By Brian Correa
Viewability is a commonly cited KPI in ad tech that is responsible for significant shifts in advertising dollars toward platforms that optimize toward viewable impressions. In previous Lift Letters we discussed viewability within the context of demands from marketers (e.g. More from Marc Pritchard, P&G's CMO). This letter will define viewability in market today, why ad tech cares deeply about this metric, and how TripleLift thinks about viewability with respect to our clients on both sides of the supply chain.
The Media Rating Council (MRC) defines a display impression as viewable if 50% of the ad appears on the screen for at least 1 continuous second, while a video impression is deemed viewable if 50% of the video player is in-view for at least 2 continuous seconds. GroupM touts stricter standards for viewability, with support from major brands like Unilever, Shell, and Volvo, where a display ad must be 100% in-view for at least 1 second and video ads must be 100% in-view for at least half the length of the ad.
As the industry continues its crackdown on fraud, viewability is often lumped into the discussion along with invalid traffic and price transparency. Depending on who you ask, roughly 50% of digital ads are never seen by a user -- a claim industry leaders use to suggest half of digital ad spend is wasted dollars. These concerns manifest themselves across the supply chain via the growth of campaigns that are optimizing toward “high” viewability rates (50%-70%) as a transaction metric to validate campaign delivery. In turn, DSPs and SSPs are required to shift delivery toward publishers and placements with the historically high viewability rates. However, when evaluating viewability in a vacuum there’s a risk of optimizing toward low-quality, ad-cluttered publishers that may outperform premium publishers on viewability metrics simply because their pages are loaded with intrusive ads that are above the fold, often to the detriment of the user experience (something Ari actually wrote about way back in 2014 in Ad Age). While 100% viewability may indicate that all impressions were viewable to the campaign audience, this metric alone lacks insight into the quality of the content of the page and context of the ad. Furthermore, an ad’s viewability only denotes that the ad loaded onto the page -- it offers no indication that the user actually saw the ad.
In market there is a chorus of industry leaders suggesting that brands should not pay for an impression when it is not viewable. At face value this seems like a logical conclusion: why would I pay for something when half the service was not delivered? However, this both overstates the significance of an “in-view impression” and undercuts the additional benefits a campaign reaps when an ad is rendered on the page, regardless of its viewability score. In cases when the ad is not considered viewable, the ad may actually have been viewable and seen but various technology issues prevented complete fidelity. Further, by purchasing the impression, the brand has purchased not only the right to show the ad to the user, but to collect the information that the user was on this page, consuming this content - which can be helpful in building a holistic profile of the behaviors of their target consumers. Advertisers can work directly with DMPs to build and enrich their target audiences by dropping pixels that don’t need to be in view to capture data. Still, the demand for a single viewability standard grows louder.
The first step in the push toward universal viewability is happening today in mobile, where the Open Measurement SDK (OM SDK), an open-source project spearheaded by the IAB, will facilitate third-party measurement from multiple vendors across iOS and Android. The SDK is scheduled for release in 2018 and its objective is to introduce standardization and scale to mobile, where app publishers can install a single library to capture viewability events for any approved measurement vendor. This will effectively introduce a “bring your own viewability tag” dynamic that, if successful, we expect will introduce a larger appetite for a single source of truth when it comes to capturing attention metrics -- e.g. fraud, user data, CTR. As Jonah Goodhart, co-founder of Moat, noted at last week’s Industry Preview conference -- Oracle’s acquisition of Moat is the first step toward the integration of measurement and audience data, where marketers will not only have insights into viewability, but also the user segment targeting through a single vendor. Furthermore, this poses an opportunity into which Walled Gardens and Connected-TV can follow a similar, open-source measurement models.
According to DoubleClick, “content that holds a user’s attention has the highest viewability,” a claim that resonates with TripleLift’s mission to put user-centric content monetization at the forefront of our business. Conversely, there exists an inherent tension between brands requiring high viewability metrics and its effect on user experience. While a publisher may be incentivized to produce as many high viewability slots as possible, this is a tradeoff between monetization and the end product -- the same problem that exists in display advertising where the business model of the internet is tilted in favor of the marketer KPIs, not the user. TripleLift is well positioned to continue addressing marketers’ viewability requirements through our Guaranteed Outcomes Initiative, integrations with MRC-accredited measurement vendors, and high-viewability deals, but as the demand to drive higher viewability numbers grows, we should expect native ads to meet this standard, while staying true to our company’s mission.