The 21st Century Kettle - ad viewability

15/12/2016

Back in the day, the kettle was often cited as the reason not everyone could watch your ad on TV (as in off making a cup of tea). Today, “viewability” is the villain being blamed for ads not being seen on the internet. Alex Radford, GM Digital NZ, explores this...

Alex Radford Alex Radford GM Digital Auckland, New Zealand
https://carat-cdn.azureedge.net/media/7067/radford_square.jpg

The other week a colleague send me an article about the fact that 100% viewability was not necessarily a good thing. As I was reading this article I started thinking about this, and all I could think of was how bonkers our industry is sometimes. I mean that in a good way, but seriously when you start to explain the question of viewability to a person outside of our industry you can come of sounding like a crazy person.

Let’s just recap the argument, and I suggest you read this a couple of times, and see how irrational you sound. Even say it aloud for extra crazy credits. Here we go: when you buy digital media placements from banners to videos we do so in the complete knowledge that on average only 48% of your ads will be seen by a human ( comScore study). This is exacerbated for video where the average viewability drops to 41%. That means that our tracking tools tell us that an impression has been served, but the reality is that that impression was triggered by a robot or some other kind of internet trickery. Which therefore means that 52% of your expensive ad campaign was seen by robots (well really just lines of code).

Even the actually definition of a viewable ad is somewhat ridiculous; as far as the IAB are concerned, a display ad is viewable if 50% or more of its pixels appear on-screen for at least one continuous second. A video ad is deemed viewable if 50% of its pixels appear on-screen for at least 2 consecutive seconds.”

So what this means is that of that 41% of video views this also includes people that saw half of the physical ad for less than 2 seconds. Nuts right? Just try watching 50% of a video and tell me how much you were able to recall…

And it gets even more like a sub-plot of Catch-22. It has been proven numerous times that actually knowingly buying ads with low viewability rates actually performs better, from the perspective of outcomes that one with high viewability. Huh? So in plain English that means that by targeting more robots and less humans you will sell more of what you want to sell. According to separate pieces of research conducted by digital marketing agency IMM and programmatic media planning and buying company the Goodway Group, the economics of delivering 100% viewable campaigns doesn’t make financial sense, at least not yet. In the words of Jay Friedman, COO of the Goodway Group “I maintain to this day that 50% in-view at $4CPM beats 100% viewability at $10CPM all day long”.

 Which really is the crux of the argument. If you focus on on-site outcomes such as sales, engagement, ROI or other clear conversion metrics then viewability shouldn’t really enter the equation, as what is really key is driving relevant traffic, and then focus energies on conversion optimisation. However the argument becomes a little more blurred when your objectives are about reach and frequency of say a video campaign where onsite objective are less important.

Research supports the fact that completed views are more valuable to advertisers than click-through rates, with brand recall being significantly higher if the entire ad was watched. So again the focus for advertisers should be around what metrics are important to them, and develop creative and targeting that works for those objectives. If you need reach then focus on viewability and completed views with shorter videos with multiple versions. If you need engagement then build creative that entices engagement, and focus the buy on Cost per Engagement. And if your goal is on-site engagement then optimise your buy on what drives whatever conversion you need.

So, I hear you ask, what should I be doing, or more precisely what should I be asking my agency to do about viewability? Should I be demanding 100% viewability? Should I invest in technology that monitors viewability? Is this a storm in a tea cup? And those of course are great questions. And I wish I had the answer, but like the client ten years ago who no doubt asked their media agency what they can do to stop people making a cup of tea during an ad break, I don’t have a perfect answer, but what I can tell you is that ensuring that 100% of your audience actually saw 100% of your ad is impossible. And anyone that tells you differently is a liar, or trying to sell you something!

However the clearest way to ensure your ads are being viewed for the vast majority of your buys is to focus on publishers who are actively trying to improve ad visibility on their site – improving the location and size of ad placements, this ensuring maximum impact. Look at premium publishers, ensure your programmatic buys have a human-vetted whitelist, and if you are using networks ask how they define viewability and ensure that it lines up with your objectives.

From a media agency perspective we know that the industry is still trying to work out a globally defined standard, and this is getting better but we still have a way to go. We also know that there are significant flaws in the technology that measures viewability – each vendor measures it differently, so the best advice that I can offer is that if when you are buying digital media viewability is a primary objective then make this very clear to your agency, and ensure you definition of viewability lines up with theirs and the technology they use to measure your ad visibility.

And if you have objectives beyond viewability then right now focus on that metric first, and who knows the robots might really enjoy your creative!

Alex Radford Alex Radford GM Digital Auckland, New Zealand
^Back to Top