Internet Metrics

views updated

INTERNET METRICS

In general, Internet metrics encompass a wide variety of measurements or assessments made on the Internet. These can be very broad measurements, applying to the traffic patterns and usage of the Internet as a whole; or they can apply specifically to an organization's Internet infrastructure, which might include the network connections, cables, workstation computers, servers (computers used to host Web sites, shared software applications, and e-mail systems), and e-commerce software.

On the broader scale, Internet metrics refers to any number of Internet aspects that can be measured or presented in a statistical format, including online advertising industry revenue reports and projections, trends about the preferences of Web site users, or other statistics about the Internet economy. Global Internet metrics also apply to different technical aspects of the Internet's infrastructure, including the miles of cable that connect the world's computers, routers (computers that relay information between devices on the Internet and different Internet Service providers).

In the early 2000s, the Cooperative Association for Internet Data Analysis (CAIDA) was one organization devoted to Internet metrics on a large scale. CAIDA collected, monitored, analyzed, and visualized data about the Internet in four broad areas. Topology measurements helped to reveal global characteristics about the Internet, describing the ways in which the many networks that constitute the Internet are joined together, and revealing information about the size and constituency of the Internet's core. Workload measurements monitored the distribution and flow of Internet traffic. Performance measurements provided a means for "isolating global problems within the infrastructure, as well as assessing service quality by country or other granularity of interest." Finally, routing data provided details about "relationships between individual Autonomous Systems at a given point in time."

Internet metrics also play critical roles at the organizational level. Just as companies use Internet metrics to measure, monitor, and report on their financial performance, successful ones also take steps to measure the electronic elements of their business efforts. This involves monitoring performance and statistics at the user (client) level, on the back end (different computer systems and databases that may be used during the company's e-commerce activities), and at a level in between these that includes things like network performance and servers.

To obtain metrics related specifically to Web sites, companies rely on software applications designed to provide details on a wide variety of factors, including everything from the number of hits, page views, and unique visitors a site registers within a specific time frame to click-through rates on banner advertising and information about the most-requested pages on a site. In the early 2000s, NetIQ's Web-Trends software was among the leading applications in this area, and was used by 55,000 customers worldwide, including many Fortune 500 companies. Pharmacia & Upjohn, a leading pharmaceutical company, used WebTrends to determine who was using its corporate intranet and what information was being viewed. The intranet was launched in 1996 to provide a means of communicating with more than 30,000 employees worldwide. The software helped Pharmacia & Upjohn improve the design of the intranet and to improve its performance based on usage patterns.

Metrics related to Web site traffic can be used to lure potential advertisers to a company's site, to understand how customers use a Web site, and to improve or streamline processes that make a Web site experience better. However, as with any statistic, the meaning of Internet metrics varies depending on how they are calculated, the context in which they are used, and the value or weight users assigned to them. Additionally, different metrics mean different things to different parties. At a large magazine, for example, the publication's marketers may rely on one set of figures for e-commerce purposes, while the editorial department and potential advertisers review other numbers.