• by December 7, 2000

Many of our readers have noted disparities in reported data from reputable sources and questioned the reliability. Seen from different perspectives, each has been correct, but perhaps misleading if misunderstood. Though we try to report an unbiased review of information available, we heartily endorse standards for research that will make this information even more reliable and comparable. Perhaps more than you want to know, but an important background both in developing research and internalizing the results of research.

Though conducted almost 2 years ago, the Future of Advertising Stakeholders (FAST) Summit developed Principles of Online Media Audience Measurement that have been approved by the FAST Steering Committee and the ARF (Advertising Research Foundation) Board of Directors making them guidelines for research online. Adherence to these principles, while in the best interest of the industry, is entirely voluntary

The central purpose of these principles is to encourage the development of accurate, unbiased, precise, reliable, and actionable estimates of consumers' opportunities to see (OTS) ads and other online media content. Underlying these principles is the expectation that online media measurement will quickly migrate from their current state to a state comparable to traditional media and then beyond. Technology is currently becoming available to enable the measurement of "ads sent" to "ads displayed" -- the online equivalent to OTS.

Three types of online measurement described in detail in the Summit report are:

SITE-CENTRIC: The website server log entries are the immediate subject of analysis

Most current systems can only count pages sent, not pages displayed. On the other hand, pages that are aborted or abandoned before display are presumed by the originating server to have provided an opportunity to be seen, while, in fact, they haven't. And, site-centric measurement has an inherent conflict of interest since the measurement is in the hands of the media sellers.

AD-CENTRIC: The ad server log entries are the immediate subject of measurement

Ad-centric measurement offers many of the same benefits and limitations as site-centric measurement, except that it counts ad exposures across media consistently. Some ad-centric systems have technologies designed to defeat caching, thereby providing a better count of ads delivered (OTS).

USER-CENTRIC: The person using online media is the immediate subject of measurement

User-centric measurement measures the display of online media content by a browser (OTS). However, given the increasing complexity of online content display styles, there is concern that an ad "visible" to the browser may not be visible to the user and thus not a true OTS.

The Principles of Online Media Audience Measurement are to further the quality and comparability of all online media audience measurement. The methodological principles developed in