Contextual advertising is moving closer to the moment.
In its new Contextual Advertising Trends 2026 report, eMarketer looks at how AI is expanding the role of contextual advertising across the media mix. As AI gets better at understanding meaning, intent, emotion, and tone, contextual intelligence can inform far more than where an ad appears. It can shape planning, creative, commerce, and brand suitability.
In video, that evolution is happening at the scene level. Instead of understanding a show or episode as a single piece of content, AI can interpret what’s happening moment by moment, bringing together signals like objects, dialogue, location, sentiment, and emotional tone.
That level of precision creates new opportunities for advertisers and more questions about how contextual intelligence gets activated at scale. KERV Chief Strategy Officer Marika Roque spoke with eMarketer about both for the report.
Context Is More Than a Signal
Consider something as simple as a knife.
As KERV Chief Strategy Officer Marika Roque explained to eMarketer, a knife in MasterChef means something very different from a knife on Dexter. Identifying the object is one signal, but understanding the scene around it is what gives that signal meaning.
KERV analyzes video at the scene level, bringing multiple signals together to understand not just what content is playing, but what’s happening in a particular moment and whether it’s suitable for a particular brand.
A cooking show may be broadly relevant to a food or kitchen brand, but a contestant celebrating a win, preparing a meal, or opening a refrigerator creates three very different opportunities.
The more precisely we understand the moment, the more precisely we can determine what belongs next to it.
When Context Shapes Creative
Understanding the moment creates opportunities that broader contextual targeting can’t.
KERV’s work with Warner Bros. Discovery and Wayfair is one example. KERV identified relevant moments within content and dynamically matched products from Wayfair’s catalog to what was happening around each ad break. So the products featured in the creative changed based on the content a viewer had just watched.
Across thousands of pieces of content and ad breaks, that turns contextual intelligence into potentially millions of individual product and creative decisions. It also shows where contextual advertising is headed: beyond simply deciding where an ad belongs to helping determine what should appear there.
From Targeting Signals to Defining Intent
Today, contextual targeting still relies heavily on predefined taxonomies. Marketers select the categories and signals that best represent the environments they want to reach. But the eMarketer report points to a different approach emerging: using natural language and agentic AI to translate campaign objectives into contextual strategies.
Imagine a brand wants to appear in reality TV, but avoid moments involving excessive drinking. Rather than manually translating that objective into a fixed set of categories, an AI agent could understand what the marketer is trying to accomplish and identify the combination of signals and moments that fit.
Instead of requiring marketers to define context signal by signal, the technology can start with the campaign objective and identify the moments that match it.
That could give marketers much more flexibility in how they define relevance and suitability, without requiring them to anticipate every signal or scenario in advance.
Understanding the Moment Is Only Half the Job
More precise contextual intelligence also puts more pressure on the infrastructure behind it. As eMarketer notes, fragmented taxonomies, interoperability, latency, and activation remain barriers to putting advanced contextual signals to work across the media ecosystem.
Understanding a moment is one thing. Making that intelligence available quickly enough to inform an advertising decision is another.
As contextual gets more granular, the systems used to activate it have to keep pace.
The industry has spent years getting better at understanding content. Now AI can understand the moments within it with far greater precision. The next challenge is making that intelligence actionable wherever advertising decisions are made.