Why Contextual AI Will Shape the Future of Advertising

A privacy-first and cookie-free way to engage consumers

As marketers hurtle towards a privacy-first future, the industry is being flooded with countless stories about new approaches to consumer engagement that will soon shake up the advertising world. What if they’re all wrong? What if we need to go back to basics? What if it’s time to return to contextual targeting?

Marketing pros would understandably have some concerns. Contextual targeting has been around for a while and fell out of favor with many in the industry. But its latest iteration––one driven by AI––is a more powerful tool than many realize. 

For decades, contextual was considered a necessary yet simplistic targeting tactic. When cable TV was at its peak, many niche channels––especially smaller ones––were using context as a proxy for audience. These contexts were based on sweeping generalizations about what an audience was likely to be interested in: men watched sports networks, women watched lifestyle channels, for example. But they often worked. Niche networks attracted small but highly engaged audiences, and brands wanted to align with people’s passions.

With the rise of the internet came the birth of the cookie, and the death of contextual as we knew it. Brands no longer needed to generate proxies for their audience; the data they collected meant they could make a fairly accurate educated guess as to who their target market was and where it could be found. As social media grew in popularity and scope and screen time increased, advertisers had even more time to serve impressions.

So, what now?

This has led to where we are today. Audiences are progressively fractured, our attention is increasingly precious and, most importantly, users are sick and tired of their data being used to serve ads that follow them around the internet. Platforms are no longer supporting third-party cookies and legislation is coming into effect around the world to better protect consumer privacy. 

Marketers need to go back to basics and return to contextual. But this time they’re going to do it bigger and better.

Like all forms of advertising, contextual has been hamstrung by standardization and strict categorization. In a way it makes sense: If advertisers can place people in convenient little buckets, it strips away a lot of complexity and makes targeting simpler. But it lacks nuance. They would be, once again, making sweeping assumptions about people without identifying what they’re actually interested in.

Let’s say, for example, a brand has a specific campaign brief and is looking to target people who are environmentally conscious but also interested in fashion. This is when traditional contextual approaches favor quantity over quality. Targeting sustainability may mean targeting content related to electric cars or renewable energy, but that has nothing to do with fashion. It’s still too blunt and lacks the context to be truly effective.

The power of AI

But that’s all about to change. Advertisers that are finally reallocating some of their budgets back to contextual are riding the latest wave of contextual targeting innovation. Contextual targeting is now being driven by AI that has been trained using supervised machine learning techniques by being fed millions of articles over the course of years. Premium AI solutions are capable of unsupervised learning, meaning they can now identify patterns in content completely independently.

All of this means AI is not only capable of thoroughly analyzing articles, but entire networks of content across different countries, languages and across standard categories. But it also allows advertisers to go beyond them, using those advanced machine learning models to do custom targeting specific to that brief, which is 92% more accurate than traditional targeting. 

Let’s go back to the fashion example. When targeting using the latest AI and machine learning models, the brand’s ad will be placed alongside articles related to a topic like fast fashion, because advertisers know that’s where the ad’s going to resonate most with the target audience.

Why should brands care?

When a brand works with a premium contextual partner, it gains access to huge amounts of data. If a brand were to collect all this data itself, it would require enormous resources and expertise. This data is precious, as it allows brands to know exactly what consumers are interested in like never before. But it also offers opportunities for scale. If a brand’s brief is too narrow, it can expand the size of this custom category while ensuring it’s still relevant to the brief. Conversely, it can narrow the category to ensure the most precise audiences are targeted.

Demographic targeting is clunky and has been changing perceptions of advertising for the worse. Consumers automatically associate all ads they see with the invasive cookie-driven techniques they’ve become all too familiar with, which leads to negative perceptions of the brands whose ads they see. AI-powered contextual targeting gives advertisers the power to turn the tide. As marketers continue hurtling towards a cookieless future, it is the only way to make advertising less awkward, more relevant and ensure brands are seen in a more positive light.

Brian Danzis leads U.S. operations at contextual advertising company Seedtag. He has extensive experience in digital advertising, having led agency solutions for VideoAmp, served as head of video and live event sales for Spotify, and as CRO of Virool.