Rob Bradley's byline article feature in The PR

Read the English translation of Rob Bradley's article in Korean trade media The PR below. 

English translation: Many news publishers are finding themselves in a perfect storm at the moment – generating record audiences and seeing strong user engagement due to the public need for trusted news and information during this unprecedented news cycle, but at the same time seeing a downturn in revenue to support essential journalism as advertisers pause spend due to wider adverse economic conditions and/or shift spend from the news space due to fears of being aligned with hard news content.

I cannot stress how important it is that the news industry addresses the second of these points to ensure that our content and inventory is being utilised in the best possible way. This way forward must be through a combination of sophisticated ad tech andreal-life human know-how to ensure that brands can be confident that their presence in a news environment is suitable and appropriate.

Arbitrary use of keyword blocklists had already been a strain on the digital news industry for many years with advertisers and agencies too often resorting to blunt instruments to ensure they are kept well away from harmful content. The problem with this approach is that it poses an existential threat to the news industry’s ability to generate advertising revenue and also means that brands miss out on being aligned with some of our most engaging and popular content on account of a single word that happens to be on a blocklist being used in an article. To address this issue, the industry must move away from keyword blockers and turn to contextual technology to increase campaign performance by protecting brands and promoting them in the right places. This is something we are committed to at CNN through the implementation of Sentiment Analysis Moderator, or SAM for short.This ensures brand safety, a prerequisite of premium publishers, along with brand suitability.

So, how does it work? Using a mix of neuro linguistic artificial intelligence, SAM intelligently analyses the context of sentences to cleverly determine when content is brand suitable.This means that all content –video, audio, text and galleries –across CNN’s properties is scanned by SAM which rates each page, pre-load and pre-bid, along a scale for how suitable it is for each campaign, based on an advertiser’s list of keywords and objectives. It provides much more nuance into the kinds of content an advertiser’s brand messages would be aligned with, rather than heavy-handedly blocking all pages that have certain keywords on them.

In addition, we use SAM in tandem with our internal content classification tool, Contextual Engagement Platform (CEP), which uses IAB taxonomy in line with industry standards. And, of course, even in this age of machines, we ensure that the human element is present as our editors classify and tag content as it is published.

Pulling together all three elements –SAM, CEP and manual tagging –we are able to utilize positive targeting. This really shifts the approach from one of using basic ad tech to avoid being next to content, to using a rich mix of tech and human understanding about content in a solution that can actively match brands with content that reflects their values and that will resonate with their target audience. It’s a win-win for all involved because the publisher is better able to utilize inventory, and the brand can be placed more intelligently alongside relevant content.

One example of how positive targeting works is how a brand wanted to be adjacent to positive health stories to support and raise awareness of cancer, particularly help guides and recovery stories. Previously, many stories would have fallen foul of blocklists as articles referenced terms such as cancer or survival, however our tech was able to ascertain the context, tone and relevance of stories relating to topics such as breast cancer awareness, which meant a positive sentiment, allowing us to appropriately target the campaign.

Utilising these contextual tools opens up much more suitable inventory and in turn generates better campaign results as brands are paired with the most pertinent content for them and their ideal audience.

In some cases we found over 50% of neutral and somewhat positive news content was misclassified by existing keyword blocklists, demonstrating the importance of how nuance and context can offer many more opportunities and inventory for brand suitability and brand preference.

For another campaign, after switching from an industry piece of tech to using SAM, we were able to identify that there was five times more inventory that was brand suitable for this specific client. It is more important than ever for brand solutions to be present in the news environment as marketers strive to communicate their messages to engaged and large scale audiences who are turning to trusted news organisations for facts and information. While it’s understandable that brands won’t want to be alongside all of our news content, we need to remember that by our very nature publishers like CNN are brand safe as they offer a premium environment, high quality content, engaged and affluent audiences and minimum fraud.

This requires a shift in mindset from all involved to think less about “brand safety” and more about “brand suitability”. Less about where do we not want to be, and more where do we want to be. However, I recognise that advertisers and agencies must feel confident that a publisher has the protocols and tech in place to deliver on this promise. With the positive feedback and results so far, we are confident that we have found the answer in a shared commitment to bring our advertisers and audiences the best experience possible.