« All Experts

Ask the Experts

We chat regularly with experts and pose them a single question to help share their knowledge and experience with you.
Robert Meth Head of Enterprise Sales TubeMogul

What is best practice in online video view-ability?

For the past year, viewability has dominated advertising industry headlines, research and happy hour musings. And while most discussion has centered on fairly nuanced aspects like standards and measurement capabilities, a troubling trend has emerged: the conflation of viewability with ad fraud. Grouping these two distinct phenomena together as a single entity neglects not only recent technological advancements in measurement and prevention, but also threatens to actually increase inefficiency and contribute to more wasted ad dollars.

First things first ­our definition of viewability: a “viewable” video impression is one where 50% of the video player’s pixels are in view on an active tab for two consecutive seconds. (They don’t necessarily have to be the first two seconds of the video, and the video doesn’t have to be audible). Whether two seconds is long enough is an issue that advertisers and publishers will continue to debate, but we’re more concerned with identifying the main reasons for low viewability.

And the main reason is you. We found that about 70% of non­viewable impressions were due to an inactive tab or window. In other words, we all start our videos, realize an ad is in front of the content we want to watch, and we switch to another tab to wait it out. But unfortunately, this level of insight isn’t available until after the ad has been shown.

This bears repeating: viewability cannot be measured before an ad is served. However, we’ve found that optimizing towards a large player size can mitigate the level of non­viewable impressions in a campaign. The more immediately engaging an ad is and the bigger the video player it is shown in, the more likely people are to remain on the active tab to watch the ad.

We found that about 70% of non­viewable impressions were due to an inactive tab or window.

Robert Meth, Head of Enterprise Sales, TubeMogul

Fortunately, player size is available before an ad is shown – and so we can use it as a proxy to maximize the opportunity for viewable ads. Marketers should demand that every inventory source they work with provide the player size of every ad impression in the bid information so they can make a decision based on player size before the ad impression is purchased.

But how can we be sure that there’s a real person looking at those large players? The thinking goes: “If viewability is about making sure that my ad had the chance to be seen, and fraud is about making sure that my ads are being shown to real humans, then can’t we just combine the two?” Unfortunately it’s not that easy. That’s because fraudulent inventory – or ads that aren’t shown to real people – can have excellent viewability rates.

Through a technique called “ad stacking,” nefarious characters can place ads on top of one another, and make it appear as though multiple ads are appearing above­the­fold or in a large player size, when in fact only a single ad is actually visible to the viewer. As a result, advertisers who are optimizing only towards viewability can end up with highly viewable ads that aren’t ever seen by a real person.

The distinction is important because the technology now exists for advertisers to understand why their ads aren’t seen. Newly developed fraud verification – like Integral Ad Science’s Bid Expert – can help ensure there’s a real person on the other side of the screen and then give advertisers the metrics they need to optimize for high viewability.

Fortunately, it looks like Canada is comparably well­off: in our latest quarterly research reports, Canada had the lowest average price per viewable impression of any major market.