The different TYPES of stats packages
Log Readers vs. Integrated Tracking (Javascript/Image)
Having read about the features offered by Log Readers and Integrated Javascript/Image stats analysers, it is worthwhile looking at the pros and cons of both methods in order to track your website's stats.
Log Readers are precise ways of looking at the visitors to your website - every page, every hit is recorded in the log, without fail. As a comparison, the Integrated methods are reliant on you personally inserting their code into every webpage on your site in order to log traffic. This can be time consuming and difficult to manage - miss one page or accidentally paste the code in incorrectly, and your stats are flawed. However, assuming every page does have the integrated code in it, this method has advantage over log readers for the very same reason:
Log Readers log everything - every image access, every flash file access, every style-sheet access - everything!. So if these elements are accessed from your web server WITHOUT being viewed as part of a webpage (it happens! see hot-linking) then this will be still be recorded as a new visitor and artificially inflate visitor records. The same is true for referer data, too. With the Integrated method, you can be absolutely sure that ONLY page views are recorded since they are the only elements that the integration code can be added to!
Using the integrated javascript method, you also have the interesting effect of automatically excluding all crawler traffic. A crawler is the search engine's robotic computer program that will access your website regularly in order to update its database regarding your website's pages, their content and so on. A crawler is, therefore, not a real visitor. Log readers are usually able to differentiate crawlers from real visitors because they have a list of search engines' known IP Addresses (amongst a few other methods, but this one is of particular importance), however this is an inexact science. Apart from this, a lot of log readers do not offer you a direct way to exclude crawler traffic from your final reports and statistics. However, in comparison, crawler robots CANNOT view or execute javascript and as such will never "trigger" a visit on your tracking website. In this way you ensure that your visitor numbers are not over-inflated by robotic visitors that are of no use or consequence to you when looking at your webstats.
Some javascript methods can cause issues on very busy sites, specifically that page loads can appear to freeze for your visitor until the external javascript or image has been fully loaded. This is not always the case but it is always worth considering the impact of something like this happening to your visitors or customers.
|
|
|