Psst … Did you know that last month my web site had 5000 hits on it? Sounds impressive doesn’t it? Well if they calculate hits, how I calculate hits, most likely it is not that impressive.
One problem with online advertising is that there is still some confusion, and no thanks to traditional media sources, regarding the proper metrics to use when explaining the value of an online property. This article is not about discounting the term “hits”, rather putting it into perspective in terms of what it actually means to your online business and how to use the information.
Before jumping to that, let me explain what the difference is between the terms.
According to Wikipedia, in web analytics, a hit is any request for a file from a web server. By request this means a hit calculates the page of content delivered, all the images delivered to complete that page and any additional files that need to be loaded to make the web page you are looking at, look the way it does.
For example, my home page (www.darbysieben.com) upon visiting this 1 page, you actually generated 12 hits. 1 page view hit (index.php), 9 images hits (located in the middle of the site), 1 background image hit (notice the leafs) and 1 style sheet hit (makes the text on my site look the way it does).
You also generated a couple of hits for The Weather Network (notice the weather information on my site) as well as 3 hits to Google (notice the ads).
So, depending the type of site, 1000 hits might actually not really mean much. A graphic intensive web site is going to generate a lot of hits.
A better metric is Page Views.
According to Wikipedia, a Page view is generally defined as a request to load a single page of an Internet site. On the World Wide Web a page request would result from a web surfer clicking on a link on another HTML page pointing to the page in question.
Going back to my example from above. When you visit my home page you generated 1 page view. Because you wanted to read the rest of this article, you generated another page view for a total of 2 page views.
Page #1 – www.darbysieben.com (index.php) my home page
Generated a total of 12 hits
Page #2 – this page
Generated a total of 2 more hits (the page load as well as the “, which is actually an image)
Therefore your stats for visiting my site would be:
2 Page Views
As you probably have guessed a unique visitor is access from a single IP to a web server whom generates page views and hits during a particular visit. So continuing the example, your stats for visiting my site would be:
1 unique visitor
2 page views
Although not a 100% fool-proof method of tracking, the unique visitors metric has limitations. However, that conversation can become quite long and technical, so let’s just say this … depending on your web server, analytic software for interpreting your server log files, you can get a fairly accurate picture of the number of unique visitors. Keep in mind that limitations can come from multiple computers sharing the same IP address, server tracking issues, or a variety of other complications.
So back to which metric you should use?
For pure calculations of where your site is growing or not, you can actually use all methods, provided you have not made any changes to your site over the timeframe in question. Hypothetically speaking, let’s assume a web site in November has the following stats:
100 unique visitors
200 page views
And then in December, there stats showed the following:
200 unique visitors
300 page views
Provided no new content added/removed from the site, regardless of which metric you used, all would indicate that a site is growing.
However, on its own, “hits” tells me nothing. If I loaded 100 images on a page and told you I had 1000 hits, what that really means is the page was loaded 10 times. If I load up the page with 200 images and then tell you I had 1000 hits, it means I had 5 page loads.
Hits is only a good measurement tool when comparing 2 separate time frames of a site, with no content changes as to whether it is growing in popularity. For sites with a lot of images, hits can be used to estimate bandwidth charges.
The better metric to use is page views. I prefer this one, because it gives me a lot of information. For example, if your site has a total of 10 pages that can be accessed and your stats for a given time frame showed 1000 page views, what this tells me is this:
At the worst, you had 1000 visitors to site who looked at 1 page each.
At the best, you had 100 visitors to your site who looked at all 10 pages.
Because of caching, once you visit a web page and then come back to you during the same visit, you are not loading that page from the server again, you are loading it from your local machine. As always there are exceptions, however, 9 times out of 10 this calculation will work.
One its own, page views can provide a lot of information, provided there are no hidden or non-public accessed page. A site that had a public front-end a back-end used by employees under the same URL can really skew the data, so the assumption is that all pages are public.
This measurement tool can be used to tell us a number of things:
1) Is the site growing in popularity?
2) What individual pages of content are viewed? (once you collect page views, you also collect the specific pages accessed)
3) How sticky is our site? By sticky I mean are people visiting our home page and leaving, or are they sticking around to view more pages.
4) What paths did they use? By path, I mean what percentage went from the home page to the products page, etc.
This data can be used to make specific decisions about the performance of your site as well as how to continue to improve your site.
Finally, unique visitors gives us more insight into page views. Using the above example, if our stats told us we have 200 unique visitors, then our median average of page views would be 5, because we generated 1000 page views. This means that for every visitor who left from our home page, another visitor looked at all our pages. The next question I would ask is why did the person leave after visiting our home page, was it because our site design, its content, they thought we are somebody else, etc?
On its own, unique visitors tells us whether our site is growing in popularity.
In conclusion, the most important metric is Page Views. This actually gives us a lot more information as a stand-alone statistic. Second I would add in unique visitors, because this helps me use page views to better unerstand visitor patterns. Hits I only use on sites concerned with bandwidth usage or simply as a gauge to whether a site is growing.
So the next time somebody tried to impress you with the number of “hits” to their website, say great, but how many page views did your site generate?