Understanding Web information search behavior: an exploratory model.

ABSTRACT

This paper develops a model of consumer Web search behavior. The model is derived from consumer external information search behavior models and also those relating to search behavior in physical and hypertext environments. Personal demographic and behavioral variables, which have been found to affect search behavior in conventional consumer information environments, are also discussed Some of the unique features of the Web search environment are identified and the navigational skills required of Web searching consumers are considered. The relevance of personal Web skill and experience variables is also reviewed. A taxonomy of Web search actions is proposed as the basis for recording Web search behavior The outcomes of information search behavior are discussed from decision-making and affective viewpoints. The variables discussed are then combined into a proposed model of Web search behavior and a research agenda detailed.

Keywords: Web; consumer behavior; information search; model

INTRODUCTION

Information search behavior is a key concept in the study of consumer behavior. Information search behavior has been extensively studied in traditional consumer decision-making environments. Studies include: Copeland (1917), Katona and Mueller (1955), Kiel and Layton (1981), Newman and Lockeman (1975), Punj and Staelin (1983), Srinivasan and Ratchford (1991), and Titus and Everett (1995, 1996). In addition, researchers from other disciplines have researched information search in other environments. Examples include: physical shopping environments (Titus and Everett,1995, 1996); hypertext (Conklin, 1987; Kerr, 1990; McDonald and Stevenson, 1996; Mohageg, 1992); and, Web browser usage (Catledge and Pitkow, 1995; Tauscher, 1996; Tauscher and Greenberg, 1997). A number of models of information search behavior have been formulated. Notable examples include Maute and Forester (1991), Moorthy, Ratchford and Talukdar (1997), Punj and Staelin (1983), Schmidt and Spreng (1996), Srinivasan and Ratchford (1991), and Titus and Everett (1995). The Web introduces a major new resource for consumers. It can be considered as an additional source of consumer information, a commercial transaction medium, and in some cases also a delivery channel (Hodkinson, 2001). Consequently, it is important to understand the nature of information search on the Web, the likely predictors of such information seeking and the outcomes of such information search. The objective of this paper is to develop a conceptual model of consumer Web information search behavior. To accomplish this it is necessary: (1) to understand the nature of Web information search, (2) to identify variables that may affect Web search behavior, and (3) to identify the outcomes of the Web search process.

JUSTIFICATION FOR THE MODEL

The Internet was originally conceived as mode of communication that would be robust in the event of an attack or sabotage (Schneider and Perry, 2001). Its development into the Web and its later conversion to a graphical medium in the mid-1990s extended its capabilities. It captured the public's imagination and with the availability of domestic modems and cable access, what was once the province of academic and military text-based communication transformed into a rich, if anarchic, information medium with public access. Its potential as a consumer information and sales medium led to the appearance of innovative on-line commercial ventures, many of which had synergies with the Web medium itself. The Internet was a phenomenon that truly sneaked up on an unsuspecting world (Schneider and Perry, 2001). During the early stages of commercial colonization of the Web, widely disparate estimates of the growth of Internet commerce were evident. In hindsight, many overly optimistic predictions were made for the growth of on-line sales (Claymon, 1998; Colarusso, 2000). While estimates of the current level of "e-commerce activity" are obtainable, the underlying definitions of "e-commerce" vary (Congressional Research Service, 2002; U.S. Census Bureau, 2000). Some statistics include transactions using older Electronic Data Interchange (EDI) technologies, while others do not differentiate between B2B and B2C sales. In addition, methodological differences also exist in the derivation of estimates. However, there is a consensus that B2C e-commerce is growing at a rate far below that originally predicted. In the USA retail e-commerce sales still account for less than 1.5% of total sales (U.S. Census Bureau, 2002) and the press is replete with articles, either searching for, or expounding reasons for this situation, for example, Claymon (1998), Colarusso, (2000) and Starling (2001). Despite its slow start, it is likely that B2C etailing will be a permanent feature of the consumer environment. The only uncertainties appear to be the duration of the adoption period and the extent to which marketing via the new media will eventually replace or augment consumer in-store purchasing.