Last week I wrote about the International
Conference on Online Media Measurement (
I-Com ). I
promised that I'd finish this up this week, so I'm writing to you from Villa Del Arcos in Cabo San Lucas, on a week's vacation with the family. We are in a wonderful two-bedroom suite
alongside the ocean. Two cruise ships have pulled into the bay this morning. But, back to work and the topic at hand, as promised.
As mentioned, a major point of discussion was the MIA
project which is the Worldwide Audience Currency System (WACS). They are talking about not a single currency but transparency between. Like $ for yen or gold for oil.
According to George
Ivie, head of the MRC, an important part of this initiative is the perspective. Ivie discussed how the perspective needs to be one of advertising orientation. This theme was echoed throughout the
conference. Much of the web research we have seen to date (comScore, NetRatings) has been site centric. Ivie and many others at the conference believe that the client, rather than server perspective,
is necessary to provide accurate advertising metrics and metrics that are as comparable as possible to other media. This is a similar approach to the one the MRC took on impression counting, not
counting the request or the send from the server, but counting the impression as far down the line as possible. Ideally, an impression should be when an ad is fully loaded on the browser.
There was also talk in the conference about the changing Web or Web 2.0. As you may be aware, the new technologies (Ajax, Flash, JSON, etc.) deliver greater interactivity within a single page. Pages
can now be entire self-contained applications. Page content can change when you click and drag or mouse-over. Flash games run in a single page of html. Because of all this, we need to move away from
page impressions and other volume metrics and find something else. This has brought into question the viability of the pageview as a metric. One suggestion is that unique users and some measure of
time spent is a possible solution.
The time-spent metric is an important issue, as it potentially the default new replacement for pageviews. One attendee debunked this in an amazingly
logical way. He opined that, in the case of an expert user who is a heavy user of a specific site, that user gets in and out of the site quite quickly and accomplishes his or her work with ease. A
less experienced user might take much more time to do the same thing. The risk then being that time spent could value occasional visitors to a site more so than the experienced ones.
An
observation separate from the conference: Take a look at Quantcast and its metric relative to the number of times a unique user comes to a site in a month. Quantcast separates it into 1x per month,
2-30x per month and 30+x per month. It stands to reason that we cannot mount an effective campaign against those who have only visited once a month. I'd like to suggest that we embrace the
Quantcast metric of 2+ times or more per month and call it "effective unique users." Use of time spent against effective unique users might be a much more relevant metric.
There
was also talk about the newer metrics companies like Quantcast tgat are attempting to measure from a census standpoint, not a sample standpoint. With these services (and there are already multiple
companies who claim this capability), there will need to be a standard established and agreed to that provides an estimation process for cookie problems, accounting for cookie deletion.
Mobile was an important topic, too. In 2007, there were 1 billion mobile phones sold in the world, while the PC number was 250 million. Active usage of mobile phones worldwide is about 2.6 billion,
with an estimated 1.2 billion with Internet connections. Of course at some point, the distinction between what is a mobile phone and what is a PC will become an increasing definitional issue. Where is
the line drawn between a fully functional PDA, iPhone and a very small computer? A topic for another day.
As with the PC-based Web metrics, there is a great need to establish standard
metrics in the mobile space. This is much more difficult than the PC world, where there are at best only 2-3 operating systems and a couple of major browser entries. In the U.S. alone there are
hundreds of different mobile configurations.
Greg Stuart, former CEO of the IAB, made a very logical presentation on this topic. He pointed out that we have a clash between media
researchers and technologists in the mobile space and that they represent very different skills that are invariably present in different people. Once again, the specter of a global tower of Babel was
presented. This, combined with multiple companies that are already providing proprietary measurement, protection of technology by the hardware and the phone companies, and the difficulty to even get
proper motivation and support, makes the mobile goal of setting up, guiding and managing the mobile industry to develop strategically advantageous global currency standards as stated by Stuart a
daunting task.
Video games are another major platform that was discussed. Once again, the fact that we are talking different research and metrics languages came into play. Mobile is facing
many of the same issues as the PC and Web. There is a need for more openness, more standards, education, and census rather than sample, which we need to measure from an advertising and user
standpoint, not site standpoint all come into play. In addition, there was the issue of frequency capping which needs specific attention in the game space due to the way that game advertising is
currently bought and sold combined with the frequency of play on the part of the consumer.
The mobile goal, as stated by Shelby Cox of EA and Ken Ripley of IGA, is to be able to book
mobile impressions like trying to fly from Memphis to New York: to go in, look at the choices and book it.
For WebTV and IPTV there was an observation that downloads are one thing, plays
are another. The panel discussing this topic wanted to see what percentage is played, when, how often, completion rates, etc...
Other issues like cross-media effectiveness, understanding of
the relationship between brand and direct response were also raised at the conference.
I'm going to wrap up this report as it is already about twice as long as the average Metrics
Insider. I feel like I have only scratched the surface of the conference, and I apologize to the speakers and topics that space did not permit me to cover. But then, as I stated last week, you really
had to be there. See you at the next I-Com.