English II

Comparison Shopping Engine

Buyers worldwide have always been inclined towards comparison shopping even before the Internet era. Price competitiveness, product attributes, and merchant reputation are some of the considerations that precede many purchasing decisions and thus, taking the trends online the buyers will be able to carry out multiple website comparisons with ease. Despite the fact that the internet has made everything easier, it has also remained a challenge to many people in that they lack the capacity of sailing through different terminals of the website and achieve their intended goal. Price comparison heavily depends on the volume of data that the involved person has. The internet remains the main source of data that one can sufficiently have to carry out these tasks. With effect to these issues, reevaluation of the challenges with which people are undergoing while trying to compare the prices online is a determining factor to ensure that they run these activities free of difficulties. For instance, growth in internet data has led to the development of various search engines that have helped people in many ways. Many people don’t really understand the merits of online shopping over the traditional one, forming the foundation of preferring the traditional system that is less convenient compared to online shopping. Comparison of product prices from a single point remains advantageous. This saves the time spent as well as reducing the resources since they are only used once. On the contrary, the internet, online things come with a number of challenges, fraud and cybersecurity are the common threats when one decides to move digital. Increased competition has forced many merchants to look for ways through which they can cut short the costs incurred. With this effect, therefore, merchants should make use shopping engines to drive exposure and sales to their stores, spreading brand awareness as well as bringing in repeat buyers (Evans & Wurster, 2016)

Literature review

Shopping engines are a two-faceted tool that on one hand has made it possible to conveniently shop and on the other hand, haven't fully addressed the problem of merchants’ listing and management costs. In this chapter, the dynamics surrounding customer and merchant relationship in e-commerce are studied using opinions and findings from other researchers. The design decisions of a shopping engine are critically evaluated and at the end of this chapter, a suitable concept is conceived to realize the objectives of this project.

Description of the current system

The European Commission (2012) asserts the benefits of online shopping as perceived savings in time, better prices, easier price comparisons, a wider selection and availability on a 24-hour basis. According to Andam (2015), e-commerce reduces the information search costs as well as transaction costs for the buyer in developing countries. E-commerce models in Kenya can be broadly categorized into; Business-to-consumer (B2C) where products and services are sold from company to the general public (end users); Consumer-to-consumer (C2C) where products and services in an online marketplace are sold by consumers to fellow consumers, Business-to-business (B2B) where one business engages another in exchange of products or services, Consumer-to-business (C2B) where consumers sell their services to a business. The B2C model is the most popular in terms of market adoption. This research uses public data from Alexa.com to access traffic levels for the e-commerce websites where Jumia.co.like, Olx.co.ke, Amazon.com, and Kilimall.co.ke are leading in that order.

How the current system works

The process of submitting product feeds involves merchants downloading an XML file of all products from their store with prices, short descriptions, colors, sizes, and other attributes. With this file, the merchants then log into their shopping engine accounts then upload the file. After the product feed has been approved by moderators, products start showing in their respective categories of the shopping engine so that customers can view them. When the merchant wants to edit a product (e.g., change the price), they have to log into the control panel of the shopping engine then use a search box to find the product before editing. This process is obviously tedious for very large online stores such as Jumia and Masoko with hundreds of thousands of products each. The manual system also poses a challenge to the moderators of the shopping engine who now have to deal with a catalog of products which may often be out of date. Buyers, on the other hand, expect the shopping engine to provide up-to-date information but may not always be the case. Furthermore, the current shopping engines fail to provide an opportunity for offline merchants to list their products for customers that may be interested in them.

Weaknesses of the current system

Due to the increase in the number of e-commerce sites in Kenya, and the growing range of products offered, the current shopping engines fall short in catching up with the exponential growth of listing and management costs for merchants. This is mainly because merchants have to manually update the shopping engines as product details change on their stores. This process is obviously tedious for very large online stores such as Jumia and Masoko with hundreds of thousands of products each.

The manual system also poses a challenge to the moderators of the shopping engines who now have to deal with a catalog of products which may often be out of date. Customers, on the other hand, expect the shopping engine to provide up-to-date information but may not always be the case. Furthermore, the current shopping engines fail to provide an opportunity for offline merchants to list their products for customers who may be interested in them.

Many researchers agree to shop engines being indispensable as the scope of e-commerce grows. Their primary value is aggregating merchant information and transforming it into useful and creative ways to aid buyer purchasing. Since typically shopping engines hold more data than the individual merchants, they can be further advanced using machine learning to recommend products to the buyers therefore further lowering their searching costs on one hand and improving lead generation for merchants on the other. Product recommendation using machine learning works based on buyers’ preferences, browsing history, characteristics (e.g. gender and age) or based on their similarities with a group of buyers (clustering). Gupta, M. et al. (2014) demonstrates a use case for machine learning in shopping engines by predicting user behaviors for example whether they are more likely to buy or leave without buying. In their quest to develop a smarter crawler, Doorenbos et al. (2017) faced the difficulty of the unstructured nature of markup language used by e-commerce websites - therefore recommending studies into natural language processing as a means to gather data semantically from merchant websites. The study of shopping engines leads to the disciplines of distributed systems, cloud computing, and big data since comparison shopping involves the collection and storage of large amounts of data. Legal implications concerning data reuse also make the domain of comparison shopping a rich area of study by legal professionals. Data reuse regulations are possibly the biggest threats facing shopping engines because of their reliance on online merchants for product information. Zhu & Madnick (2018) provide an elaborate history of suits against shopping engines as well as good practices for shopping engines to mitigate such losses and remain in business. Finally, the role of economists in directing the next phase of comparison shopping cannot be ignored. This is because shopping engines act as intermediaries between buyers and retailers and therefore there is the need for creation of a neutral environment for merchants to ensure business activity thrives without creating a cesspool of price wars, monopoly or diminishing returns as is always the concern of especially smaller merchants on shopping engines. Fowler (2014), addresses in depth most of these economic issues that are directly linked to shopping engines.

References

Andam, Z. (2015). E-Commerce and e-Business. E-ASEAN Task Force UNDP-ADIP (pp. 1–47).

Doorenbos, B., Etzioni, O., Weld, S. (2017): "A Scalable Comparisons-Shopping Agent for the World-Wide Web", University of Washington, Seattle, WA 98195

Evans, P, Wurster T.S (2016) Blown to Bits: How the New Economics of Information Transforms Strategy.

Fowler, G.A. (2014), “Auctions Fade in eBay‟s for Growth” The Wall Street Journal, May 26, 2009, A1.

Gupta, M., Mittal, H., Singla, P., Bagchi, A. (2014): “Characterizing Comparison Shopping Behavior: A Case Study”, Indian Institute of Technology, New Delhi, India

Zhu, H. and Madnick, E. (2018), "Legal Challenges and Strategies for Comparison Shopping and Data Reuse", Journal of Electronic Commerce Research, Vol 11, No 3, 2010