This paper studies the relationship between Web page download time and various elements' of user satisfaction. These elements include content, format, ease of use, appeal of graphics and responsiveness. The approach that was followed to test the hypotheses was a controlled experiment where the users were provided with several Web pages. The download time of the Web pages was varied and controlled by the administrators. A questionnaire was used to assess the user's level of satisfaction. Surprisingly, delays of 15 seconds or less did not appear to impact a user's overall level of satisfaction. In the process, the paper highlights the strategic implication of corporate Web sites and makes several recommendations in that regard.
The Internet, which has been referred to as the Information Superhighway, is probably the fastest growing information infrastructure in our society. The active user population has been estimated to be 37 million in December 1998 (up from 27 million in December 1997) and is estimated to grow to 142 million by 2002 (eMarketer, 1998). According to New York-based research firm Jupiter Communications, in 1!)98, U.S. Web retailers sold $7.1 billion worth of goods and services, compared with $3 billion in 1997. Retail on-line sales are expected to reach $41.1 billion in 2002 (Hong, 1999). Business-to-business electronic commerce forecasts are even rosier. Business-to-business electronic commerce has been forecasted to grow to $140 billion in 2000 from $500 million in 1996 (eMarketer, 1998a). Due to this tremendous growth, many companies have discovered the potential of the Internet as a marketing tool. As a result, the variables that determine Web-user satisfaction have become a very important area of research. Among the most important variables that have been identified are Web page download time, appeal of graphics, ease of use, format and content of Web pages.
This paper will focus on the effect of Web page responsiveness (download time) on overall user satisfaction. Excessive download times may create consumer frustration leading to cancellation of a Web page load and perhaps alternatively exploring a competitor's Web site. In a recent survey conducted by the Graphics, Visualization and Usability Center at Georgia Tech (GVU, 1998), more than half of the respondents (53%) indicated that while shopping, they left a Web site that was downloading too slowly in order to visit some other site.
It is well known that limited bandwidth is one of the primary reasons that Web pages can take a long time to download. Graphics, animations, frames, sound, video and other multimedia effects can add to the appeal of a Web page; unfortunately they can also add to the time it takes to download and view the page. There is an inherent trade-off between many of these multimedia effects and Web page responsiveness.
Currently the majority of home computers are using modems over telephone lines and are thus limited to 56.6 kilobits per second (Kbps). Even with new Internet connection technologies, download speeds may be limited by Internet backbone bandwidth. Keynote Systems (1998) performed a study of Web page...