When you consider the reality, a circus is actually a tightly organized, colorful and poetic mosaic of movement, served up in digestible pieces that stir the senses. A “well run circus” (almost qualifies as an oxymoron, doesn’t it?) really means that the circus producers REALLY have their act together. What other kind of organization has so many moving parts, with live animals no less, well timed, where every sliver of the event is produced with certainty and great fanfare.
Though to the uninitiated, describing last week’s ARF Online Reach and Frequency committee meeting as a circus might be perceived as disapproving, in reality it’s used here to describe the proceedings with much respect. It was entertaining, informative, larger than life and very well orchestrated.
The other element about this meeting, which made it so counter-intuitive, is that the topic was focusing largely on media research. Normally, media research is not covered with as much scrutiny or high drama. When we think of research, we normally think of the small type methodology in the back of a database description, which most people gloss over.
So what made this event so provocative? This was the meeting where members of the ARF, clearly the smartest brains in the media business, met to hear solutions from the leading online media research and server companies designed to measure Reach and Frequency. A meeting so important that it was specifically announced only to a small group of people in the know, and had an invitation-only audience list.
I was allowed to sit in because my agency is a member, plus I was invited by the chairman of the committee, Dave Smith of Mediasmith Inc. to attend. He and I are working to keep attention on learnings that are produced relative to Reach, Frequency and GRPs. We both believe these metrics are going to help improve the current online business condition by reducing the confusion and hurdles around how to measure the Internet.
The meeting began and online media-related presentations were made by Jupiter Media Metrix, AtlasDMT, IMS Media Plan, DoubleClick, comScore and Telmar. Nielsen was conspicuously absent, though their announcement about measuring online usage in the TV sweeps periods just the previous week generated much buzz. Much of the focus was on the differences between the "currency" for buying and selling online media, which is Server-Centric Measurement such as from Atlas and DoubleClick vs. User-Centric Measurement from comScore and Jupiter Media Metrix.
There are four factors that go into any reach and frequency analysis, which impacts the outcome. They are: (1) the media vehicle audience, (2) the probability of audience duplication, which impacts the frequency distribution, (3) the actual media schedule/buy to be analyzed, and (4) the marketer’s target audience definition.
The “death defying moment” of the meeting came at the end when the ARF had to vote together about the whether to back one methodology over another. Here were experienced research minds, many insanely bright people who could juggle concepts easily and cut through the data to get to the essential principles, who were being asked to consider if one process was superior to the other. My sense was that few in the room would be motivated to pick one vs. the other based on political issues. Their intellectual prowess and objectivity would prevent them from being swayed by the latest fashionable thinking.
The consensus was that neither is better vs. the other, which signified a breakthrough for the online media business long term. The group unanimously agreed that the best way to measure reach and frequency is through a combination of both server side and user side data - hardly an easy proposition or message to relay back to an advertising industry fed on a daily dose of oversimplification, but for me it was another small indication that the lion is on its way to being tamed.
The ARF membership vote in that room essentially signified that a combination of research processes reflects the best measure of reach and frequency for the online field. Inherent in this statement is that the vast majority of offline research processes is only measured by half the requirement. If online research is best measured with both processes, at some point the ARF will surely raise the issue on offline research, which has escaped greater scrutiny for too many years.
One of the other learnings confirmed by all parties, was that the 80/20 rule applies to Internet usage, same as it does to traditional media. In essence, 20% of the online audience is largely responsible for 80% of the page views. Given tools like frequency capping, the sheer number of different sites and different kind of applications, one might assume that online advertising impressions would be more dispersed.
There will certainly be much more reported on this important topic. We’ve only just begun to formally create the context to compare online and offline in an apples to apples kind of way. However, from my position in the peanut gallery, we have just laid the groundwork for the research industry to confirm what most of us already know, that the Internet is the greatest (media) show on earth!