Back to Member Vault

Keeping brands safe doesn't have to mean a media lockdown

Tags:

Brand Safety
Brand Safety

This content was created by an IAB UK member

Members of IAB UK can contribute to the Member Vault. Log in to submit your content.

GumGum's Managing Director for EMEA Peter Wallace explores how to construct a media plan that incorporates brand safety and suitability measures while still maximising impressions

 

Like the UK population, brands are starting to cautiously emerge from nearly three months of isolation. Although ad spend is still significantly pared back compared to February, the economy is beginning to move again, and marketers are considering how to reach back out to consumers.

But media planning remains rife with new obstacles, chief among them is the fact that a large proportion of online content is still dominated by virus-related articles. When this trend was at its height in March and April, advertisers responded swiftly; many of them used keyword blocking technology to stop their ads appearing next to mentions of anything connected with the pandemic.

It was the media planning version of social isolation and it worked in the sense of keeping brands untainted by virus associations. The trouble is it also removed enormous amounts of media inventory. 

Brands lost out on huge audiences just as readers and viewers were hungry for information and entertainment, while publishers lost so much revenue that the UK’s culture secretary Oliver Dowden was forced to make a statement encouraging advertisers to bring back their ads in order to protect the media.

 

Looking at the big picture

So, what can we learn from this panic-driven period? We conducted research at the height of the pandemic's spread in the UK, to look closer at the effect of this stringent brand safety activity. Over two months between March and May, we used our proprietary machine-learning tool Verity to analyse over 7 million unique pages of internet content containing various COVID-related words.

The majority of these pages would have been ad-free because they contained “unsafe” words associated with the pandemic. But running them through our machine learning contextual analysis system, which takes into account the broader context in which these words appear, we discovered that a massive 67% were completely safe environments for advertising.

Context is everything and this, we believe, is now the key to planning brand safety. Imagine an article about gardening that begins with a reference to the fact that the lockdown has led to many people spending much more time outdoors and experimenting with growing plants and vegetables. Blunt keyword blocking tech would detect the word “lockdown” and remove this article from an inventory despite the fact that the virus association was only indirect, and the article is actually a particularly useful host for an ad for certain brands.

The solution is found in the use of AI-powered systems, like Verity, that use Natural Language Processing (NLP) to look at the broad meaning of an article. Add in the ability to tap into computer vision tech to analyse pictures, and you have a system capable of detecting the context and nuance of the entire page in a similar manner to how the human brain processes it. 

So, we can avoid unsafe media environments for brands, but without avoiding so much content that campaigns lose their power and publishers face a big drop in income.

 

Tailoring safety to fit

The real beauty of moving towards this more nuanced form of analysis is that it can be adjusted according to a brand’s relative sensitivity to risk. 

Depending on the product categories and audience profiles, brands will vary in marketers' willingness to expose them to certain types of content. An ad for a pharma brand might be OK appearing in pandemic-related content, for example, while an ad for a cruise line would not. So, the ability to dial up and down the sensitivity to risk will help both brands find the appropriate content they need.

In a nutshell, AI technology now allows us to move beyond just brand safety and instead to gauge individual brand suitability

By employing AI systems to look more intelligently at the context of content we can minimise risk while still allowing brands the freedom to mingle with a volume of audiences. 

By Peter Wallace, Managing Director EMEA

GumGum

GumGum is an artificial intelligence company with a focus on computer vision and natural language processing. Our mission is to solve hard problems across media by teaching machines to understand the world. Since 2008, the company has applied its patented capabilities to serving media-related industries, including advertising and professional sports. For more information, please visit gumgum.com

Posted on: Tuesday 16 June 2020