Re-Fact-Oring Viewability Measurement

Challenges in the measurement of viewability continue to be ironed out.  The MRC is well underway with the analysis of campaign-level data for billions of impressions in the current round of reconciliation work.  The IAB and its members are working together to develop educational tools and harness the accumulated knowledge from the publisher community about viewability implementation. And, to try to ensure that the meaning and value of viewability are clearly understood by both buy and sell-side stakeholders, the 3MS team is developing more ecosystem-wide educational programs.

Nearly every day, the same questions arise about why viewability matters -- and even what it is.  Despite the manifold efforts to explain, and the unmeasured though undoubtedly high reach and frequency of trade press articles, blog posts, social media posts, real-world conversations, panels, and outright disputes, there is a reservoir of misunderstanding and misinformation.  Among my favorites are:

1. The industry standard is not enough.  

Fact #1: The standard is sufficient for what it is intended to measure.  That is the “opportunity to see.”  It is backed by empirical science.  The MRC has presented and written about the underlying work that found that in 80% of cases when an impression meets the requirements for viewability, the ad fully renders.

Fact #2: If your creative is unappealing and/or not effective, no amount of being viewable will help the brand.  

2. Viewability is about being viewed.

Fact #1: Viewability is about a digital video or display ad rendering on the screen, and providing an “opportunity to see.”  

Fact#2:  This is a more precise level of accountability for the measurement and payment of an individual ad than in other media.  Even TV, which has not had a problem providing an “opportunity to see,” has commercial pod or average commercial minute ratings as currency, not individual commercial ratings.

3. GRPs measure quality.

Fact #1:
GRPs have nothing to do with quality.  GRPs are solely a way of counting exposures.

Fact #2:
There are many -- perhaps too many -- metrics that are being used to measure engagement.

Fact #3:
Engagement may not be the only or even the best construct for a set of variables and metrics that define advertising impact.  

4. Viewability in and of itself is the end game.

Fact #1:
Under the auspices of 3MS, a new definition of engagement was published by a task force of publishers and agencies in a paper entitled “The Advertising Engagement Spectrum: Defining and Measuring Digital Ad Engagement in a Cross-Platform World.”

Fact #2:
Without a comparable core currency unit for exposure measurement, R+F/GRPs and effectiveness metrics cannot be standardized.

5. MRC accreditation for viewability vendors holds for mobile.

Fact #1:
As yet, there is neither a standard for mobility viewability measurement, nor is there even one single accredited vendor.

Fact #2:  The MRC issued Mobile Viewability Measurement Interim Guidance on May 4, 2015.  

Yes, we have issued, stated, written about these items before.  That’s the point -- in order to  break through the clutter on everyone’s computer and calendar and solidify what is known, we recognize that we must reiterate the facts on a regular basis until they become rote.  

It is worth noting that the progress that is occurring is palpable. Across the ecosystem there is consensus that for brand advertising, viewability is table stakes, and that we must -- and will -- continue to move forward.

8 comments about "Re-Fact-Oring Viewability Measurement".
Check to receive email when comments are posted.
  1. Douglas Ferguson from College of Charleston, June 10, 2015 at 2:33 p.m.

    If a standard TV commercial appears on my live TV, it only takes a second or two for me to make a decision whether I should take advantage of the "opportunity to see" it, regardless of whether I dash to the kitchen or bury my head in my email until the commercials blow over. Advertisers receive no more of my attention (no more viewability) from an ignored 30-second full rendering than they do from an unwatched "opportunity to see" on some device. But maybe they feel comforted that with regular TV they have plausible deniability that no one saw the ad, simply because someone might have watched. Not in my house. And on my computer, I have two screens or at least two windows. Even if I can't skip a roll-in on YouTube, I can pop over a more entertaining screen until it's safe to watch the content I sought in the first place.

  2. Cary McGuckin from ServiceMaster, June 10, 2015 at 2:48 p.m.

    Surprising there is any confusion about this metric - the opportunity to be seen.  Did your ad ever have a chance.  It is a good metric that helps determine amount of totally wasted spend and I applaud the industry for adding this level of transparency - those who are opposed, are probably with reasons beneficial only to them.  The best creative and perfect targeting won't ever work if the ad isn't even visible.

  3. Ed Papazian from Media Dynamics Inc, June 10, 2015 at 3:54 p.m.

    Can't argue with that, Gary but wouldn't it be even better----especially for a TV-style storytelling commercial--- if the whole message could be seen from start to finish. Displaying disjointed parts of such a commercial on the screen for a few seconds does not give the advertiser an opportunity to have a cohesive message seen----even if the potential viewer was inclined to watch. This "standard" is simply not comparable to what has always been the acceptable norm in TV, radio and print media. Indeed, it falls far short of that and advertisers are idiots if they pay equally for each "impression" that qualifies under this definition of "viewability".

