The State of the Art in Contextual Video Intelligence and Targeting
Beyond genre: Meet the new standard for contextual advertising on CTV
When Harry met Sally
Contextual advertising at its best
Meet Sarah, an avid runner who gained the confidence to sign up for her first marathon after years of jogging for fitness. She just searched online for "shin splints during marathon training” – just three weeks into her training program, the pain shooting down the front of her lower legs in becoming unbearable. She clicks on a fitness blog that explains the common causes and solutions. Right there in the article, she sees an ad featuring the most popular running shoes for preventing injury among long distance runners embedded in a listicle of expert recommendations. She clicks. She buys. She feels excited and hopeful about continuing training. The ad worked because it found her in the perfect moment—one when she was already thinking about running, already engaged with relevant content and primed to take action towards solving her problem.
This is contextual advertising at its best. Content and commerce aligned naturally. Sarah didn't feel interrupted or sold to as she searched; she felt helped.
Contextual advertising is the art of placing the right message in the right moment. It relies on content—keywords, subject matter, themes—rather than user behavior or personal data. When done well, it feels native and relevant in a way that improves both effectiveness and user experience.
Think about the last time you searched for instructions to complete a simple home repair such as "how to fix a leaky faucet" or “how to find wall studs” and immediately saw ads from home improvement retailers. That's contextual advertising at work—meeting you exactly where your interest and need intersect. It doesn't feel pushy or random. It feels like help arriving right in time.
On the web, contextual advertising is everywhere. Hiking boots appear on outdoor adventure blogs. Cooking equipment shows up alongside recipe articles. Travel insurance ads run next to destination guides. The power lies in relevance. When someone is already thinking about fitness, they're more receptive to fitness-related products. When they're reading about travel, their mind is open to luggage, hotels, and travel experiences.
But here's what makes it even more powerful: contextual advertising doesn't have to work alone. Smart marketers layer it with other targeting methods. A car brand might use first-party data to find people already shopping for SUVs, then add contextual targeting to reach them during family road trip content or adventure-focused articles.
Across web and mobile environments, contextual advertising is a standard part of campaign planning. But in connected TV (CTV), the concept is less familiar and often misunderstood. Many advertisers don't realize that contextual advertising is even possible on CTV. And until recently, that misunderstanding wasn't entirely wrong.
Meet the Flintstones
Moving past the primitive era of CTV contextual advertising
It's fair to say we have been living in the primitive days of CTV contextual, but that is not the future. Back to our friend Sarah. Now she’s at home after a day at work and training in the evening. She settles in to watch Silver Linings Playbook on her smart TV. Jennifer Lawrence's character is out for another training run, pushing through the Philadelphia streets with determination. The scene captures the same determined spirit Sarah has as she pursues the marathon. But instead of a message showing fitness, adventure or the drive to succeed in the ad break, there’s a diaper commercial.
That running scene in Silver Linings Playbook? It gets labeled as "romantic comedy” and misses the athletic opportunity entirely.
This is the gap that exists in CTV today. While the web has sophisticated contextual advertising that matches content and brand messages, CTV contextual advertising has been stuck in the Flintstones era—still largely relying on broad, program-level information from a show or movie’s genre or text-based description.
Until recently, the contextual advertising opportunity in CTV has been hamstrung by technology that limited advertisers to program-level metadata such as genre and category. This broad-brush approach made sense when technology couldn't do better. But advances in AI make relying on program-level information alone a bit like using a paper map to navigate today’s roads. It's technically functional and can get you places better than directions from the clerk at the gas station, but it misses all the information you can get in Google Maps—the nearest restaurants, traffic, accident reports, and so much more.
Consider innovation patterns that have transformed other industries. Electric vehicles existed for decades before Tesla made them desirable and accessible to mainstream consumers. Now, people buy Teslas not just for environmental reasons, but because they want the features they offer. Smartphones existed before the iPhone, but Apple's breakthrough device made the entire category explode. The same catalyzing moment is happening now with AI. With it, the very definition of contextual advertising in CTV is changing.
A New Hope
Entering the Golden Age of AI
AI stands to change almost every facet of human life. New tools are created every day, with the AI market projected to grow at a compound annual growth rate (CAGR) of 28.46% from 2024 to 2030. But the most exciting thing about AI isn’t the latest new use case.
Here’s why AI represents such a breakthrough: it can accomplish tasks that would be impossible for humans alone to execute at the scale modern society demands.
From a business standpoint, artificial intelligence allows tasks to be carried out with precision at speeds and volume that wouldn't be feasible with humans alone. While it's common to worry about AI replacing humans, what we're seeing is AI allowing humans to do new tasks that would not have otherwise made sense because of the sheer human power required to execute them adequately.
Consider the simple task of reading. AI can analyze thousands of pages in minutes—a task that would take humans many hours to complete. It can process legal documents, research papers, and books at speeds that make previously impossible projects suddenly viable. Financial firms use AI to read through millions of earnings reports, news articles, and analyst notes to identify market patterns that no human team could spot. Medical researchers deploy AI to analyze thousands of research papers simultaneously, finding connections across studies that might take human researchers months to discover.
But for CTV advertising, the challenge is far more complex than processing text.
The Matrix
CTV, meet multimodal AI
Video content presents a unique challenge for computers: instead of numbers or text, video inherently contains multiple different elements that need to be understood simultaneously to get the full context of what is happening in any scene. Because of this, video requires a sophisticated form of artificial intelligence called multimodal AI that can understand video in much the same way humans process what they watch. It can understand the visuals, the dialogue, the music, the emotions, the objects on screen, even the subtle sentiment shifts that occur moment by moment.
