• by Josh Chasin , Op-Ed Contributor, March 24, 2010

Last week I had the good fortune of attending the I-COM Global Summit in Estoril, a lovely coastal town outside of Lisbon, Portugal. I-COM, the International Conference on Online Media Measurement, is " an industry backed global forum for exploring measurement issues facing the Digital Media Industry, with the goals of international cooperation and understanding, information sharing and working toward consensus on Best Practices." That's what they say. Here's what I say: I-COM is a forum for bringing together global online metrics providers, as well as their constituencies (primarily agencies and publishers), in addition to ancillary interested parties (e.g., measurement auditors) to share information, learn about best practices, debate issues, do business -- and, perhaps most important, become a community.

Let me briefly hit a couple of key takeaway points.

First off, the worlds of Web analytics and audience measurement will continue to converge; one speaker postulated that by 2012 (assuming, I gather, the Mayans were wrong about the end of the world) all Web audience measurement systems will include a census component. There are several different iterations of hybrid audience measurement around the world; of course I'm partial to comScore's Unified Digital Measurement implementation, but the global, communal groundswell for such integrated, unified solutions was heartening.

EMarketer's Geoff Ramsay and our own Magid Abraham kicked off the conference proper with what was supposed to be a point/counterpoint set of keynotes. But when Geoff announced, "Finally we'll get over this scourge, the click!" I was pretty sure there would be violent agreement. Magid gave us the bad news and the good news on clicking. The bad news is, almost no one clicks on display ads. The good news is, most advertisers don't want the people who do.

Magid also pointed out an alarming but compelling point about digital display campaigns: that in many campaigns today, as many as 60% of the impressions are served to persons exposed to the campaign 10 or more times. By any definition, this is wasteful. Of course, one of the problems with cookie-based targeting (say it with me now) is that cookies are not people, and people delete cookies. And, the ad servers can't distinguish between a browser that hasn't yet received an ad and a browser whose ad-serving cookie was just deleted.

There was a surprising amount of talk about the GRP as a metric, with the usual two sides: "We need to know how many people saw the campaign, and how often," versus, "We have to get past the GRP." About which, allow me to say, "Hold on -- you're both right!" Microsoft's Beth Uyenco said she doesn't give a -- let's just say "hoot" -- about frequency, but reach will always matter. I say, you will always need to understand reach and you always know how many impressions you bought. As long as you can know the population and multiply by 100, you've got GRPs. But, there are a plethora of additional relevant metrics for ad and campaign evaluation.

GRPs are a measure of quantity that will not go away, but which, I believe, will be supplemented with newer metrics associated with engagement, effectiveness, and ROI.

Not surprisingly, everyone wants three-screen measurement. Thus far, perhaps our eyes are bigger than our stomachs. But soon, soon...

Social media monitoring, or customer listening, is clearly a dynamic and blossoming component of the online metrics space, with several companies doing interesting stuff. Personally I was intrigued by Attentio, but that may be because CEO Amaia Lasa was so charming (am I allowed to use my Mediapost column to flirt?).

I tend to think that this space -- let's just call it customer listening -- is a key plank in 21^st century market research. Once you can combine fluctuations in intensity, directionality and tonality of customer dialogue about a brand with empirical exposure to marketing variables (e.g., advertising), then you have a new response mechanism for evaluating the effectiveness of those marketing variables (that is, tracking the impact of advertising on customer dialogue.). And when you can associate movements in customer dialogue with movement at the cash register, you've got ROI. Assuming, of course, that the dialogue is a positive one!

Conversely, I heard a lot of excitement about data, data, data! And, specifically, about the potential of using all this data for targeting. Once, we media researchers saw data as a ubiquitous commodity, to which we added value by turning it into information, then insight. But the new generation of quant seems to revel in the abundance of data, sometimes to its own end, and we are increasingly seeing advertising and targeting systems that rely almost exclusively on the clever deployment of data.

I don't mean to dismiss that; the Google juggernaut is after all the clever deployment of data. But targeting data, especially when tied to a cookie, is not necessarily transformed into insight. As much as we'd like to turn advertising into a science, it remains part science, part craft, and part art. ARS has repeatedly found that the creative execution is four times as important as the media spend in determining the effectiveness of an ad. At I-COM I couldn't help thinking (and, full disclosure, tweeting) that without the right creative execution, all that targeting science is just the sound of one hand clapping.

On the Site Centric panel moderated by ABCe's Richard Foan (which, with seven panelists and a moderator, was less like a panel than a committee) there was some talk about the confusion arising from conflicting data sources. There seems to be some sentiment that elimination of multiple data sources provides clarity, but I've never believed that to be the case. Just ask the U.S. TV networks; when you're stuck with one data source, all you've done is eliminated any opportunity for validation. You didn't make that solitary data source any better or closer to the truth. All you've done is guaranteed that you're stuck with it.

On Friday, during lunch, industry groups from four countries announced Joint Industry Committee (JIC) tender offers. Consider the implications here: At one event, the industry currency audience measurement business in four different international markets was essentially put up for bid. The four countries were the Czech Republic; Slovakia; the Netherlands; and South Africa (although that was positioned as a "Strategic Review.") Among us online metrics providers, this was exciting news.

But all this, I think, was secondary to the opportunities to convene with peers, colleagues, clients, prospects, and old friends and new. In 2008, DoubleClick's uke-toting Rick Bruner and former Online Metric Insider Dave Smith both mildly chastised me for missing I-COM. So now I'm paying it forward. Really good event, great turnout, great locale, great business. If you or your company want to be a part of the conversation about online audience measurement, and if you want to spend a few days in a beautiful place rubbing elbows with online metrics nerds just like you, don't miss the next I-COM.