With addressable impressions on the open web set to fade out of view, advertisers are increasingly looking to contextual targeting to deliver campaign results without using personal data or identifiers.

Those who’d written off contextual a decade ago as static, limited and outdated, may find themselves pleasantly surprised. For, while the industry has been busy profiling and following audiences based on past browsing behavior, some interesting new capabilities have emerged from the contextual camp which are offering alternative ways to look at relevance.

Marketing Technology News: MarTech Interview with D. Scott Bowen, CEO at Tenovos

This is mostly being driven by developments around keyword targeting – the method traditionally used to match brand to page topic in the hope of finding relevant users.

On the one hand, more nuanced metrics are now being layered onto this process which offer deeper ways to categorize content into segments. Meanwhile on the other hand, keywords and third-party audiences are being partly replaced in favor of using live contextual insights to work in real time.

Keywords and page categorization

Keyword targeting has traditionally been useful for marketers looking to guarantee the contexts within which their ads will run. But keywords also limit a campaign unnecessarily, narrowing it into themed corners which take no account of trending topics, nor the fact that individuals have multiple interests and may notice ads in multiple and non-matching environments.

Accordingly, work has gone into developing more effective ways to segment content. Specialist providers are now categorizing pages by looking at things like implied sentiment and intent, and using tools which can also read images, videos and metadata. Some use machine learning and natural language processing (NLP) to establish the weight of themes and phrases. Others have the capability to work across languages.

It’s also increasingly possible to gauge user mindset from context, by scoring pages according to whether a person is likely to, for example, be browsing with intent to buy or inform themselves; whether they’re researching, evaluating or using content with a view to creating something. All can be tagged using complex algorithms and their likely influence on campaign performance measured and understood.

Marketing Technology News: MarTech Interview with Rachel Meranus, CMO at Transfix

Real-time contextual analysis

At the very vanguard of AI-powered contextual, it’s possible to untether campaigns from keywords altogether and instead use this context-mindset influence along with live data to steer campaigns in real time. No segments or static lists – instead, pivoting on the fly to target contexts based on what’s working for a campaign while it’s running.

Of huge interest to marketers are the growing insights coming back from these more fluid contextual campaigns.

First up, the content itself which someone is viewing when they show interest in an ad can give a good indication of where other relevant audiences might be found online. This lookalike-style relevance is essentially contextual targeting powered by live browsing behavior – and it’s a giant step forward when the industry needs it most.

Secondly, the contexts which are proving effective for a campaign can also pinpoint the less-obvious, unpredictable interests of a brand’s potential customers, which can be incredibly useful when planning wider marketing activity.

What’s more, targeting powered by mindset, sentiment and NLP can transcend seamlessly into newer areas such as CTV and audio.

However ambitious your outlook, the contextual sector is now collectively offering some interesting solutions to the thorny issues of how to overcome the limits of keyword targeting, and how to find relevant audiences without using cookies.

Right time, right technologies – contextual itself has never been more relevant.

Marketing Technology News: MarTech Interview with Deepak Anchala, Founder and CEO at Slintel