On-line personalized sales promotion in electronic commerce

https://doi.org/10.1016/j.eswa.2003.12.017Get rights and content

Abstract

Electronic Commerce encompasses all electronically conducted business activities, operations, and transaction processing. With the development of electronic commerce in the Internet, companies have changed the way they connect to and deal with their customers and partners. Businesses now could overcome the space and time barriers and are capable of serving customers electronically and intelligently. However, it is quite a great challenge to attract and retain the customers over Internet due to the low barrier of entrance and severe competition.

Personalization, a special form of differentiation, when applied in market fragmentation can transform a standard product or service into a specialized solution for an individual. In this research, an on-line personalized sales promotion decision support system is proposed. The proposed system consists of three modules: (1) marketing strategies, (2) promotion patterns model, and (3) personalized promotion products. The marketing strategies contain sales promotion strategies and pricing strategies. Promotion patterns are generated according to various sales promotion strategies, and the promoted prices for the promotion products are generated by considering both the current stages of business life cycle and product life cycle. In the promotion patterns model, by segmenting the market, customer behaviors of three categories can be analyzed by utilizing data mining techniques and statistical analysis to generate personalized candidate promotion products. Finally, multiple evaluation indicators are used and adjusted to rank and obtain the final personalized promotion products. With the promotion products based on customers' past frequent purchase patterns, it has the potential to increase the success rate of promotion, customer satisfaction, and loyalty. In this paper, a prototype system was developed to illustrate how the proposed on-line personalized promotion decision support system works in electronic commerce and a simplified case of performance analysis was conducted for evaluation.

Introduction

In addition to providing a new channel, Electronic commerce (EC) has the potential of serving customers better if it can take good advantage of information technology and develop EC-specific marketing strategy. Marketing is a social and managerial process by which individuals and groups obtain what they need and want through creating, offering, and exchanging products of value with others (Kotler, 1997). EC, not just the purchase of goods and services over the Internet, is a broad term. It encompasses all electronically conducted business activities, operations, and transaction processing. With the development of the Internet and EC, companies have changed the way they connect to and deal with their customers and partners. Businesses hence could overcome the space and time barriers and are now capable of serving customers electronically and intelligently.

Internet marketing and commerce has shown many cases of uncertainties, potentials, and impact. It expands the opportunities for branding, innovation, pricing, and selling. However, the exponentially increasing amount of data and information along with the rapid expansion of the business web sites and information systems makes a business hard to manage and leverage the potential power of EC. Therefore, emerging data analysis techniques such as data mining are capturing researchers' and businesses' great attention. One main purpose of utilizing data mining technology in EC is to attract and retain customers. While there are diverse approaches in Internet marketing, one solution is to provide personalized information services (Schafer, Konstan, & Riedl, 2001).

Personalization, a special form of differentiation, when applied in market fragmentation can transform a standard product or service into a specialized solution for an individual. Through personalization, businesses can get to know customers' buying behaviors and accordingly develop more appropriate marketing strategies to attract each customer of a specific type and efficiently deliver the suitable information and products/services to him/her. The customer's satisfaction and loyalty can thus be enhanced, and the increase in each customer's visiting frequency can further create more transaction opportunities and benefit the Internet businesses (Lee, Liu, & Lu, 2002).

Surprenant and Solomon (1987) proposed three types of personalization: option, programmed, and customized personalization, while Dean (1998) classified web site personalization into three categories: rule-based, collaborative, and learning-agent personalization. A number of web-based personalized systems have been proposed recently (Borchers, Herlocker, Konstan, & Reidl, 1998). Personalization usually works by filtering a candidate set of items (such as products) through some representation of a personal profile. The technology challenges to supporting personalization include the need to perform clustering and searching in a very large dimensional data space with huge amount of data. In general, there are two major approaches to provide personalized information: content-based and collaborative filtering (Yu, 1999, Aggarwal et al., 1999). In the content-based approach, it provides items that are similar to what the user has favored in the past. Some recommendation systems operate based on this approach, such as NewsWeeder (Lang, 1995) and Infofinder (Krulwich & Burkey, 1996). However, there are some shortcomings: implement difficultly to several non-text multimedia resources, like movies, music, etc.; moreover, a user's preferences localize one specified domain, unable to make other classified recommendations. As a result, collaborative filtering approach is presented. In the collaborative filtering approach, it identifies other users that have showed similar preference to the given users and provides what they would like. A lot of recommendation systems are developed based on this approach, such as Tapestry (Goldberg, Nichols, Oki, & Terry, 1992), GroupLens (Konstan et al., 1997), Ringo (Shardanand & Maes, 1995), PHOAKS (Terveen, Hill, Armento, McDonald, & Creter, 1997), and Siteseer (Rucker & Polenco, 1997). But, some drawbacks come up, e.g. unable to provide new items to a user, and unsuited to a user whose liking is different from other users. After analyzing the existing approaches, in this paper, we integrate the two approaches and present an on-line personalized promotion decision support system (DSS), which uses the data mining techniques to help the business discover suitable promotion products for each individual customer.

This research aims to propose a personalized promotion DSS, which can provide personalized promotion products at customized prices for each specific individual customer. By utilizing powerful data mining techniques, more suitable promotion products can be selected even better than experienced professional sales persons. For achieving the purpose of personalization, we must cluster all the customers firstly. By using ART (Adaptive Resonance Theory network) (Carpenter & Grossberg, 1988), we cluster the customers according to two types of attributes, one is the factual attributes which include demographic information such as gender, age, income; and the other is transactional data, which consist of the customers' purchase records, for instance, the purchase date, amount paid, etc.

