ツイート

このエントリーをはてなブックマークに追加

↑PAGE TOP

無駄の少ない経路にむけて

2012.11.27

“Data Driven Thinking” is written by members of the media community and contains fresh ideas on the digital revolution in media.

Today’s column is written by Jessica Sanfilippo, group media director at 360i.

Below the fold, partial ad loads, and other related (and unwanted) occurrences on the island of misfit banners are the digital equivalents to the TV commercial bathroom break. What’s the value in an ad that no one ever sees? With all of the advancements we’ve made in digital tracking technologies, we have not, as an industry, been able to assure our clients that they are paying for ads which are truly viewed by consumers. Let’s face it. We operate in a view-through attribution world, where the term “view” is loose at best. With the public release of the pilot results from the “Making Measurement Make Sense” (3MS) initiative, we finally begin to scratch the surface of a viable means of defining ad viewability, and potentially eradicating the plague of unseen banners that pervade our industry and numb our consumers.

As it turns out, the results are a bafflingly broad range of anywhere from 7.3% to 78.6% of measured (that is an important distinction – a large gap in measurability is a separate challenge) impressions in the pilot that were actually qualified as “viewable.” The variation was perceived across campaign, ad unit size, and category of site (i.e., network or publisher). Viewability is categorized, based on working definitions arrived at during prior 3MS pilots and adopted by the IAB, as an ad that is at least 50% in view for 60 seconds or more. The inconsistency alone pinpoints exactly why this is such a pivotal issue for our industry to tackle.

Even more illuminating is what was found when looking at back-end metrics. 360i participated in the study (you can view our full report here). While deeper analysis into conversion activity was understandably not included as part of the pilot program, we took a closer look at one of 360i’s participating campaigns, run exclusively on ad networks. What we found was that the network with the highest viewability (46.8%) also incurred the highest post-view conversion rate (0.0029%), and the lowest viewability performer (20.2%) had the lowest post-view conversion rate (0.0001%). This may be pure coincidence, and as viewability becomes more widely available it is strongly encouraged to perform clean A/B testing, but it may very well be a sign of correlation between viewability and conversion.

The implications are wide sweeping, but one of the more pivotal shifts we should see with the emergence of viewability is the redefining of our digital real estate. This may disrupt not only how the currency of digital media is exchanged, but also the overall experience we present to consumers. In an era of programmatic buying, where the post-view activity is king, buyers will be compelled to optimize away from those non-viewable ads. Some of 360i’s media team members are already doing so for our most response-oriented clients, using one of the handfuls of MRC-accredited viewability metric providers such as comScore and RealVu.

If more media folks mimic that pattern, the infinite universe of remnant inventory may become more finite. Sure, other optimization levers we have at our disposal are likely more aligned with bottom line performance (CPA, conversion rates, etc). However, most of them hinge on post-view activity measures as inputs. True attribution accountability compels us to ensure those inputs are valid, particularly if we are paying on any form of CPM rate model.

The balance of supply and demand could be altered if the industry is no longer able to assign value to non-viewable ad impressions. This could place digital on a much more even playing field with TV, whose upfront marketplace relies on competition for its fixed pool of inventory. Optimistically, the future outlook could reveal a digital site layout with minimized banner clutter. The concept of negotiating for above the fold placements is, in essence, a proxy for ensuring that a client is only paying for a viewed ad. Perhaps we have even unfairly punished below the fold banners. It’s not that they necessarily have any less impact when viewed, it’s that counting methodologies record an impression based on ad server calls, and below the fold is simply a risky path for a waste-averse buyer. With viewability stepping in to replace this now irrelevant system, we will change the language we use to negotiate and construct our ad buys. What may emerge are ad formats that are strategically designed to meet the 50% in-view requirement of the viewability definition, or modifications to editorial layouts and scroll functionalities that keep ads in view for longer periods of time, so that the 1 second requirement is met. Less “overmonetization” of our web pages and bottom-feeding of our banners translates to an ultimately more palatable digital experience for consumers.

Viewability is here to stay, and those who quickly adapt can only benefit. After all, our entire system of back-end digital metrics is built on the very foundation of the impression. If we get that part wrong, everything else follows suit.


  • このエントリーをはてなブックマークに追加