Every day I am confronted with representatives from websites, wether it is from sales or management, telling me how many readers they have or how many visitors their website has. I also read a lot of marketing and new media magazines and blogs and they all have one thing in common – they treat the term unique visitor as equal to the number of people visiting their site. Now that is a grave mistake and I will tell you why.
Unique Visitors, or Clients as many measurement systems will call them, are almost always based on the measurement of a certain cookie. The cookie is a small text file that is issued either by the measurement tool or by the site itself and is stored in the browser that reads the site’s measurement code. Each cookie has a unique ID, often hashed, and this ID is the actual measured unique visitor. Every pageview with a statement in the measurement system containing this ID is then aggregated to that unique visitor.
Keeping that in mind, the main problem with mixing unique visitors/clients with people is obvious. One person can operate several browsers and in doing so they actually represent several unique visitors. For example, at work I have two computers, each with at least four browsers. If I visited this blogpost from all four browsers on each computer I would be counted as eight unique visitors/clients. That is without counting the browers on my home computer, or my iPad or iPhone.
To make matters worse you have the issue of cookie deletion which makes tracking visitors over time an activity with a significant margine of error. Or to be honest, a waste of time. During a single day I can represent lots of unique visitors/clients from a single browser if my cookie is deleted. Most measurement systems compensate for this my modeling the visitor from other sources such as IP adress, environment variables from the browser or other available variables. However, this becomes more and more difficult as users from mobile devices, especially iOS devices, can change IP address by just moving a few meters and then find a better GSM signal.
My point is that we need to be more acurate in what we’re talking about. We should avoid using the term visitor or reader when talking about cookie measured sites.
I’m just sayin…
PS! The only way to measure people is by establishing panels that measures the entire internet use by pre recruited panel members. Some does this par exellance and some not.
