Conclusions and Call for Future Research


The number of unique domain names with hosts , almost two hundred million in January of 2003 <www.isc.org/ds/>, suggests a lower estimate of the number of Web sites. Similarly, the popular search engine Google's index of over four billion Web pages in 2004 serves as a lower estimate of Web pages. These online repositories have at least two things in common. One, minute by minute a Web server is automatically logging visits and thereby creating a valuable data set for improving that site or page. Two, each site owner is a potential partner for collecting data that might be of theoretical interest to marketing academics and of practical interest to the site owner.

In our opinion, much of the current electronic marketing research attempts to recreate familiar mass media situations. For example, Internet user panels track how a particular individual navigates online, going from one site to another. Other research uses surveys to capture the attitudes of Web visitors . Such data are certainly valuable, albeit expensive.

We argue however, that most marketing researchers are missing an important Internet component, namely its direct and interactive nature. The marginal cost of acquiring and analyzing Web log data from individual sites is essentially zero. These data are valuable enough as is, describing what individuals do on that site. Combined with experimentation on the site, the data become markedly more valuable to both the practitioner and academic.

Rather than argue or guess what might work best online, practitioners , along with academics armed with the appropriate theory, could run experiments. Similarly, industry already running these experiments could discover additional insights by sharing the results with academia. There are numerous research possibilities.

At the level of individual links, experiments could test wording, the icon or image that anchors the link, text vs. image anchors, the type of appeal , and a host of other psycholinguistic and format issues. At the page level, we know little about the effectiveness of simple format variables and the appearance of pages, supplementary images and text that might provide a context for the links on the page, the position, number and type of links, and how one link might cannibalize another.

Finally, at the level of the whole site, productive research could investigate site structure, and how to best present that structure to the visitor. Issues pertaining to the former include modularization vs. larger pages, semantic distinctions to define sub-pages, the information space (hierarchical, linear, serial, factorial, or semantic network) and menu structure. Issues pertaining to the latter include navigation bar content and format, representing relationships between pages, and how visitors receive feedback on where they are, where they have been and where they can go.

Live Web or e-mail experiments can test virtually every consumer marketing theory, as well as theories in other disciplines. Possible theories could deal with exposure, perception, cognition, scent, learning, persuasion, elaboration, reasoning, memory, choice, heuristics, risk, flow, influence and compliance, motivation, culture, communication, hedonic shopping, economic rationality, pricing, effort and commitment, and judgment.

With even a modest Internet site, and a partnership with the site owner, academics can create multiple versions of the site, sit back, and let the data roll in. The winning experimental condition gives the practitioner a better site and gives the academic a valid test of the theory that suggested it.




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