The Appeal of Neuro-Contextual Advertising: A Q&A With Seedtag’s Brian Danzis
Seedtag is an ad platform that introduced “neuro-contextual” advertising that is powered by its AI, Liz. “We decode real-time interest, emotion, and intent to make context the foundation of planning and activation,” the company states. “Our vision is a world where advertising is built on context and moves beyond surveillance to create experiences that are relevant, respectful, and genuinely human.” Seedtag uses minimal personal data to provide an alternative, neuroscience-based approach to serving relevant ads.
I spoke to Brian Danzis, Global Chief Revenue Officer, about what’s involved in Seedtag’s technology and where the company’s vision fits into the streaming industry as it evolves.
Brandi Scardilli: Why does Seedtag believe that moving beyond broad audience targeting is important? Tell me about the company’s shift to employing a neuro-contextual advertising approach.
Brian Danzis: Traditional audience targeting often relies on broad demographic or behavioral signals, based on the assumption that someone’s age, gender, or income reveals their passions, or that past interests accurately predict what will capture their attention today. But people aren’t static, and neither is their attention or priorities.
With third-party data increasingly restricted across the digital ecosystem, particularly in CTV and in-app environments, brands and agencies are looking for ways to deliver personalized, scalable advertising with precision, beyond the limits of personal data. In fact, as EMARKETER shared in its Contextual Advertising Trends 2026 report, 50% of privacy-regulated industries in the U.S. rely on contextual as their primary method of ad targeting.
Neuro-Contextual advertising is an evolution of the contextual targeting Seedtag has been offering since we started in 2014. We think the industry has been asking the wrong question. Instead of “Who is this person?”, we ask “What does this moment mean to them?”
Traditional contextual advertising tells you what a piece of content is about, often by analyzing keywords. Neuro-contextual advertising, based on neuroscience, goes beyond traditional contextual targeting to capture the moment itself, decoding signals of interest, emotion, and intent in real time. It helps brands understand what someone is engaged with right now, rather than relying on keywords, demographics, or past behavior. This enables more personalized advertising based on the moment, not the individual.
Seedtag’s main technology is a proprietary AI named Liz. What challenges did you run into with creating this AI, and how do you monitor its continuing accuracy?
The biggest challenge was recognizing that understanding what a page or video is about isn’t the same as understanding how people will respond to it. Two video scenes can cover the same topic and trigger completely different reactions: One might spark curiosity, while the other creates frustration. Standard contextual AI models can struggle to distinguish between these nuances.
To address this, Seedtag built its own Neuro-contextual embeddings specifically for media, drawing on more than a decade of contextual intelligence, analysis of more than 100 million pages and 100,000-plus videos and shows. We’ve expanded beyond standard IAB categories to more than 11,000 contextual classifications, with signals of interest, emotion, and intent built into Liz’s understanding of the content.
Accuracy is a continuous process. Real campaign performance, validated by third-party measurement partners, provides feedback that helps us refine Liz and understand which contextual matches perform best. We also test her interpretations against real human responses, including through a neuroscience study with a renowned neuroscientist. And because we built Liz entirely in-house, our 100-plus engineers and data scientists have direct control over how she is trained, evaluated, and refined, including monitoring for potential bias.
Liz connects “real time content signals to broader user interests, helping brands reach people based on what truly matters to them.” What are some of these content signals and user interests? How does this technology make CTV more dynamic?
Liz looks beyond what a piece of content is about to understand the broader context around it. At the core are three signals: interest (what the content is relevant to), emotion (the emotional tone it conveys), and intent (whether someone is browsing, learning, or close to taking action). This becomes particularly valuable in CTV, where the signals available to advertisers can be relatively limited. Liz can turn fragmented data points into a richer understanding of the viewing environment and the interests it connects with. We can then extend that understanding beyond the TV screen. By connecting CTV signals with insights from the open web and querying trusted sources for additional content metadata, Liz can fill in gaps and build a more complete contextual picture of each opportunity.
In practice, this means moving beyond the basic information typically passed in a CTV bid request, such as a domain, IAB category, or keywords. Liz can enrich each opportunity with additional context, including the show name, genre, interests, topics, intent, emotions, and brand safety, making it more meaningful and actionable for brands.
How do you think the neuro-contextual approach to serving relevant ads fits into the evolving state of advertising on CTV and streaming?
CTV has the attention of the living room, but it doesn’t always have the identity signals advertisers have traditionally relied on elsewhere. Across parts of the streaming ecosystem, user-level signals can be limited or fragmented, making it harder for buyers to balance precision and scale. That’s where context becomes increasingly valuable. Every stream carries its own signals: what’s being watched, what it’s relevant to, its emotional tone, and the intent it may indicate. These signals are available at the content level, without relying on individual identity, and can provide a consistent foundation for reaching viewers based on the moment they’re in.
The shift we see is from buying TV primarily by audience demographics to buying it by the moment. People don’t necessarily have the same interests when they’re on their phone during the morning commute versus late at night on the sofa, even if they’re consuming the same content.
How does Seedtag prioritize protecting company and viewer privacy?
Privacy isn’t a feature we added. It’s how Liz was built. Seedtag was born contextual, so we’ve never relied on cookies, device IDs, or personal data. Liz reads the content, not the person.
We comply with data protection laws such as GDPR and CPRA, while developing Liz in-house with rigorous reviews and controls around how the AI is trained, evaluated, and refined. This gives us direct oversight of how the system is developed and helps us identify and address potential bias.
For brands, protection also means understanding where their ads appear. Liz’s brand safety classification goes beyond traditional blocklists by understanding the context around content, helping brands avoid unsuitable environments without unnecessarily excluding quality content.
What else should brands know before signing up to work with Seedtag?
First, bring us the brief, not just the audience. Our planning starts with what a brand wants to achieve and the moments where that message will land best. The Liz agent translates the brief into the content, emotions, and intent signals most likely to deliver it.
Second, this isn’t only an awareness play. Neuro-contextual works across the full funnel. United Airlines saw a +48% lift in Recognition with its CTV campaign, while Conagra Brands achieved a 3x return on ad spend (ROAS).
Third, NeuroX is how we bring Liz’s neuro-contextual intelligence to market at scale. NeuroX connects Liz’s intelligence across 30,000-plus premium publishers and broadcasters, helping us understand every advertising opportunity, whether or not identity is available. There are several ways in: fully managed campaigns, curated deals, or premium marketplaces across CTV, video and the open web.
Related Articles
Audience data, reach, and targeting still matter. But the strongest campaigns aren't choosing between those things. They're targeting within environments where people have already told you what they care about.
29 Jul 2026
Marketers are reconsidering how to capture attention and provide significant outcomes as consumers shift to streaming platforms. At this year's Advertising Week New York, Reed Kiely, Director, Data Insights & Trends, VAB, moderated the panel "Streaming's Strategic Advantage: Precision. Scale. Results" with panelists from Seedtag, Affinity Solutions, Premion, and Digitas.
17 Oct 2025
The 2024 edition of Streamticker, recapping the streaming industry's most momentous mergers and acquisitions of 2023, kicked off with Disney's gobbling up the last extant portions of Hulu, as Comcast ceded its remaining 33% stake. But if the late-2023 deal marked only the quiet conclusion of an already-silent partnership, the much noisier news of Jan. 6, 2025, found Disney absorbing sports-centric streamer Fubo and merging it with Hulu Live + TV. Here we'll review this and other done (or more nearly done) M&A deals that reshuffled the streaming industry in 2024.
28 Mar 2025