Cybermetrics in business: methods exist for tracking various Web site statistics, but not all methods are effective. Make sure you get what you're looking for in a Web analytics tool. (Information Technology).

The Internet is only a prototype for the future Information Highway: it's constantly evolving and expanding and assuming a central role in society comparable or greater than other mass media.

A Web site is an organization's extension into the virtual world. Establishing a Web site is similar to opening an office on another continent, with technological and philosophical rules to which a company must adapt while maintaining its own culture and image. The quality of the content -- the information and services available -- determines its value within the medium. Quality and ease of use define the content's value.

Any technology should only be used if absolutely necessary, and the best technological tools are not necessarily the newest or "hottest," but those that are easily available and useful for satisfying companies' expectations.

To date, none of the traditional media have offered the possibility of targeting and quantifying a customer base so precisely as the Internet. By assembling individuals with similar interests in one virtual location, the medium enables companies to deliver their message to a more uniform target public.

Web analytics are very reliable because they are based on counters, not on the probabilistic extrapolations used by traditional media based on audience samples. However, since there is no industry standard, a company can easily falsify the statistics collected if it wants to.

Visitor traffic (a site's audience) is one of the Web's economic models in the sale of online publicity. Traffic analysis is at the heart of strategic and media planning and is of interest to advertisers, advertising departments and agencies. The value of advertising space is based on the site's visitor traffic, and this depends on the site having a reliable tracking utility.

The success of online partnerships (swapping links with complementary finns) also relies on accurate statistics to provide the number of visits generated by each partnership.

Publication of the number of "hits" means that people can compare their own site's performance with that of their competitors or partners, nationally and internationally.

Counting hits is just a primary measure, however. Companies should also check more specific success metrics, such as visitor-to-customer conversion. This method analyzes the access path of each user and generates reports with information on viewing time, the user's location, and the user's operating system. This enhances your understanding of site visitors and can be used to accurately determine the effectiveness of a promotion and eventually how to improve the conversion ratio and the return on investment for an online business. Such research answers questions like: "How many visits result in purchases?" "Which ad campaign generated the sales?" "What pay-per-click (PPC) key word generated the sales?" "Which search engine generated the sales?" "Where do the customers come from?"

Companies must optimize the return on their investment in Web initiatives by assessing the effectiveness of their marketing campaigns, content, business operations, and affiliated programs.

When companies look for analytic software for their Web sites, it's important to select services that are independent, user-friendly and reliable. These tools will become indispensable for companies and organizations that invest in the Internet and need independent, accurate information on the behaviour of surfers on their site. These tools will become an added value and an excellent way to understand Web investments.

The techniques used by some Web counters are changing, although the hosts continue to offer statistics based on the simple analysis of log files directly from the Web server.

There are two main types of Web analytics systems: static and dynamic. Static analytics analyze connection or log files. A log file is just a series of lines, each corresponding to a "hit" by a Web surfer. To be able to really use this raw data, you need special software.

But the log file alone leaves out too much information, such as surfer profiles, and an accurate page count of pages actually viewed (fully downloaded). It includes hits generated by page-indexing robots and falsifies counts by not counting pages that have been cached in the computer or have been viewed through supplier proxies.

Dynamic analytics are very different. They use tools that function with "tails" (markers) at the bottom of each Web page. This method, developed by third-party companies, is more credible and much more user-friendly. It shows real-time operations based on such variables as visitors, hits, and pages viewed.

With this second technique, when a surfer calls up a page, it triggers the display of the tail at the bottom of the page, which sends a request to the site's independent counting servers to count a viewed (fully downloaded) page. The viewed page is then included in the site statistics, even if it comes from a proxy server. In the other log file analysis technique, a page called from a proxy server, used as cache memory, is not counted.

Also, visits by robots that index pages for search engines and falsify statistics are not counted.

Some tools available using this technique provide time-based data that site managers can use to analyze surfer behaviour for each page viewed, i.e., a particular schedule, daily, weekly, monthly or periodically. The visitor's access path and the time spent viewing a Web page can also provide useful behavioural information.

Visitor analysis justifies the human and technological investments required to support a Web site. A better understanding of the site's visitors helps develop content and site structure.

Today's Web analytics tools meet internal and external objectives for companies. Internally, these tools authenticate and evaluate site performance so site content can be made more dynamic. Externally, data on visitor traffic gives credibility to a site, for the purpose of creating partnerships and attracting advertisers.

It's therefore possible to analyse partnerships with other sites, surfer profiles, visitors' provenance and geographic targeting, the most requested key words, site referrals, whether visitors' systems can handle the online applications, as well as charting the impact of email, responses from a mailing list, and referral by search engines and the validity of key words.

Interpretation of Web analytics results should be done by someone who knows the company well. It doesn't necessarily require a lot of technology but does require reflection, to develop the intellectual content of the survey and to analyse the results.

All companies need a Web site to increase their visibility, but often they don't know the real impact of the site on their respective markets. Web analytics tools are becoming essential to understand the impact and measure visibility. Today, there are many services that specialize in traffic analysis. Offers are extremely diversified, so everyone should be able to find a service to suit their requirements. u

Loic Guillard ([e-mail]) is manager of business development at Marigny Solutions Internet inc.