Viewability is the most meaningful metric buyers have ever had – but it needs to be used wisely
It’s the best thing to have happened to digital media metrics, but focusing entirely on viewability at the expense of other metrics is a mistake, writes Wavemaker’s Greg Cattelain.
At last month’s Programmatic Summit in Sydney, a panel tackled the topic of “Kill the metrics that don’t matter”. At one point, the moderator asked each participant to nominate the two metrics they would “kill”.
One of the responses was “viewability and fraud”. I fell off my chair. The panelist justified their answer by saying viewability should be a given. But is it?
Viewability verification has been a standard practice for three years now, so it’s a good time to step back and look at what we’ve learned from it and how it can be used to better understand channel impact in an environment where the user is in control.
Viewability is not a given
Unlike other mediums, the internet is a world in itself: a world where the user is in control. Where people can scroll, flick, skip, move on. Where people choose to visit sites that carry no advertising, or use ad blockers to avoid having ads.
And on top of that, you can hack the system and make fake websites and fake users. Imagine if there we so many TV channels that you could create fake ones and nobody would know.
So our role is not just to say: “I’m going to select channels that are 100 percent safe and provide 100 percent viewable impressions”, our role is to ask, “How can we give brands an opportunity to be seen in digital? What is the value of viewability on what we are trying to achieve? Is there any relative value in viewability rates between websites and platforms? Are we working with publishers that have minimal fraud rate?”
As media planners, these are the questions we now have to answer. Involving clients in that conversation is a good way to increase their knowledge and work in full transparency. Clients need to be aware of the fraud rate of all partners they are working with and understand that not all publishers or trading desks are the same.
Fighting fraud is an industry effort that includes clients, agencies, publishers and ad-tech, as not one specific actor is responsible for it and there is no worldwide web police.
It makes our job far more interesting
In the nascency of digital media, we bought millions and millions of impressions and then realised that most of them didn’t even have an opportunity to be seen. With viewability, the conversation is focused on the quality of media exposure, which allows us to understand the correlation between viewability rates and ad recall per platform.
We can now measure the tension between “opportunity to see” and reach, on-target and contextual relevancy – all of which is important, because trying to improve one particular metric has a detrimental impact on the others. In short, if you are obsessed with one metric, you lose control over other metrics that matter to you.
So choosing a viewability rate of, say 70 percent, could compromise your reach by a factor of two. Do you choose to reach 100,000 people (for argument’s sake) at 70% viewability, or would you rather reach 200,000 people with a viewability rate of 50 percent?
Understanding how the different metrics interact allows us to reframe the conversation with clients and introduce more meaningful metrics or indexes that reflect individual campaign objectives. For instance, we can talk about “cost per quality reach” (which includes all quality metrics above) or customise indexes to take ad recall into consideration.
But what happens if you want 100 percent viewable impression and zero percent fraud?
If that’s the case, you may not want to invest in digital media. And that removes a lot of opportunities to talk to your audience.
The biggest mistake buyers and clients can make is to evaluate a website only through the viewability lens. Buying only toward 100 percent viewable placements makes the web shrink dramatically. Your opportunity to get your message out there at scale becomes very hard (outside maybe one publisher here in Australia). And what does 100 percent viewability to a very limited number of people do to your brand awareness? Not much.
Back to the earlier example of reaching 100,000 people at 70 percent viewability or 200,000 at 50 percent – in the second case, you actually deliver more viewable impressions and your message is more visible overall. So it might be the right thing to do for your branding campaign.
If we refocus our conversation with publishers and advertisers around exposure quality and media outcomes, we can start to put this in perspective with the price we pay for the media. And by doing so, we can trade and plan media with impact, scale and business outcomes in mind.
That’s where media buyers make a difference and what separates effective planning from dumb planning.
As per zero percent fraud, it’s an objective. The same way eradicating criminality in general is an objective. We must continue using safety tools and rigorously assess the publishers we are buying from. But there will always be a degree of fraud by the very nature of the medium.
Yes, it’s the most important metric of all, but aiming for 100 percent viewability in isolation – and above all other metrics – could be the wrong thing to do.
Greg Cattelain is programmatic and social director at Wavemaker Australia.
No matter what the medium. I am always in control. I can choose to change channel, radio station or turn a page. Switch on or off at will. Find something without ads (the ABC springs to mind). Choose to ignore something or not as I please.
“Viewability”. Yet another useless, idiot buzzword to justify something. Anything. Next thing along with “Crisis Evaluaters” we’ll have a CVO…
User ID not verified.
How fun living in 1980 must be David.
Viewability measurement allows for as you admit yourself, transparency like we’ve never seen before.
The days of blind buying TARPs off OzTAM data are over.
The quicker this can become standard across all of digital (along with IVT and ad fraud measurement) the quicker we can clean up our industry and continue growing buying efficiency and effectiveness.
User ID not verified.
Isn’t it fun how insults are used to make a point. From people who are supposed to be “communicators” and expert “marketers”.
HA!
User ID not verified.
“100 percent viewability in isolation – and above all other metrics – could be the wrong thing to do.”
Greg is from Group M. Isn’t Group M’s viewability definition 100% in view for one second?
User ID not verified.
I do not come from the participation award generation David.
You spoke rubbish and it got responded to accordingly. Either build a thicker skin or get more intelligent opinions.
User ID not verified.
Agree with Hank, David.
Please enlighten us with your alternative to provide transparency across our industry?
User ID not verified.
The GroupM 100% viewability product is impressions only on Outlook – arbitrage?
User ID not verified.
If the only objective is awareness, then I can see that perhaps it is more cost efficient to lower the viewability and work to a vCPM. However, if you are wanting to do any kind of attribution, or measure the performance of the display activity on website visits or actions (post view) then having anything short of the highest viewability makes working out which impressions actually made a difference impossible.
User ID not verified.
@Greg The lede would make more sense if it said “Viewability is the most meaningful DIGITAL metric buyers have ever had”. Though you may not have written the lede.
@Hank. Psssst. TV ads are 100% viewable. And yes I know people go make a coffee or go to the toilet in ad breaks. But I’ll let you in on a secret – people also do other things when digital ads are streamed.
@Hank again. One day you will learn that an ad served is not the same as an ad seen – especially when it is splashed around multiple devices in the spray and pray programmatic world.
User ID not verified.