There are few things getting as much attention, or indeed that are as important to the industry at the moment as the topic of viewability. It’s been quietly bubbling away for some time now and it garnered its share of attention at the recent IAB NewFronts in NYC.
I’ve been quite vocal on the topic and my commentary late last year around our support for the move towards viewable impressions stands, with several caveats relating to the technical challenges and setting reasonable expectations, still applies.
Viewability is the core currency unit that will be built into the digital and cross-platform GRP standard. Viewable digital ad impressions defined by a pixel and duration metric will permit equalising of exposure metrics across screens; and so the shift from a served impression to a viewable impression will be yet another step to greater accountability in digital media.
Yet viewability is not the ultimate goal as a viewable impression means, in essence, that there is the opportunity for the ad to be seen, but being considered viewable is in no way a guarantee of performance.
We are in a time of transition for viewability as a metric and it’s important that the industry understands that 100 percent viewability is unreasonable at present. Indeed it has several weaknesses that still need to be addressed.
First of these is that different ad units, browsers, ad placements, vendors and measurement methodologies yield wildly different viewability numbers. U.S publishers are observing variances in measurement of 30-40 percent. These variances likely represent the limitations of current technology.
Additionally there are still technical obstacles in accurately tracking what is viewable and what is not viewable – and hence a lot of inventory is simply not measurable. This non-measurable inventory should not be treated in the same way as non-viewable inventory, and these non-measured impressions shouldn’t be free of charge as they might actually be viewable. At this point it is impossible to know.
Finally we still need to address the issue that a non-viewable impression does not equal a fraudulent impression. The extent to which viewability and fraudulent impressions intersect is not clear, but IAB Australia has made it clear that they cannot simply be bundled together. A fraudulent impression is one that an actual human being could never have viewed. A non-viewable impression should be considered one that a human didn’t see, but could have if various circumstances were different (circumstance being broad, but generally encompassing factors such as how the site is designed and the length of time a user was on the page).
Our view is that the whole industry needs to pull together, working with what is currently technically and commercially feasible while this transition is occurring and following the prescribed IAB Australia guidelines.
Most importantly we strongly recommend that the market should refrain for trading on viewable impressions until the aforementioned shortfalls in viewability are addressed.
We need to cut through the confusion and develop a marketplace that works toward the common goal of ensuring quality advertising that is viewable and measurable.
Randall Rothenberg, IAB USA CEO described our current situation best when he said: “Publishers, agencies, marketers, and ad tech companies can resolve these differences by working collaboratively to make measurement make sense. We won’t do it by holding guns to each others’ heads.”
Sage words indeed – I hope our industry heeds them well.
Alice Manners
CEO
IAB Australia