Overstated visitor numbers obscuring ecommerce metrics

In a recent press release by Internet Information company comScore, it has been suggested that audience sizes could be overstated by up to 2.5 times. This is largely due to the regular deletion of cookies by around 3 out of 10 internet users.

For internet advertising, reach is a key measure which determines the value of websites or web pages. The measure of reach is unique visitors. Web analytics software like Google Analytics tracks unique visitor numbers by using cookies. The cookies are used by the analytics software to identify the site user for tracking purposes.

There is a growing trend for internet users to limit the use of cookies on their machine as modern browser technology makes it quite easy for people to do this. The problem for analytics software is that a user who visits a site regularly but also regularly clears their cookies will appear as more than one unique visitor, even though they are the same person.

The comTest study examined the degree to which users cleared cookies from their computers and found that server logs that count unique cookies to measure unique visitors are likely to be exaggerating the size of a site’s audience by a factor as high as 2.5, or an exaggeration of 150 percent.

For internet advertisers, it is in their interest to have high visitor numbers as they can demand more money for advertising space on their websites. But how does this matter to e-commerce site owners? I guess it depends on what way you choose to look at it. On the one hand, you may want to use these numbers just as a guide to benchmark performance from one month to the next. But what if you want the figures to be more realistic?

Consider your conversion rate KPI’s for example. If the number of unique visitors you’re measuring is wildly inaccurate then so are all of your KPI’s. Your current analytics is telling you you’re converting at 1%, but it could be 2% or higher? Your repeat visitor numbers are looking healthy at 10% but they could be less than 5%. See what I mean, it kind of screws things up.

At the end of the day, until analytics vendors achieve a more reliable method of measuring unique visitors, the most important thing is to compare apples with apples by choosing a system for your comparisons and sticking with it.