21 July, 2022.

Disclaimer:

At the outset, I would like to make it clear that this is a blog, and I have expressed my personal opinion. I fully appreciate that others may differ from my opinion, and I do not confront them.

Introduction

In the present era of COVID and digital marketing, people’s lifestyles have changed; they are now more prone to online shopping than before (UNCTAD, 2020). Besides protecting us from the vulnerabilities of the COVID environment, this’ armchair shopping’ provides great convenience in shopping – free from driving, traffic and parking troubles. Online shopping also allows us to browse through a vast number of products closely, which would be impossible in a physical store. Of course, the icing is the price (Jadhav & Khanna, 2016).

In tune with the above, the businesses do all that is necessary to keep the prospect’s experience of online search satisfying and enjoyable. A popular strategy used in digital marketing is the use of recommendation agents. The recommendation agents help the prospects to arrive at the right product by showing various options. The goal of the recommendation agents is not only to make the search experience pleasant but also to drive the prospect to buy the product. Hence, the recommendation agents are critical in digital marketing.

Designing a good recommendation agent is challenging (Xu, Benbasat, & Cenfetelli, 2014). An inappropriate recommendation agent will not only fail in its objectives but may also drive users to shift to another website. The extant literature has plenty of papers on recommendation agents, their importance and their usefulness. Nevertheless, the question of what value of a recommendation agent is important has missed the attention of the researchers so far. But not now. Therefore, a study has been proposed to examine users’ online search experience by investigating what values of recommendation agents are essential to the users.

Recommendation agents’ values

The purpose of a recommendation agent is to help the user by recommending the correct product after understanding the preferences. Initially, the website gathers the users’ information, such as their location and preferences, either during the search process or from their earlier browsing/purchasing data. Thus the recommendation agents make a personalised recommendation to suit the users’ liking (Xiao & Benbasat, 2014). A surprising find is that while the recommendation agents seem to work well in a laboratory setup, it is not the case in the actual marketplace. While most (52%) of the prospects are unhappy with the online buying experience, even the B2B buyers switch to competitors’ websites (Fletcher, 2021). The cause is not known. Hence this study has been proposed to understand the recommendation agent’s values such as a) match accuracy, b) recommended products and c) image appeal by considering factors such as diagnosticity and serendipity that influence the prospects’ decision satisfaction and buying behaviour.

Generally, it was considered that the more accurate a recommendation agent offers an appropriate product to the customer, the better it is. However, we know that people browse and search online with an intention to buy (35%) and without (25.5%) (Niu, Huang, & Chen, 2021). Under this context, firms soon realised that the people welcome unexpected discoveries that give them an opportunity to get to know a new product accidentally. Hence, it has evolved that to give a good online search experience, the people should be given a recommendation that is both high in diagnosticity and serendipity. Diagnosticity means the recommendation agent’s ability to correctly suggest a suitable or an appropriate product (Jiang & Benbasat, 2004). Serendipity means the ability of the recommendation agent to put forward products that the user is not searching for but may be relevant (Yi, Jiang, & Benbasat, 2017).

Many studies have established the usefulness of recommendation agents by evaluating their diagnostic value where they guide the user to discover the product of their choice. Serendipity, “an incident-based, unexpected discovery of information” (Agarwal N., 2015), has also attracted the researchers’ attention and several papers are available on it. Similarly, many have looked into improving the recommendation agents’ algorithms.

A potential customer, who intends to buy a product online, begins by searching online. When he gets a product of his liking, he may decide to buy it after evaluating it. A recommendation agent is “an interactive decision aid that assists consumers in the initial screening of the available alternatives in an online store” (Haubl & Trifts, 2000). The interactivity of recommendation agents increases the customer’s satisfaction. Moreover, Kaminskas and Bridge (2016) think those recommendation agents should include novelty and diversity for added customer satisfaction. The recommendation agents also display products under “Most wished for”, “Hot new releases” (Amazon), “Top Seller” (IKEA) and “Best Seller” (Pepperfry). Displays such as those mentioned above expose a wide range of products to the users. This kind of promotion enables the consumers to browse more than what they would be capable of examining physically. The numerous choices displayed may also lead the customer to the serendipitous discovery of a product the customer had not thought about earlier. Besides the accuracy of the recommendation agents and the product variety suggested, the picture of the products has a role in swaying a consumer’s decision and providing a satisfactory search experience (Christel, 2006).

Keeping the three aspects of the recommendation agents, i.e., match accuracy, product variety and image appeal, the proposed paper will also look into the customer’s decision satisfaction that finally leads to purchase intention.

Conclusion

My personal feeling is that the paper would shed light on the aspects of the recommendation agents’ values that are important to the customers. The diagnostic and serendipity of the recommendation agents have also been considered separately. For this research paper, primarily social media has been used to collect the data. Initially, no students have been contacted, albeit our alumni have been reached out to.

Works Cited

1. Agarwal, N. (2015). Towards a definition of serendipity in information behaviour. Information Research: An International Electronic Journal, 20(3), 12. Retrieved from http://www.informationr.net/ir/20-3/infres203.html.
2. Fletcher, H. (2021, September 23). The B2B Future Shopper Report. Retrieved March 20, 2022, from https://www.wundermanthompson.com/insight/the-b2b-future-shopper-report-2021
3. Haubl, G., & Trifts, V. (2000, February 1). Consumer decision making in online shopping environments: the effects of interactive decision aids. Marketing Science, 19(1), 4-21. doi:10.1287/mksc.19.1.4.15178
4. Jadhav, V., & Khanna, M. (2016). Factors influencing online buying behavior of college students: a qualitative analysis. The Qualitative Report, 21(1), 1-15. Retrieved from https://nsuworks.nova.edu/tqr/vol21/iss1/1
5. Kaminskas, M., & Bridge, D. (2016). Diversity, serendipity, novelty, and coverage: a survey and empirical analysis of beyond-accuracy objectives in recommender systems. ACM Transactions on Interactive Intelligent Systems, 7(1), 1-42. doi:10.1145/2926720
6. Niu, W., Huang, L., & Chen, M. (2021, October). Spanning from diagnosticity to serendipity: An empirical investigation of consumer responses to product presentation. International Journal of Information Management, 60. doi:10.1016/j.ijinfomgt.2021.102362
7. UNCTAD. (2020). COVID-19 has changed online shopping forever, survey shows. Ginebra 10, Suiza: UNCTAD – Palais des Nations, 8-14, Av. de la Paix, 1211 Ginebra 10, Suiza.
8. Xiao, B., & Benbasat, I. (2014). Research on the use, characteristics, and impact of e-commerce product recommendation agents: A review and update for 2007–2012. In F. MARTÍNEZ-LÓPEZ (Ed.), Handbook of Strategic e-Business Management. (pp. 403-431). Berlin: Springer Berlin Heidelberg. doi:10.1007/978-3-642-39747-9_18
9. Xu, J. D., Benbasat, I., & Cenfetelli, R. T. (2014). The Nature and Consequences of Trade-Off Transparency in the Context of Recommendation Agents. MIS Quarterly 38 (2): 379–406., 38(2), 379-406.
10. Yi, C., Jiang, Z., & Benbasat, I. (2017, April 20). Designing for diagnosticity and serendipity: An investigation of social product-search mechanisms. Information Systems Research, 28(2), 413-429. doi:10.1287/isre.2017.0695