Don't Take a Hit on Website Visits - Correct Interpretation of Website Statistics

Website Hits versus Visits - Correct Interpretation of Website Statistics

It is useful to monitor traffic to your website in order to determine the volume of traffic and which pages are most visited. Such monitoring also helps determine the effectiveness of Search Engine Optimization (SEO) as applied to the website and specific posts.

On Unix-based servers website statistics are typically collected with the Webalizer package, however a number of also packages can be used. More information on Webalizer-based reporting is available in the document How to Interpret Website Statistics.

The main determinant of website performance is how many unique visitors view the site and/or its subsidiary pages, blog posts, etc. When looking at statistics for the website home page or for any of its subsidiary pages, it is important to know what you are measuring in order not to inflate results. There are two basic measurements of traffic to a website (or web page): hits and visits. Only one of these is a viable measurement.

Website hits measurement

A unique visit to a website is sometimes incorrectly referred to as a "hit". In actuality, a hit is one thing - anything - being transferred from the website hosting server to the user's browser. Although a web page looks like a single indivisible entity in a browser, it in fact consists of numerous components, including:

• The html file for the page.
• One or more cascading style sheets (CSS) that control page presentation.
• Each individual image, photo, chart, banner artwork, icon, etc.
• Javascript file(s), if included on the page.
• Audio file(s), if included.
• Video file(s), if included.

Each one of the above items is contained in a file on the server. Each access of each file is counted as a hit. Thus, browsing a single web page can result in 5-10 hits for simple pages, or several dozen hits for more complex pages.

Hits are useful for measuring the gross impact of a website on a server, but they are not useful for measuring website traffic. Hits can not even serve as a proxy for website traffic because each site is unique and each page is unique - there is no uniform number of hits that is required to produce a web page.

Website visits measurement

A visit is one user browsing website or page in a given timeframe (typically 30 minutes).

Webalizer defines a visit as a request for a page for the first time in a pre-defined time interval (the default is 30 minutes). If additional requests are made by the same user within that time period, only a single visit is recorded. Only website page requests are considered when counting visits. CSS, images, audio, video, etc., are not counted.

So for example, if a user browses many pages on a website for 30 minutes, one visit would be recorded. If the user returns later that day and browses more pages for another 30 minutes, another visit would be recorded. (The unique user is identified by their IP number). If five people browse the website for 10 minutes, five visits would be recorded. If the same five people browse the website for one hour, ten visits would be recorded.

Although not a perfect measurement, the visits statistic results in a more accurate counting of unique visitors to a website.

Accurate website statistical reporting

Hits are sometimes used as opposed to visits to report on website traffic, for several reasons:

• A comprehensive statistical package (Webalizer) is not installed.
• Hits is a "feel good" measurement that inflates the apparent number of visits.
• Hits makes the reporting agency (website designer or search engine optimizer) look more effective.

Accurate and realistic reporting of website traffic must be based upon the visits, not the hits statistic. Use of the hits statistic will result in grossly over-reported website traffic.

For more information, see How to Interpret Website Statistics.