Internet is believed to be one of the most measurable mediums; however, the fact that almost everything on the internet can be measured makes understanding internet measurement all the more complex. There are several methods of measuring audiences online; however, none of these methods in isolation give one a complete understanding of online activity.
The four distinctly different types of online audience measurements available are:
1. Site-Centric, in which the website server log entries are the immediate subject of analysis;
2. Ad-Centric, in which the ad server log entries are the immediate subject of measurement;
3. Browser-Centric, in which the contents displayed in PC browser are the immediate subject of measurement;
4. User-Centric, in which the person using online media is the immediate subject of measurement. (Richard W. Goosey)
This article shall focus on Site-Centric and User-Centric, which are the most common methods followed in the industry and analyze their advantages and limitations.
Site-Centric Measurement
Site –Centric measurement is a census –like method which can measure almost every site related metrics like no. of visitors, no. of page views, clicks, time spent on the site, time spent on each page, bounce rate, etc. This kind of measurement is very useful for website publishers as it gives them a very comprehensive way to measure all the activity on their site. This method can track page by page activity of a visitor right up to the conversion page in case of an ecommerce site. This data is really useful to understand which sections of one’s website are visited most and identify sections which may not be doing very well. It is one of the most reliable and accurate methods which captures both in house and out of house usage of the internet. Capturing out of house consumption of media is generally in most other mediums.
Site-Centric measurement generally works on a cookie mechanism. The moment one visits a site, a cookie (a small file) gets downloaded on their system and tracks all their navigation on the site. This data is reported back to a server which aggregates all this information and makes it available to the advertiser or the website owner. There are several paid and non-paid third party packages that one can use to track their website data. Google Analytics is the most commonly used application which gives one a very comprehensive analysis of their website traffic and activity free of cost.
Though this method is one of the cheapest and most accurate ways to measure internet audience, it does have certain limitations. The biggest limitation of site-centric data is the fact that it does not tell us anything about the demographic profile of the user. One does not get to know the age, gender, etc. of the user whose website activity was tracked. For Advertisers, knowing the demographic profile the visitors of a website is a must. Also, this method cannot differentiate between two users working on the same machine; hence, it is believed that site-centric method measures devices and not people. Apart from there are some other limitations, like a refreshed page or aborted pages are also counted. Since the system mainly relies on the cookie mechanism, in cases where users delete their cookies often – data can get skewed.
User-Centric Measurement
User-centric measurement is the more traditional panel based method, where a sample is recruited and their internet behaviour is tracked. This system gives one detailed demographic data about the users in the sample and hence advertisers find this data very useful. Also, because the demographic data available in this method is similar to the data in most other traditional mediums it is easier to do a comparative study of media consuming habits of particular audience using this method.
In this method a panel is recruited and they are generally asked to download a tracking file on their computers. This file then helps track all the online activity of the user. Though this method too can track out of home activity of a user but this is subject to the fact that a respondent downloads the required file on their computers. There have been several cases where it has been known that users do not download the file on all the systems they use and this again brings in error in measurement. One of the advantages of this method is that it is conducted by third party and hence any biases are believed to be absent. ComScore and Neilsen /Net Ratings are some of the common bodies providing this kind of measurement data.
The biggest limitation of this method is the fact that the kind of data tracked is restricted by the size of the sample. It is expensive to recruit a large sample and hence data captured for a smaller sample is projected over a larger universe. Also, the motivation for respondents to download the file that allows tracking of their online behaviour is often questionable. Many people are unwilling to have their online usage tracked due to privacy concerns.
Way Forward
Both user-centric and site-centric measurement methods have their own advantages and disadvantages; however none of them in isolation is a complete solution. There is a need to integrate the advantages of both the methods and come up with a more comprehensive solution that will not only accurately capture all kind of online activity but will also give the publishers and advertisers rich demographic data about users. It is this comprehensive data that will help people better plan their media campaigns and will also allow for comparability of activity and behaviour across all mediums.
Several research organizations are coming together to share their expertise around digital media tracking and measurement. Efforts are also being made to link it with traditional media metrics. Hence, one expects an integrated method of measuring audiences online to soon evolve and provide publishers and advertisers with a more complete solution.





