Viewability – the metric of whether an advertisement is visible to users – is a hot topic in the digital ad space. We’re seeing an increase in advertisers partnering with third-party viewability measurement providers, which comes as no surprise. It’s hot off the heels of a turbulent 12 months in digital, as brand safety concerns and issues with self-reported platform metrics have dominated industry news.
A static digital ad impression is typically considered viewable if 50% of the ad appears in an open browser for at least one second, continuously. Social ad delivery mimics the look and feel of organic content, and ad delivery is typically served “above the fold” as a user scrolls through the platform’s feed-based interface.
Video adds further complexity, as the Media Rating Council (MRC) defines a viewable video impression at 2-seconds continuously, 50% in view – a definition not consistently shared by the different platforms’ own video view metrics.
Within a feed-based environment, this definition could arguably be interpreted as more of a performance metric, comparable to view rate given the 2-second continuous view stipulation.
This focus on viewability raises two key questions:
1. Is it fair to apply the MRC viewable video definitions to social?
2. Do we have the tools to accurately and fairly measure viewability?
What counts as a view
Each social platform offers its own ad experience, which means they also have different ways of describing and measuring impact. The definition of a video view differs both in “amount in view” and “time in view” across each platform, also deviating from the MRC classifications.
Without consistency it is understandable that advertisers might look to a third party to create a level playing field. However, the adoption of a consistent viewable definition assumes identical impact across all formats and platforms, as well as the wider online ecosystem. That’s why it’s important to consider whether the definition of viewable impressions undervalues social and whether transacting on it could impact performance.
Viewer behaviour on social
A recent Nielsen study concluded that 38% of brand recall, 23% of brand awareness, and 25% of purchase intent are driven by video impressions shorter than two seconds. And on Facebook, mobile news feed recall has been found to improve at a statistically significant rate after only 0.25 seconds of exposure. Both of these studies bring into question the incumbent MRC viewable video definition of 2 continuous seconds, 50% in view.
Studies observing scroll speed on social also suggest that there are viewability differences based on a user’s age and device. Nielsen found that people process mobile content faster than desktop content on Facebook, at 1.7 seconds vs. 2.5 seconds. Research by multiple platforms revealed that younger users scroll at a much quicker pace than older audiences. Bearing these points in mind, the current viewable video definition seems out-of-step with how people consume content on social platforms. A blanket definition for all of digital might substantially undervalue or overvalue certain channels, meaning a campaign run purely on viewable impressions could see poor performance or a significant drop in reach.
Third-party measurement vendors like Moat and IAS use inconsistent methods to establish whether impressions meet viewable criteria, resulting in a wide range of performance. Such discrepancies draw into question the accuracy of results, specifically when it comes to video. Data integrity and consistency is key so that marketers and advertisers can rely on these stats to inform planning and buying decisions.
It’s worth asking if viewability is even the right metric for social media. Rather than trying to force a traditional display metric into the social advertising space, we should be asking measurement partners to redefine the metrics based on user experience and behaviour on social platforms. If 38% of brand recall happens before the defined two seconds, that brand value should be tracked and acted upon, not pushed aside.
Considering both performance and measurement concerns, it’s clearly still too early to transact on viewable impressions – especially in relation to video. The current definition of a viewable video impression has been proven to ignore a large portion of attributable awareness, preventing clients from achieving their true business goals on social. The continued inconsistency across third-party measurement providers also draws into question whether reported results can even be relied upon.
While viewability is undoubtedly a worthwhile concern, the metric needs to be redefined and great care should be taken when applying the current definition to social platforms.
By James Rayner, director of media operations at Adaptly
PrivSec Conferences will bring together leading speakers and experts from privacy and security to deliver compelling content via solo presentations, panel discussions, debates, roundtables and workshops.
For more information on upcoming events, visit the website.
comments powered by Disqus