Viewability is a key metric to monitor when running programmatic display campaigns. It can help monitor against poor quality and low value inventory and also act as a counter balance when tracking post-view conversions.
Viewability benchmarks and measures are something I advise all of my clients to introduce.
There remains however, a lot of misunderstanding around viewability and what the metrics actually mean, or the implementation of standards will actually cause.
- Viewability means your ad was seen
A viewable ad, is not necessarily one that was seen. The best way to look at viewability is that is a measure of whether your ad had the opportunity to be seen.
The ad appeared on the page, for the appropriate amount of time (as defined by the IAB or vendor settings) and had the opportunity to be viewed by the user. Whether or not they viewed it is not part of the metric and not something you will know.
- Viewable = above the fold
Any ad placement on the page can be viewable, and any can be non-viewable. It all depends on the website layout and the general user experience. If you have a website layout where most of the useful content is lower down the page or it has a large header section. Then the user experience will be to immediately scroll upon loading. So, the placements higher up may have the least viewability.
- Viewable = good
Pop ups are highly viewable. Ads that take up half of the page and don’t disappear when you scroll, are highly viewable. Tiny ads that are barely legible, are highly viewable.
Just because an ad is viewable, doesn’t mean it is effective.
- Implementing viewability benchmarks will see your costs skyrocket
This is a myth that has been repeated by programmatic suppliers for a number of years. That buying more viewable inventory will see your CPM sky-rocket. This is mostly because what it will more likely impact is the profit margins they are able to make, especially when charging on a flat CPM.
As a generalisation, more viewable inventory will cost more. But it is not as cut and dried as that, and it certainly isn’t a linear relationship.
What implementing viewability benchmarks does in the first instance, is eliminate poor quality inventory. The ads that never had a chance of being seen anyway. Yes, this will likely mean your average CPM rises a little, but the ads weren’t being seen anyway.
- The higher the viewability the better
As mentioned I am an advocate of implementing viewability benchmarks. But I believe it is important people understand how these metrics work and the impact it has on your programmatic activity.
As you test different viewability levels you will get an idea for the impact it has on cost. As this is unlikely to be non-linear, when you equate costs back to a cost for each viewable impression, you will find out where the most cost effective level might be. Here is an illustration:
|Viewability||CPM||Viewable Impressions per thousand||Cost per viewable impression|
In this example, you will see that the most effective impressions are bought at the 60% viewability threshold, before the price spikes.
Performing your own testing in this way will help you find your own sweet spot.
Seek to understand before you implement
Viewability is a key metric when running programmatic activity and a measure you should be introducing. However it is important you send some time understanding it before blanket implementing minimum thresholds which could have a detrimental impact on your activity.