MEASURING THE EFFECTS OF INTERNET ADVERTISING: PART I
by Larry Coleman
I have been following the exploits of advertisers on the internet with keen interest. Initially, their reaction to the Internet was one of euphoria. After all, here was a medium capable of capturing nearly every interaction with its user. It promised to provide measurement superior to any mass media product. The consequence was an overly aggressive expectation of what and when the Internet can deliver to facilitate media planning and evaluation. In the last year or so, expectations have sobered and, perhaps as a consequence of this new-found maturity, realistic efforts have emerged with a serious focus on solving the fundamental problems. In this two-part article, I describe these efforts.
Organizations like the Internet Advertising Bureau (IAB) and the Coalition for Advertising Supported Information and Entertainment (CASIE) -- a standards organization formed by two major advertising industry associations, the AAAA and the ANA -- have been working together on laying the foundation for realistic Internet measurement standards.
Simultaneously, new technology is being developed and intellectual capital spent to define this medium and integrate its valuation within the complete media fabric. There is still a lot of work that needs to be done but for the first time, the industry appears to be developing a rhythm to which everyone can keep time.
Broadly speaking, there are three measurement methodologies that have evolved, namely, site-centric measurement, user-centric measurement and survey-centric measurement. In this article, I will cover site-centric measurement, while the remaining methodologies will be covered in Part 2 next week.
Site-centric measurement aims at measuring traffic flow at a web site, using the voluminous types and amounts of data in the server logs as the data source. The real-time collection of data makes it possible to generate many different and interesting types of reports. However, this scheme has certain limitations. For starters, it involves computers measuring computers. In other words, site centric measurement cannot identify users or qualify them demographically. This has been a major problem with this technique. Unless an individual user registers and provides accurate data -- which happens rarely -- there is no geo-demographic context for the data supplied. In addition, site-centric measurement does not record what users do between server document requests and it does not record requests from cached files on the client or the proxy server. Measurement of human behavior is generally lacking.
Another problem, which the industry is aggressively addressing, is the comparability of data collection methodologies -- how "page views" and "clicks" are measured and reported. The IAB and CASIE have developed voluntary baseline standards for Internet site-centric measurement that the industry is adopting. This body of standards, like the Internet, is designed to adjust as technology overcomes some of these problems.
In summary, site-centric measurement is an evolving methodology that the Internet is adopting as a baseline standard of measurement. It has become the de facto baseline measurement standard for the industry and all major commercial sites use this methodology as the basis of their media valuation. Companies should adopt and incorporate a standardized site-centric measurement methodology for their sites.
---
Larry Coleman can be reached via http://www.virtualgold.com
---
The concluding portion of this article will appear in the next edition of D S * .