Web Analytic Matt

Views? Visits? Visitors? You What…?
Matt Hopkins posted in Web Analytics on March 12th, 2007

If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!

In my current job I give demonstrations of web analytics software and part of that demo is to explain some of the terminology. One that often crops up is the difference between Views, Visits and Visitors. I would say that about 70% of people that I speak to don’t know what the difference is so I thought i’d clear it up.

When someone goes to your website and lands on your homepage they make a request to your web server for a page. Once served up this can be classed as a page ‘view’. Obviously it is possible that the same person continues to look around your website, maybe look at 5 further pages. Each time they see a page in any order, they cause a page ‘view’.

So someone has looked at 6 pages, lets say that they did this in a 10 minute duration, we can group all of those page ‘views’ and call them a ‘visit’. The web analytics industry standard says that a ‘visit’ is a collection of page ‘views’ that do not have a gap of greater than 30 minutes between them.

What this means is that someone could look at a few pages on your site then go for lunch for 31 minutes and continue looking around your site. In analytics terms this would be classed as 2 visits because they took over 30 minutes between 2 page views. Obviously this default of 30 minutes can be changed in most modern analytics applications, and you may want to do this if your site has attracts certain types of users. For instance if you have a website with lots of PDF documents which take on average an hour to read then you may wish to increase the acceptable gap between page views to 1.5 hours to account for people reading a PDF then continuing to look around your website.

And…a visitor is the user of the website who made all those visits and page views. There are a number of ways of capturing visitors accurately.

1. 1) Usernames
2. 2) Cookies
3. 3) IP address and Useragent
4. 4) IP address

When a user logs in to a website under a specific username you know with some certainty that it is a single user so this is the best visitor identification that we have.

Cookies are small files sent and stored in your web brower. 1st party cookies are good and most 3rd party cookies are rejected by modern web browsers. Cookies are not great for tracking visitors because a web browser/computer could be shared by many people (think internet cafe). Another scenario could be that someone visits a website from the office using their office PC then goes home and visits the same website from their home PC, the website would see them as two different visitors because they would have two different cookies.

IP address and useragent tries to get around the problem of multiple people using the same IP address as most people will use a slightly different variant of web browser. This is the case for consumers of popular broadband providers who all go through a few number of proxy servers.

IP address on its own is just a very poor way of tracking visitors and you shouldn’t believe your unique visitor numbers based on this methodology.

I have simplified these ways of visitor identifcation so if you have any questions then please leave a comment below.

Thanks