With digital advertising spend expected to hit $US300 billion by 2020 — representing almost half of the total advertising spend by brands globally — the quality of online environments will come under ever closer scrutiny.
That point was reinforced in May this year when Cisco CMO Karen Walker pulled her company’s advertisements from one of the world’s largest video site over brand safety concerns. In a company blog, she explained her reasons why (although she didn’t mention them by name).
“At Cisco, we would rather not wait for something bad to happen. We are working closely with all of our media partners to ensure that Cisco’s online advertising meets our stringent standards. We only advertise where those standards are met and where we can ensure inappropriate content is not shared.”
Walker noted that the speed of change — which includes things like the velocity of content creation and the speed of the news cycle — meant that sensitive issues in the media can spread faster than any media platforms’ algorithms can update.
This, she suggested, can lead to a brand-tarnishing experience. “Similarly, some content platforms are not properly monitoring and categorising the content on their sites as it is posted.”
It was no doubt a sinking feeling for the giant video publisher and unfortunately also a familiar one as it is not the first time it has had to deal with the consequences of its major brand partners taking flight over concerns about brand safety.
In early 2017, businesses as diverse as Adidas, Amazon, Hilton, and Nordstrom (and incidentally Cisco) discovered their advertisements were being run on channels promoting white nationalists, Nazis, paedophilia, conspiracy theories and North Korean propaganda, according to a report on CNN.
This is simply the latest example of the commercial impact that brand safety concerns can have on both the brands themselves and also the platforms and publishers.
While the cost in lost revenue to the publisher is clear, it is important to understand that a business like Cisco would not easily withdraw from a such a critical channel. Cisco itself is likely to lose sales from leaving such an influential channel, though the impact will be harder to measure. Cisco is the world’s 16th largest brand according to Interbrand, which values the Cisco name at $US32 billion.
Nor is this a problem only for a few data-driven global advertising platforms and their customers. All brands that utilise digital channels need to mitigate brand risk — particularly in a world of programmatic advertising which operates in real time and mostly without human intervention.
And it is made more difficult by the reality that one brand’s risk is another’s opportunity — so that programmatic strategies need to reflect the individual philosophy of brands.
Perhaps the biggest change in mindset in recent years is the response of consumers to the placement of ads.
In the world of print, ad placement adjacent to stories did not tend to raise the ire of consumers, while television advertising took viewers away from the content altogether.
In the digital world, however, advertising is seen as an endorsement of content. It is a subtle change in consumer thinking, but an important one. Perhaps it is the impact of years of retargeting, or perhaps social media now gives a voice to concerns that were always present but lacked an outlet.
Either way, as advocacy campaigns against sites like Breitbart or particular shows on Fox News have shown, brands need to add a new layer to their decision-making. The American ABC network’s decision to shut its top-rating Roseanne sitcom after the eponymous star posted a racist tweet shows that digital culture is starting to extend back into traditional media. The network was widely applauded, and it made clear its decision was values-based — but it is likely it also anticipated the reaction of its audience had it failed to act swiftly.
The problem of brand safety is exacerbated by user-generated content, where there is little control over the context in which an advertisement might appear and where emotional responses to news as it happens can change the risk profile of a site in real time.
Take a media service like the web site of a national daily newspaper. Its news pages might present a risk with violent content at one time of the day, while its lifestyle page may be perfectly safe. However, those positions could reverse at any time during the day due to the vagaries of the news cycle and the editorial choices of section editors.
For this reason, any technical solution needs to act at a page level and not a domain level, dramatically increasing the complexity of risk.
To each its own
Likewise, each brand has its own particular tolerance and definition of risk and safety.
While it is easy to find agreement on some content — hardcore pornography, child abuse sites and sites built around violent content — you don’t have to move far from these to run headfirst into the complicating factor of individual subjectivity.
Is Breitbart a hate site, or a legitimate media player with a valid political view? World-renowned media services in some parts of the world are much more willing to display violent news content than, say, the ABC locally.
While many people assume programmatic brand safety is built on keywords or subject tags, the truth is less nuanced. Brands typically define their degree of brand comfort with a subject such as alcohol, adult content, and violence and then a service like IAS provides scoring against content. This score fluctuates in and out of the brand’s zone of tolerance, depending on the content cycle.
Video, which represents the next battleground for brand safety, is harder to protect because of the nature of the content and the difficulty of determining what represents a useful signal. However, even on this front progress being made. For its part, Google is working with third parties to build new solutions.
Don’t just detect, mitigate!
Many brands still need to get the basics right, and that starts with moving beyond monitoring brand safety to blocking ads in unsafe environments. We typically find that brands with a measurement mindset recognise the need to take control of managing brand safety in a programmatic world, and that this will deliver the best outcome for the business.
Brand safety requires a level of accountability by companies that cannot be outsourced to an agency. Nor can brands assume that social media platforms will take care of the issue for them. For that to happen, the social media platform would have to understand the brand guidelines of all advertisers in all contexts.
To protect the bottom line, brands need to maintain control themselves.
About the author
James Diamond is the managing director of Integral Ad Science Australia and New Zealand. IAS is a corporate member of the Which-50 Digital Intelligence Unit. Our members provide their insights and expertise for the benefit of the Which-50 community. Membership fees apply.