Friday, March 22, 2024
More
    HomeBusinessUnderstanding Online Consumer Behavior Regarding Purchase Intention and Recommender Systems

    Understanding Online Consumer Behavior Regarding Purchase Intention and Recommender Systems

    Understanding online consumer purchase intention is difficult. Choosing the most suitable product or service online can be a tiring process, given the overwhelming variety of products on retailers’ websites. Making the right online product selection is a time-consuming task due to the abundance of information sources, products, and store alternatives. As a result, buyers cannot evaluate all available options, leading to an increased interest in examining consumer motivations and identifying determinant factors influencing decision-making in online shopping.

    Before making a purchase decision, buyers search for product information online, analyze various alternatives, and may review product feedback from other customers. Additionally, electronic word of mouth (eWOM) and online advertisements play a role in influencing decision-making behavior. The characteristics of products on the manufacturer’s website also significantly impact consumer purchasing behavior. Once research on utility, available features, or desired quality is complete, questions about price arise. Therefore, information sources providing price aggregations, special offers, and discounts are among the most popular.

    From a different perspective, recommender systems are employed by companies to assist customers in reducing time spent searching for the right product. Personalized content delivery has become a highly publicized topic, and recommender systems play a crucial role in the evaluation of alternatives during the purchase process. Recommendations have a positive impact on sales and influence purchasing decisions. Consequently, various recommender systems have been developed to provide personalized information to shoppers.

    Purchase Intention: Understanding Online Consumer Behavior

    In their study titled “The Effect of Perceived Usefulness of Recommender Systems and Information Sources on Purchase Intention“, researchers Daniel Mican and Dan Andrei Sitar Taut address gaps in existing literature by theorizing and empirically validating the influence of different information sources and the persuasiveness of recommendations on consumer purchase intention. The study investigates how consumers’ attitudes toward the perceived usefulness of various types of personalized and non-personalized recommender systems, as well as the relevance of recommendations and information sources, impact purchase intention. The authors propose a structural model that sheds light on emerging integrated models of buying behavior and decision-making.

    Results of this study offer several key implications for theory and practice. Purchase intention is particularly influenced by the perceived relevance of recommender systems, information provided by manufacturers, and online reviews. Additionally, the impact of the perceived usefulness of personalized recommendations strongly affects purchase intention and is mediated by the perceived relevance of recommender systems. Moreover, the results emphasize that recommender systems providing non-personalized recommendations do not affect purchase intention. Electronic word of mouth (eWOM), price, and discounts for favorite products also influence consumer behavior and purchase intention. Online store managers and recommender system developers should consider hybridizing recommender systems, aggregating not only individual shopping behavior but also that of friends and influencers followed on social networks. Developers should capture informative signals from multiple sources (reviews, comments, manufacturer pages) to generate diverse explanations within recommender systems. Additionally, when recommending discounted products, it should be noted that only discounts for preferred products influence purchase intention.

    Furthermore, the study provides practical insights into consumer behavior for better understanding purchase intention and offers suggestions for decision-makers. Conclusions highlight that the perceived relevance of recommender systems mediates the impact of perceived usefulness of personalized recommendations on purchase intention. To maximize business opportunities, online retailers should adopt a highly adaptable and goal-sensitive personalization strategy. Thus, developers should opt for hybridizing recommender systems, aggregating both individual shopping behavior and that of friends and influencers followed on social networks. Store managers should ensure customers receive personalized recommendations and promotions based on individual shopping patterns, aiming to make them feel unique and positively impact purchase intention. Moreover, developers should incorporate social signals, considering the strength of connections, homophily, and geographic proximity of network members to provide the most personalized recommendations. The inclusion of social context in recommender systems is particularly crucial when offering new recommendations that enhance users’ purchase intention.

    Capturing informative signals from multiple data sources is essential for providing explanations. Therefore, recommender systems should include links to product presentation pages on manufacturers’ websites, offering detailed and quality information. Additionally, manufacturers should be encouraged to provide images and useful information about product characteristics, explaining how they meet consumer needs. Retailers should also make efforts to manage consumer feedback, as it influences their willingness to make a purchase. One strategy to encourage buying behaviors is to change the order in which online reviews are displayed based on consumption type and gender. For example, for hedonic purchases, reviews with subjective expressions should be highlighted for male readers, and for utilitarian purchases, for women. Furthermore, hybrid critiquing systems can aggregate feelings and opinions extracted from product reviews, providing an overview of other customers’ experiences with different features. Recommendations can be organized into different categories to highlight representative benefits in terms of feelings and specifications, encouraging users to compare multiple items. Regarding recommendations, adopting a strategy to dynamically adjust the degree of serendipity based on the curiosity of the target user is beneficial. Additionally, the study demonstrated that discounts and reductions influence purchase intention, but only when offered for products buyers have already expressed interest in.

    The complete study has been published in the Kybernetes journal, and the authors can be contacted regarding this study on ResearchGate.

    Daniel Mican
    Daniel Micanhttps://scholar.google.com/citations?user=OJVwSwwAAAAJ
    Daniel Mican is an associate professor and PhD supervisor in the field of Business Information Systems. His research interests are in the area of recommendation systems, web usage mining, collective intelligence, user behavior, and social media. In his free time, he writes articles for Gherf.com.
    RELATED ARTICLES

    LEAVE A REPLY

    Please enter your comment!
    Please enter your name here

    LATEST ARTICLES