A Look Into RTB's Quality, Ad Impressions, Data & Future
Some call it real-time bidding (RTB) 2.0, and others simply believe that RTB has not yet reached a level of maturity satisfying for brands and agencies. In a recent panel at OMMA RTB in New York City, debates broke out over issues surrounding metrics, content, quality of inventory, and how the world of RTB has actually evolved.
With digital advertising's adoption of audience targeting, it’s become common to hear the phrase, “reach the right audience, with the right message, at the right timeand in the right place.” However, there has been a lot of debate over the value of “place” or “content and context” within RTB. If you are reaching the right audience at an optimal time, does inventory matter? Does media placement make much of a difference in RTB -- or is it just the data?
On the one hand, as long as agencies are meeting KPIs, they don’t care about controlling content, as long as it’s not anything offensive. On the flip side, many marketers disagree and say that the context of the placement and publisher that an agency works with is critical – but in a way where you can still pick the user. Rob Griffin, EVP/director of product development at Havas Media, and Joel Nierman, marketing and media director at Critical Mass, later noted that there really are no bad impressions, only bad valuations. You are either willing to pay a higher rate for selective content and/or you are willing to pay more on an RTB platform for a specific user.
Another hot-button topic for the panel was viewability. Anne Hunter, SVP advertising effectiveness at comScore, stated that the average percent of viewable ad impressions is between 60%-70% on premium inventory and 40%-50% on non-premium/exchange inventory. The problem here is that any ads that were not viewable by the user still placed an exposure pixel, even if the user never actually saw the ad. Thus, if that user proceeds to fill out a survey or buy a product, the non-viewable ad will appear to have had real business value for the brand and will be given credit, when in fact it was never seen by the user.
In order to get an accurate account of performance and set pricing according to market demands, marketers and publishers should focus on only counting ads that are seen. At the publisher level, the infinite amount of pixel dropping hurts a publisher’s price, since it shows a massive number of impressions available for cheap, making it harder to price the real quality inventory. Therefore, publishers providing quality inventory still might not look as if they're performing.
Peter Naylor shared NBC Universal’s story of adapting to the viewable impression. As an early adopter, NBC Universal began making changes for viewability back in 2010. Initially, there was a 30% reduction in inventory, but also a substantial increase in ad effectiveness, which in turn led to higher pricing opportunities, according to Naylor.
For marketers, the viewability issue affects performance and spend. If a high percentage of ads are not even seen, then there's a large amount of money spent on wasted impressions. Additionally, since agencies rely on what the conversion data tells them in order to determine what an impression is actually worth, they may be allocating money and optimizing to a bot vs. an actual end user.
The overall sentiment on OMMA’s panel was that issues of inventory quality, ad impressions and the use of data will remain top-of-mind for marketers. However, depending on a client’s goals, some will care more about some issues than others. As Griffin said, “History does repeat itself.” The reality is that we are now solving the next stage of challenges for RTB 2.0.