The 3MS initiative is comprised of the Interactive Advertising Bureau (IAB), the Association of National Advertisers (ANA) and the American Association of Advertising Agencies (“4A’s”). Under this initiative, one of the key principles of digital measurement is to move to a “viewable impressions” standard and count real exposures online.
Currently, served impressions are recorded by ad servers (either publisher-side or third-party servers such as DART, Atlas or MediaMind) at the point at which an ad is displayed on a Web page. However, ad units are often served in spaces on the page that are not viewable to the user or do not fully load on the screen, potentially resulting in over-counting of viewed impressions. Ultimately, the 3MS initiative will strive to develop an industry standard around the counting and valuation of each impression served.
The preliminary definition of the viewable impression focuses on two factors: area and duration. The minimum area of the ad that must be in view is 50%, while the minimum duration of the view is 1 second (see Figure 1). During the first half of 2012, 3MS will be testing viewable impressions in a “pilot” program, with the intent of rolling out the soft and full launch in the last half of the year.
Figure 1. The spaces within the black boxes represent the user's view (above the fold). The ads within the red boxes indicate the impressions that would be counted in each example.
According to a recent study from comScore, 31% of served and measured impressions in the study were not considered viewable based on the criteria of area and duration. By current “served impression” standards, any impressions within that 31% would be counted and would receive credit for serving on the page.
When accounting for post-impression conversions, this becomes even more problematic, as the ad server will cookie the user – regardless of whether he or she saw the ad – and then give credit to the resulting latent conversion. To this end, we believe that ad networks and exchanges that manage performance-based buys will likely experience the greate st impact, as the ad placements are often remnant positions appearing below the fold. Additionally, we anticipate that the true visibility on long-tail sites would decrease, as users rapidly click through pages during their daily online navigation. But under the 3MS initiative, networks, exchanges, and other publishers will need to work to ensure that their served ads appear in viewable areas against our target audience to receive credit.
In addition to direct-response campaigns, we expect to see an impact on campaigns that are designed around exposure (reach and frequency) and awareness metrics. Currently, standard practice is to report on total impressions served, but in reality, to make a statement such as, “we served 5 million impressions last week” is misleading. While technically the ad was served 5 million times, this does not necessarily mean that all of these impressions were seen by the target audience. Although reporting on viewable impressions will reduce the number of exposures, it could help to provide more context around the true delivery of the campaign as well as the likelihood that someone saw a display ad and that it had an impact.
Of course, there are drawbacks to the 3MS viewable impressions initiative, from both a publisher and advertiser standpoint. From a publisher standpoint, this may result in less sell-able inventory and increased demand for viewable impressions. However, from an advertiser point of view, the impact would be an increase in CPM for those impressions that can be confirmed as in-view.
From a user perspective, we may see some evolution of how publishers redesign their sites to accommodate more above-the-fold ads (impacting user experience, given the potential for extremely cluttered Web pages). With this, we may also see the expansion of “fixed” placements (such as a branded “wallpaper”) that remain visible whether the user scrolls down the page or not.
Finally, this would be a major shift in the way we’ve measured impressions to date, meaning historical comparisons will be difficult. Advertisers would need to adjust their reporting and measurements to properly compare results, for example, from one year (using “served” impressions) to another (using “viewable” impressions). This poses a challenge in properly analyzing results, not only for impressions delivered, but for metrics such as post-impression conversions or even click-through rates (there’s a major difference in the click-through rate against “viewable” impressions versus “served” impressions – likely, the former would be higher).
In the digital space, we have the ability to be smarter and more accurate in the way we record impression delivery, and therefore, the push for viewable impressions makes sense for several reasons. It provides a way to ensure that our media budgets are spent more effectively, to get the most out of publisher optimization and media placement, and to bring more value to the “impression” metric.
This initiative encourages publishers to work harder for our clients’ display media programs. Whether we’re dealing with a brand campaign or a direct-response campaign, having the ability to validate where on the Web the client’s ad appeared would move us toward better placement control, verification and optimization.
While still in the infancy of development, the viewable impressions initiative will certainly present new opportunities for accountable media buying as well as challenges and learning experiences for industry development. We will continue to monitor the progress on this initiative, and also on the pilot program later in 2012, and we will evaluate specific impacts on our media buying approaches as further developments are made.
Questions regarding the 3MS initiative, the viewable impressions principle, and Rosetta’s POV can be directed to Mike Pocci, Digital Media Manager, Rosetta, firstname.lastname@example.org.