Secondly, according to the traditional marketing techniques, we embed three sales promotion strategies into the promotion decision support system: general promotion, cross-selling and up-selling, to find personalized promotion products by analyzing customers' purchasing behavior using data mining, where customers are divided into three categories: all customers, customer cluster, and an individual customer. That is to say, we not only analyze the all customers' purchasing behaviors, but also extract the purchasing behaviors of different customer clusters and individual customers. The first strategy, general promotion strategy aims at finding the best-selling and worst-selling products. Cross-selling strategy is to sell additional cross-related products to the customers. In this paper, we will discover the association and sequential products by utilizing association rule mining and sequential patterns mining, respectively. The last one, up-selling strategy is the closely related case of getting existing customers to trade up to more profitable products. Here, up-selling is applied to the best-selling, association, and sequential products restricted within the same ‘brand’. In short, we integrate data mining techniques and cross analysis to carry out the three strategies and propose a new pricing strategy to provide personalized promotion products for each customer. Accordingly, an on-line personalized promotion DSS for EC is developed and proposed in this paper.

In Section 2, related data mining techniques are introduced. Section 3 proposes an on-line personalized promotion DSS and Section 4 describes the developed DSS. A performance evaluation of the proposed DSS is discussed in Section 5. Finally, Section 6 concludes the paper and presents some future research directions towards the further development and enhancement of the system.

Section snippets

Data mining

Data mining refers to extracting knowledge from a large amount of data (Han & Kamber, 2001). Data mining by automatic or semi-automatic exploration and analysis on a large amount of data items set in a database can discover potentially significant patterns inherent in the database. Kleissner (1998) defined that data mining is a new decision support analysis process to find buried knowledge in corporate data and deliver understanding to business professionals. Hence with data mining analysis,

An on-line personalized promotion decision support system

In this section, an on-line personalized promotion decision support system (PPDSS), which uses data mining techniques in accordance with the proposed marketing strategies to help the business prepare the highly potential and suitable promotion products for each individual customer, is presented. Fig. 1 shows the architecture of the proposed PPDSS, which consists of three modules: (1) marketing strategies, (2) promotion patterns model, and (3) personalized promotion products. Each module of the

Development of the PPDSS

In order to illustrate how the proposed PPDSS can function and work well with EC virtual stores, a prototype is developed using programming languages JAVA and PHP. In the prototype system, the number of products is 50, the number of customers is 1500, and there are 10,000 transaction records in the experimental database. 4.1 Generating personalized promotion products, 4.2 Pricing of the promotion products describe all the interfaces for the backend decision maker, and Section 4.3 introduces the

Performance evaluation

In this section, the proposed PPDSS is evaluated in terms of profit gain. In the simplified simulation scenario, the costs, within the range between 30 and 250, of 10 promotion products are randomly generated and their list prices are also generated accordingly with various amounts of gross profit per unit as shown in Table 13. Since different personalized promotion products will be provided for different customers, the PPDSS will actively provide some of them as promotion products when each

Conclusions

It is quite a challenge that a business will face more competitors in Internet than in traditional market, and the customers' loyalty in the Internet is low compared with traditional market so that it is a difficult problem for a business to attract and retain customers in EC. Traditional mass marketing is no longer effective for EC in the Internet, and thus more precise on-line one-to-one marketing for better suiting each customer becomes more and more important for competing in the Internet,

Acknowledgements

This research was supported by the National Science Council, Taiwan, R.O.C., under contract no.: NSC 92-2416-H-005-005.

References (35)

  • K.R. Davis

    Marketing management

    (1981)
  • Dean, R (1998). Personalizing your web site. Available at...
  • R.J. Dolan et al.

    Power pricing: How managing price transforms the bottom line

    (1996)
  • D. Goldberg et al.

    Using collaborative filtering to weave an information tapestry

    Communications of the ACM

    (1992)
  • J. Han et al.

    Data mining: Concepts and techniques

    (2001)
  • Hsu, C. H (1998). Study of sales promotion models on the internet. Master Thesis, Department of Information Management,...
  • G.R. Iyer et al.

    Linking web-based segmentation to pricing tactics

    Journal of Product and Brand Management

    (2002)
  • Cited by (70)

    • The value of personalized dispatch in O2O on-demand delivery services

      2023, European Journal of Operational Research
    • Direct marketing campaigns in retail banking with the use of deep learning and random forests

      2019, Expert Systems with Applications
      Citation Excerpt :

      This task is to some extent similar to detecting the best moment for a direct marketing camapaign call. The use of historical transactional data is widely used in previous works (Changchien, Lee, & Hsu, 2004; Liao & Chen, 2004; Rossi, McCulloch, & Allenby, 1996; Weng & Liu, 2004). Using transactional data is considered to be an advantage that allows to personalize particular campaign in order to improve cross-selling and up-selling efficiency.

    • A strategy-oriented operation module for recommender systems in E-commerce

      2012, Computers and Operations Research
      Citation Excerpt :

      However, for an RS constructed from the viewpoint of system developer, issues should be considered that not only to fulfill the user's needs (preferences, budgets) but also to raise the supplier's profit. Changchien et al. discussed sales promotion based on businesses’ marketing strategies, pricing strategies, and users’ purchasing behavior, which could potentially be a win-win situation [9]. However, the study prioritized the probability of an inequitable supplier so that it may be difficult to keep a user's loyalty.

    View all citing articles on Scopus
    View full text