  4. Ed Papazian from Media Dynamics Inc, June 10, 2015 at 3:56 p.m.

    Ooops! Sorry for the typo. It's Cary, not Gary.

  5. Ed Papazian from Media Dynamics Inc, June 10, 2015 at 4:14 p.m.

    @Douglas, it's very true that an advertiser gets virtually no value when a viewer mentally tunes out one of his commercials after noting it for a few seconds. I often do this myself as, I'm sure, is the case with all of us. However, with digital, the user is denied even the opportunity to ignore the commercial about half of the time and it's a certainty that no matter how awful or irrelevant the ad is, that some of those who were denied the opportunity, would have watched it play out in its entirety.

    The term "opportunity to see" should be redefined, depending on the type of ad. If it's a static banner, then, perhaps being "visible" for only a few seconds is appropriate. But if it's a TV-style presentation, where much of the impact is lost if not seen from start to finish, then we need a "standard" that reflects this---like 90-100% "visibility" ----or the seller doesn't get paid.

  6. Cary McGuckin from ServiceMaster, June 10, 2015 at 6:11 p.m.

    @Ed in the world of online people have a higher expectation of being able to choose whether or not to engage with an ad, or skip it.  therefore a measure of "view to completion" rate is more the metric I think you are referring to, and in my opinion that moves more into the measurement of the success of your creative, rather than if someone had an opportunity to be exposed to it.  but these types of pay models certianly exist,  YouTube has for years now offered "TruView" ad formats, where you only pay IF someone watches more than 30 seconds.  But to me, and I am not a publisher, that puts more risk on the publisher for how engaging your content is.  In a similar vein, people have the freedom to change the channel the moment your commercial started playing, yet you still pay for that.

  7. Ed Papazian from Media Dynamics Inc, June 10, 2015 at 7:40 p.m.

    Cary, I get what you are saying, however there is plenty of evidence that TV viewers are very selective about which commercials they pay attention to----just as is probably the case with digital. For example, commercial recall studies invariably show that some messages in a break will be recalled---with some degree of proof----by as much as 50-60% of those who had an opportunity to watch while others perform at levels only a third as high. The variations are mainly due to choice of subject matter and, not surprisingly, to commercial execution, the relevance of the claims, etc. Where I part company with those who appear to be defending the proposed 50%/2- second in-view "standard" for digital is its lack of validity for TV-style ad messages. I believe that digital video advertisers who, in effect, are trying to present TV commercials to consumers, should hold sellers to a tougher standard than having their ads on the screen for at least two seconds.

    You mention YouTube's "TruView" ad format, which is comparable to TV as the audience has the same chance to see the entire message, as putting more "risk" on the publisher. I dont buy that. When a video commercial is "viewable" for at least 30 seconds, that should be all that a branding advertiser using a "15" or "30" should want. After all, we dont know if the digital ad is being "watched"---just as we dont know whether a TV commercial is being "watched". We do know that both had exactly the same opportunity to be "watched", however, which goes directly to the comparability issue.

  8. Joshua Chasin from VideoAmp, June 12, 2015 at 12:11 p.m.

    Sherrill's done a great job here of providing some demystification around Viewability.

    I think there are two things to keep in mind: first, that Viewability is designed to get digital on an equal footing with other media, and especially TV, for valuation; and second, that Viewability is "table stakes," not the ultimate measure of impression value or quality.

    In TV, we understand that when you prchase an ad, it makes it to the screen. That doesn't guarantee that it was seen or noticed, or even that there was anyone in the room; but if an advertiser buys a network spot and for some reason that spot doesn't clear in Des Moines-- i.e. it didn't make it to the screen there-- then they don't have to pay for Des Moines. In the chain of advertising metrics, it might be instructive to think of viewability as akin to network clearance data. On the Internet, where ads are delivered one at a time instead of via a one-to-many "broadcast" model, viewabiloity mesuement is essentially providing advertisers with that same service.

    As such, the point of viewability is to establish the threshold-- really, the minimum threshold-- for impressions against which publishers should be paid, This is why the definitions are what they are-- 50% of pixels, 30% for a large ad; and 1 second for display, 2 seconds for video. If advertisers and agencies want to negotiate payment against different criteria, or to differentially value impressions based on time on screen or measured engagement with the ad-- that's entirely in play, and nothing about viewbility precludes any of that.

    (By the way, I know Sunday content in June sucks, but Sherrill and I, as well as George Ivie, Bob Liodice of the ANA, and others will be at the ARF Audience Measurement conference this Sunday beginnig at 2, talking about Viewability, fraud, and other important issues. If you're bored with the beach or sick of the family, come on over.)

Next story loading loading..