Dynamic Chiropractic – December 1, 1999, Vol. 17, Issue 25

Hits vs. Visits: Which Are More Accurate?

By Editorial Staff
One of the best things about the Web is its ability to provide massive amounts of information. Understanding the context of that information and its usefulness can sometimes be challenging. This is especially true with website "traffic" data.

When you visit different pages on a website, you are really requesting files from a particular computer (sometimes also called a "server") or group of computers. As the server sends you the files, it records data: your URL; how many and which files you ask for; what site you just came from; your computer's operating system; your web browser, etc. This is all recorded in huge files called "log files." (ChiroWeb's log files have grown so large that they are now kept on a separate computer. Even then, we can only keep 14 weeks of data.)

At the end of a chosen period of time, these log files are used to create a report that helps the website administrators understand how many files were downloaded, how much traffic occurred per day or hour, what files (pages) are the most popular, etc. From this, website developers have sought a meaningful number to quantify the amount of use their website has gotten.

Initially, the term "hits" (or "accesses") was used. A hit can be defined as a single file downloaded by a single URL. Unfortunately, in their effort to promote their site, many sites used the term without a clear definition of what is and isn't included in that number.

A single page that you download from a site can have multiple files. Each graphic on a page, for instance, is a separate file. This partially explains why pages with lots of graphics take longer to download than those without. (File size is another issue. Poorly designed graphics can create monstrous-sized files.)

For example, the front page of Dynamic Chiropractic On Line currently contains 29 files: one for the page itself, one for the advertising banner, and one for each graphic. So while you might view this page only once, some sites might count it as 29 hits.

There are a number of ways to make hits a more honest number and reduce downloading time as well. One way is to use "include files." These files contain graphics or content that are used more than once on the site. When you download them the first time as part of a page, they are kept and reused as they appear on other pages without duplicative downloading of that file.

On ChiroWeb, the toolbars on the left and at the bottom of most pages are "include files." You only have to download them once per session, making transmission time of most pages much quicker.

Another situation that swells the number of hits unnecessarily occurs when search engines enter a site and begin to review files. They use programs called "spiders" or "robots" that search each page to catalog the contents. Obviously, if they search a few hundred pages, this can make the number of hits jump, but they aren't reflective of real visitors.

Because of its popularity, ChiroWeb can have as many as 500 such spider searches in a given month. It's easy to see how this could bloat the numbers unless they are filtered out.

To help provide more accurate data, webserver statistics programs have been developed that will screen out the number of graphic files downloaded and search engine spider accesses. The problem then becomes how to distinguish between a site that screens out these numbers when reporting hits and one that doesn't.

In response, the internet industry has moved away from using hits as the traffic benchmark to reporting "visits." A visit is a single downloading session with a single visitor. No matter how many pages you download, until you leave that site, everything you do is counted as just one visit. Thus, even those sites that don't use include files or screen out graphic file downloads and spider searches can't inflate their numbers too much. At worst case, a site might show a few hundred extra visits from search engines that should have been screened out.

The graph on the front page of this issue relates the growth of ChiroWeb.com. In October, ChiroWeb.com enjoyed 157,454 visits, with an average of almost 5,100 visits per day. To avoid any confusion, this is the number that we will be reporting each month in our effort to fairly quantify our traffic. And while the numbers aren't in the millions (yet), this should allow you to better appreciate how many DCs are enjoying ChiroWeb's services, and how many consumers are leaning more about chiropractic on our consumer sister site, ChiroFind. ChiroFind was visited 15,225 times in October. Look for a report on ChiroFind.com next issue, in which we will examine how the Web is educating consumers about chiropractic.


To report inappropriate ads, click here.