "chain-of-thought conjoint," can mitigate, but does not eliminate, discrepancies between LLM and human responses. Although directly eliciting preferences using LLMs may yield misleading results, combining chain-of-thought conjoint with topic modeling aids in hypothesis generation, enabling researchers to explore the underpinnings of preferences. Chain-of-thought conjoint provides a structured framework for marketers to use LLMs to identify potential attributes or factors that can explain preference heterogeneity across different customers and contexts.