Tabloid news and false positives: Examining the brand safety cycle
Earlier this year, broadcast giant Sky declared it is yet to resume spending with YouTube after stopping advertising on the site 18 months ago. This delivers a clear message: while the dust may have settled on those infamous ad misplacements, brand safety remains paramount for the big spenders. Moreover, protecting brands is still cause for concern across the wider industry, with 94% of advertisers worried about ad context.
Yet it seems the industry is still struggling to find a way of ensuring brand safety that works for all. Currently, the challenge is multifaceted; finding the right technological solution is one issue, while creating suitable content for ads to run beside is another. From the advertiser perspective, the ‘go-to’ tools are negative keyword lists, or black and white lists; but these rigid structures don’t provide the flexibility or accuracy required to support a fruitful relationship between publishers and advertisers.
What is a false positive?
Negative keyword checklists also have a tendency to trigger false positives, which miss the true context of a page, limiting reach for advertisers and delivery for publishers by inadvertently blocking safe content when a negative keyword is found, but in a positive context.
Take for example the headline: IKEA kills it in retail! This is clearly a positive story about the success of the Swedish retail giant, which would usually be considered desirable inventory for many brands. However, a keyword technology would likely want to target away from content containing the word ‘kills’, perceiving it as negative and unsafe for most brands. Consequently safe pages, or even entire websites, where the context is lost due to a keyword-only approach are blocked. The impact of this on publishers’ reach and advertisers’ delivery distorts campaign and revenue performance for both parties.
Tabloids and breaking news
Hard-hitting news is a part of everyday life, whether consumed via the radio, online, TV or newspaper – everybody wants to know what is happening in the world. But while these may seem to be obvious places for brands to target a huge pool of potential consumers, breaking news is never a clean-cut page; often containing controversial, negative or upsetting content.
When advertisers run searches for negative keywords and banned terms on breaking news and tabloid sites, it’s not surprising that they come up with multiple red flags. Indeed, these factual reports are often automatically avoided, as common words such as death, crash or sex are guaranteed to set off sensors. As a result, keyword tools will see many terms that are completely harmless within the context of a story cause ad placements to be unnecessarily avoided. In fact, many advertisers avoid advertising on news content entirely, due in part to the perceived complexity of avoiding negative content.
Though this may seem like a drop in the ocean when considering the vast amount of content available on the open web, cutting off tabloid content significantly reduces advertiser reach — as well as reducing essential revenue flowing into news sites. Plus, when considered as a wider trend, the use of keywords and banned lists is damaging to the entire online content landscape.
As more terms are unwittingly added to broad ‘bad word’ compendiums, there will be very few words left that are still in the safe zone. So, to avoid losing more ad spend, publishers may have to compromise editorial integrity and move towards generating acceptable content for most advertisers. Although this may sound alarmist, we should not lose sight of the threat publishers (especially news portals) and an informed consumer internet face, if their ability to grow advertising revenue is hampered further.
While no one disputes that brand safety is of key concern and advertisers have every right to be firm with their requirements, this doesn’t necessitate putting pressure on publishers. Instead of forcing content producers to censor their output, advertisers and publishers should be employing intelligent tools that not only help improve content for consumers, but also boost reach for campaigns and improve the overall revenue opportunity for publishers.
For example, solutions have been developed that use artificial intelligence (AI) to take a closer look at content. These cognitive semantic technologies conduct in-depth content assessment at page level; using natural language processing (NLP) that understands grammar and the complex relationships between words in the same way the human brain does, picking up subtle differences in context that can completely change the meaning of words.
By using these advanced tools, advertisers can judge whether the sentiment and subject matter of content make it safe for their particular brand and related values, based on its true meaning. This increased insight will prevent ad misplacement, improve audience targeting and the effectiveness of the campaign, and avoid false positives.
Protecting brand safety on an ever-evolving entity like the internet will never be a one-dimensional process. Consequently, it is vital the industry works together to sustain the power of both journalism and advertising. To achieve brand suitability and brand safety requires a two pronged approach, making sure that only content that is unsafe for your specific brand is blocked, and using the right AI technology. By doing this you will open up a plethora of inventory, avoid false positives, and ensure the industry can progress in a mutually beneficial way.
Find out more about Digital Marketing World Forum (#DMWF) Europe, London, North America, and Singapore.
- » For brands with purpose, the future is authentic
- » Mind your language: Using multilingual websites to increase eCommerce engagement
- » Brand suitability: Exploring the next generation of brand safety
- » The 2019 data outlook sees spending rocket and multi-channel key amid regulatory concerns
- » Barclays: Agencies need to evolve as digital marketing spend rises – but don’t get too close to the behemoths