Now imagine trying to hire enough people to watch every hour of content streaming across platforms today to find a special moment like Jennifer Lawrence running in Silver Linings Playbook and adding the metadata to the video specifically so it can be identified as an advertising opportunity for a running shoe brand.
For a single movie, this analysis might take a human expert days to complete thoroughly. Now multiply that by the millions of hours of content across FAST channels, SVOD platforms, and streaming services. The math becomes impossible. No company could afford to hire enough experts. No team of humans could work fast enough to keep up with new content being added daily.
This is where multimodal AI proves transformative—not by replacing human judgment, but by making human-level analysis possible at previously unimaginable scale.
Here's another crucial distinction: multimodal AI doesn't just identify that a dog appears on screen. It understands subtle emotional context like a human viewer does. The beloved family pet in Marley & Me creates a completely different advertising environment than the trained assassin's companion in John Wick. Both films feature dogs prominently, but the sentiment, tone, and brand suitability couldn't be more different.
Unlocks monetization
opportunities
This emotional intelligence unlocks monetization opportunities that publishers couldn't access before.
A pet food brand would never buy genre-targeted inventory in a romantic comedy. But with scene-level analysis revealing the movie's premium pet content, that inventory becomes valuable to a wider variety of advertisers and brands. For advertisers, this precision eliminates the guesswork and risk of broad category targeting while opening entirely new creative possibilities. For publishers, it unlocks new monetization opportunities. For viewers, it means more resonant, less disruptive ads.
Almost Famous
Your CTV targeting playbook
As you explore contextual advertising for your CTV campaigns, here's what to look for and what questions to ask:
Ask the Right Stuff.
Instead of simply targeting female viewers aged 25-54 for your women's clothing brand, consider the scenes that would resonate: the iconic shopping montage in Clueless, the fashion industry moments in The Devil Wears Prada, or the style transformation scenes in romantic comedies. Your running shoe campaign shouldn't just target sports programming—look for characters jogging in films like Silver Linings Playbook or Forrest Gump.
Think Beyond The Usual Suspects
When evaluating contextual platforms, dig deeper than surface capabilities:
Can it analyze actual scenes, not just programs or genres?
Does it understand emotional context, not just objects on screen?
Can you set parameters for tone and sentiment to ensure brand suitability?
Does it explain why an ad was placed in a specific moment?
Was it built specifically for streaming video, or retrofitted from web advertising?
Plan for Layered Targeting
The most successful contextual campaigns combine scene-level intelligence with audience targeting. A pharmaceutical brand might target health-conscious consumers while also seeking scenes about resilience, family support, or medical breakthroughs. A financial services company could target homebuyers while looking for scenes featuring beautiful homes, family milestones, or characters making important life decisions.
Look for Transparency
Your platform should explain its decisions. When your ad appears next to content, you should understand why—what specific elements in that scene made it relevant for your brand.
The ContextIQ Chronicles
From interruption to resonance
Context changes everything. Anoki's ContextIQ platform represents a fundamental shift in how CTV advertising can work—moving from interruption to resonance by ensuring ads appear at the most impactful moments.
Scene-Level Intelligence Across Multiple Dimensions
ContextIQ analyzes every scene across multiple signals: visual elements, objects, dialogue, music, emotion, sentiment, and narrative context. This creates opportunities that genre-based targeting simply cannot access.
Purpose-Built for CTV
Unlike web-based tools adapted for video, ContextIQ was designed from the ground up for streaming television, understanding the unique characteristics of long-form content and the connected TV viewing experience.
Results in Action
The proof is in the performance:
A QSR brand targeting "pizza occasions" saw a 59.5% uplift in shop rate and 61% increase in new shoppers by reaching viewers during late-night snacking scenes, sports watching moments, and social gatherings.
A financial services brand achieved a 13.2% increase in brand consideration by aligning with life milestone scenes—weddings, career changes, budgeting conversations, and family planning moments.
A pharmaceutical brand targeting health-conscious female audiences saw a +28% lift in awareness and +16% in consideration by appearing during hopeful, family-centered, and health-focused scenes.
These results didn't happen because of better genre targeting. They happened because the message met the moment with unprecedented precision.
Back to the Future
The turning point is now
Contextual advertising in CTV exists, but we're at a pivotal moment where it can finally deliver on its promise. The technology has caught up to the opportunity. Multimodal AI makes human-level content analysis possible at the scale modern CTV advertising demands.
This shift comes at a critical time for the industry: CTV is the fastest-growing major channel for marketers today, with ad spend projected to reach $48 billion by 2028. To continue relying solely on broad genre categories is to leave massive opportunities on the table—opportunities that are sitting there right now, waiting to be activated.
This isn't just about improving ad targeting. It's about creating a better experience for everyone: advertisers get more relevant placements, publishers unlock hidden value in their content libraries, and viewers encounter ads that feel like natural extensions of what they're already watching.
This isn't just about improving ad targeting. It's about creating a better experience for everyone: advertisers get more relevant placements, publishers unlock hidden value in their content libraries, and viewers encounter ads that feel like natural extensions of what they're already watching.
At Anoki, we're building this future every day—one scene, one moment, one perfectly placed ad at a time. Because when advertising fits the moment, everyone wins.