|
Generating TrafficAt the end of 1997, the Web supported over 175,000,000 (yes - one hundred and seventy-five MILLION) discrete WebPages (estimate by HotBot). Yahoo! estimated that by the year 2000, there will be in the region of 800 million pages. (Although, as noted above, these are, in reality, no more than "guesstimates".)It therefore becomes imperative that your site - and its content - is easy to find for your target audience. The only way visitors will find their way to your site is through linking; that is, the presentation of unique URLs which will lead potential visitors to your site. In reality, a visitor will most likely find your site by one of the following means: It is undeniable that Search Engines and Directories remain a major source of traffic to websites. By the same token, it is reasonable to assume that, as the Web grows in size, these sources will become less significant because of : Nonetheless, for the foreseeable future, both Engines and Directories will remain invaluable sources of generating traffic to your site and your job postings. What is "traffic" and how do you measure it?At its purest, "traffic" is the number of individual visitors to a site (or page) within a given period. Which sounds easy enough in theory. But when translated into practice, presents a host of virtually insurmountable problems.In common parlance, "traffic" may refer to:
"Hits" are counted by the number of individual files which are downloaded by a visitor to a site. As such, they are notoriously unreliable except as a measure of server usage. For example, a homepage may contain the
following:
A small graphic of your company's logo at the top of the page (eg "smallogo.gif"); Your homepage comprises nine unique files. A visitor to your site will therefore register nine "hits" when they download your homepage. And so on for each and every page that they visit. Measuring visitors is more problematic. Most dial-up ISPs (Internet Service Providers) assign user identities "dynamically". That is, each time a user logs on, he or she is assigned a "label" from a bank of same. For example, the appearance in one's user logs of: p0696c01.kpmg.com merely indicates that an individual from "kpmg.com" who was assigned the identity "p0696c01" visited one's site. There is no clue to who this individual was. Indeed, this entry may well appear on a future occasion - and there is no reason on earth why that unique identity should refer to the same individual (other than sheer serendipity). By the same token, an individual may well return to one's site, but be assigned a different "identity" by their ISP, although you would never know this from an examination of your user logs. As if these obstacles weren't enough, both of these admittedly crude measures are subject to the phenomenon of "caching". Here's how "caching" works. When a page is loaded, the files are downloaded to a directory on the user's computer hard-drive ("cache"). If a user revisits a page, the browser will check the page files on the Web against the "cache". If they are identical, the browser will load from cache, for the simple reason that it's quicker. (It also relieves stress on the page's server). There may be several levels of "caching" between your page and the user. i) Local site cache In the example cited above, "kpmg.com" may cache the site, which, in the absence of modifications to the page, would be loaded by another user logging on through the KPMG server. ii) Local regional cache iii) Large regional cache In none of these situations will a visitor appear
in one's user logs.
So how do you measure traffic?There's no simple way. Liken a Website, if you will, to a billboard. If it's in a popular, well-trafficked location, you know that lots of people see it, but you don't know exactly how many or who they are.Having said that, Relevant Knowledge (http://www.relevantknowledge.com/) claims that, by monitoring the surfing habits of a randomly selected "media panel that is representative of the entire Web population" it produces the most comprehensive and reliable statistics about the Web to date. This service is used is used by sites which generate huge amounts of traffic (like Yahoo!, Netscape and Microsoft), to whom accurate site statistics are vital in order to sell advertising space on their sites. In point of fact, for an averagely trafficked site which is not selling ad space, there is little need for 100% accurate statistics. You will know whether your site is being visited by the amount of feedback you receive and by a comparative examination of your user logs ("did visitors download more than last month?"). The Web, by its essentially decentralized nature, makes accurate number crunching a virtual impossibility for all but the very most sophisticated site operators. [Next - Search Engines and Directories]
795 Mammoth Road
"Less is always more."
Home
|| Articles || Background ||
Less Text at a Discount
|
