Introduction


Industry (Newell, 2000) and academics (Hoffman & Novak, 1996; Kotler, Jain, & Maesincee, 2002) argue that interactivity makes the Internet unique and more powerful than the prewired mass media. This interactivity extends beyond consumer interaction with Web sites and incoming e-mail, to include marketers interacting with the firm's Web site and outgoing e-mail. The Internet gives marketers direct and instant feedback on tactical decisions for their Web site and outgoing e-mail. Unlike most marketing, Internet marketing offers a fast-paced cycle of optimizing based on objective feedback.

Direct Internet connections, especially Web and e-mail, support this fast-paced cycle and contrast markedly with traditional mass media. Internet channels flow directly between the marketer and the audience - without third party intermediation from a broadcaster , network, entertainer, publisher or billboard owner. In addition, this software-based channel can track and store communication with each consumer. In stark contrast, marketers can attempt to track what happens when a particular mass media advertisement travels towards a particular consumer, but it is an extremely expensive process.

Similarly, Internet sales channels can flow directly between marketers and consumers without intermediary wholesalers or retailers (Alba et al., 1997). In theory, manufacturers can track the activities of individual consumers, but once again, it is an expensive process. In contrast, Web servers automatically and inexpensively track each encounter, even encounters that do not end in a sale. Figure 11-1 illustrates the typical information contained in a server log file.

click to expand
Figure 11-1: Sample log file entry

The direct Internet marketer-to-consumer connection parallels direct marketer-to-consumer connections (Hofacker & Murphy, 1998; Marinova, Murphy, & Massey, 2002; Novak & Hoffman, 1997; Tizende, Smith, & Murphy, 2002). The direct marketing field has long noted the ease that one can test everything from envelope formatting to offer pricing (Kotler, 2003). The Internet offers millions of opportunities to do essentially the same thing, along with added benefits such as automatic data entry and real-time feedback.

Server log data provide this instant, direct and automatic Internet feedback. While authors promote mining huge quantities of online data (Cooper & Giuffrida, 2000 ; Newell, 2000; Scharl, 2000), log data are also invaluable for fine grained analysis with little data and a priori hypotheses (Burton & Walther, 2001; Hofacker & Murphy, 1998, 2000; Marinova et al., 2002; Murphy, 1999; Murphy & Hofacker, 2003; Murphy, Hofacker, & Bennett, 2001; Tizende et al., 2002). Despite the potential of log data though, most marketers are content with a cursory glance at simple canned reports of their Web site activity (Sterne, 2002).

Several technical hurdles, albeit surmountable, mystify practitioners and academics and thereby limit Web log data analysis. This is unfortunate for practitioners because their online efforts are far less efficient. For academics, even small Web sites can generate large amounts of rich data in short periods. Emerging digital and interactive applications, such as mobile phones (Barwise & Strong, 2002; Newell & Lemon, 2001) and interactive TV, highlight the importance of analyzing these new interactive and digital data sources.

This chapter de-mystifies Web site log files and calls for academics and practitioners to perform experimental research using this data. The chapter begins by placing online marketing and log data into context with four classifications of online marketing: communicating, selling, providing content and providing a service (Hofacker, 2001).




Contemporary Research in E-marketing (Vol. 1)
Agility and Discipline Made Easy: Practices from OpenUP and RUP
ISBN: B004V9MS42
EAN: 2147483647
Year: 2003
Pages: 164

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net