The online advertising industry is quick to become obsessed with the fad of the hour. The moment a topic becomes hot — be it fraud, viewability, private marketplaces, native ads or transparency — everyone suddenly has an opinion. Articles get written, startups get funded and brands demand to be on top of this new trend.
From the perspective of advancing digital marketing, this behavior is positive. After all, we’re working in an industry that’s fast paced and highly dynamic. Where this behavior becomes problematic is when the trend of the moment takes precedence over the actual goal of online marketing. That is, driving incremental revenue or increasing brand awareness.
Viewability is the latest online marketing trend that falls into this trap. Viewability itself isn’t a bad idea. After all, which marketer doesn’t want the ads that they’re paying for to be seen. The problem that emerges, however, is when viewability itself becomes the main goal / KPI of a campaign.
Increasingly, brands and agencies are demanding that viewability minimums be met for their campaigns. These minimums are required even when better performance could be obtained for the campaign’s true KPI at the expense of viewability. Let’s analyze this through some actual campaigns.
Campaign 1: $10,000 campaign with a goal of a $200 CPA.
Let's say there are two publishers that this marketer can target for their campaign. Publisher one has 70% viewability with a $7 CPM and a .004% conversion rate. Publisher two has 30% viewability with a $2 CPM and a .005% conversion rate.
How should this campaign’s marketer analyze the performance of the campaign? If viewability is mistakenly made to be the primary KPI, Publisher 1 clearly outperforms Publisher 2. However, this would be a grave mistake as Publisher 2 is clearly providing better performance (its CPA is ~46% lower).
This is where viewability has become a red herring for online marketers. Rather than focusing on viewability as a KPI on its own, marketers need to ensure that they’re focusing on the KPIs that they’re actually trying to drive.
Does the above still hold true for branding, rather than DR campaigns like the one referenced above? It certainly does if the marketer is approaching the campaign with actual KPIs ( i.e., increasing in brand awareness, brand recall) rather than simply pretending that viewability is the KPI they’re aiming for.
Where does this leave viewability?
Although viewability doesn’t make sense as a KPI in and of itself, viewability does serve a very important role from an attribution perspective.
For instance, should an ad that drives a conversion count if it was served but not seen? Should a user be given a brand awareness study if they were served an ad but never saw the ad? In both cases, the answer is no. Performance shouldn’t be attributed to a served ad if that ad wasn’t seen. That’s why in the above example, conversions rate wasn’t applied to the number of impressions served but rather to the number of viewable impressions served.
Viewability as an input to attribution uses the best aspects of viewability — it’s ability to more accurately determine if an ad was performant — while avoiding the red herring of viewability as its own KPI.
As a marketer, a great question to ask your vendors is “are you determining campaign performance based off viewable impressions or served impressions?” Having conversations from this perspective rather than solely chasing viewability benchmarks ensures that you’re optimizing to the actual goals you’re trying to achieve.