The CTV Trends Report

Share:

Introduction

Ad-supported streaming TV is entering its next phase of growth. Daily hours of viewing are up nearly 12% year over year, with a growing number of households tuning into ad-supported content daily.

But as streaming audiences and inventory continue to expand, advertising dollars are lagging behind the audience. According to eMarketer, CTV now accounts for about a fifth of the time adults spend with media each day, but less than a tenth of ad budgets. One reason spurring this disconnect is the dearth of information available to buyers about what they’re actually buying. CTV still lacks consistent, scalable access to signals that buyers have come to expect across other media channels, such as information about the content the ads will appear within.

And the industry is already feeling the effects of that disconnect. According to recent research from Wurl and AdExchanger, 79% of respondents said inefficiencies around data quality, fragmentation, and interoperability are limiting the scale and effectiveness of streaming TV campaigns.

Wurl’s analysis found significant gaps in the metadata flowing through the CTV bidstream. An overwhelming share of inventory is missing basic program information, genre classifications are fragmented and inconsistent, and even when metadata is available, it doesn’t always accurately reflect what’s actually on screen.

Key findings:

89% of ad requests were missing a usable program name

3,600+ different genre labels appeared in a two-hour window

35% of analyzed inventory carried a declared genre that did not align with what Wurl detected in the actual program

Together, these findings point to one of the fundamental challenges facing CTV advertising today: The information buyers need to understand and value an impression doesn’t always make it into the bidstream.

That’s a real constraint for a channel that has become increasingly sophisticated since the days of traditional television – enabling more granular targeting, greater flexibility in ad delivery, and real-time, data-driven optimization that can drive more precise and effective campaigns. Buyers can apply advanced targeting strategies to streaming campaigns, but those capabilities depend on the quality and availability of the signals attached to each ad opportunity.

And today, those signals often paint an incomplete picture. 

In this report, we examine the gap between what viewers are watching and what buyers can see – and why better content intelligence will be critical to CTV’s next phase of growth.

Growth continues as the market matures

Ad-supported streaming TV continues to attract significant audiences.

According to Wurl’s analysis, comparing viewership trends over the first eight months of the year, between January and August, hours of viewing have increased nearly 12% year over year from 2025 to 2026.

Of note, this growth in streaming viewing hours has been primarily driven by new households turning to streaming, while viewing time per household has remained relatively stable. These trends point to a CTV ecosystem that continues to expand while entering a more mature phase of growth – one where the opportunity will increasingly be in creating more value from the viewing already taking place.

At the same time, the content itself is evolving. For instance, publishers and rights holders are increasingly leaning into FAST (free ad-supported streaming TV) as another way to reach streaming audiences, bringing more premium programming – like live sports – into the ecosystem. As the value of the content increases, so does the need to ensure that value is visible to buyers. A premium viewing environment can only command its full value if advertisers have enough information to understand what they’re buying.

To attract advertiser investment, the industry has to look beyond making impressions available and start accounting for their quality and context. Buyers need to know what they are purchasing, publishers need ways to demonstrate the value of their inventory, and programmatic platforms need richer signals to inform increasingly sophisticated decisions.

And that starts with something surprisingly basic: knowing what content is actually on screen.

FAST has spent the past several years proving that audiences will show up. As the market matures, the next question is how we create more value from that viewing – for advertisers, publishers, and the broader streaming ecosystem. That means getting smarter about the inventory itself and giving buyers a clearer understanding of what they’re buying.

Dave Bernath

CEO

Wurl

The visibility problem in programmatic CTV buying

Pick an ad opportunity in free ad-supported streaming TV and ask two basic questions: 1) What channel is this airing on? And 2) what program is the viewer watching?

The answers aren’t always available in the programmatic ecosystem. Wurl analyzed ad requests flowing through the CTV bidstream and found substantial gaps in these foundational content signals.

45% of ad requests had no channel name, while 89% were missing a program name. That creates a fundamental disconnect: the viewer knows what they’re watching and the publisher knows what content is being delivered, but by the time the corresponding ad opportunity reaches a buyer, much of that information can be missing or obscured. And even when that information is available, inconsistencies in how content is identified, labeled, and shared across platforms and partners can create additional challenges. Different data-sharing agreements and levels of access only add to the complexity, making it difficult to establish a consistent, reliable view of the content behind every ad opportunity.

That matters because better information enables better decision-making. Audience data can tell an advertiser something about who may be watching, but content intelligence adds another dimension: what that viewer is watching in that moment. Without reliable program- and content-level information, buyers have fewer signals to evaluate and optimize their campaigns for impression quality, suitability, and relevance.

Genre data doesn’t tell the full story

Genre is one of the most familiar ways to describe television content – and one of the most common contextual signals available to CTV buyers. But because of CTV’s fragmented supply chain, that signal doesn’t always arrive the way it started. As an ad opportunity moves from a publisher through SSPs to demand partners, each platform may receive different information – or apply its own taxonomy and requirements to organize it. Along the way, useful content signals can be changed, simplified, or lost before they ever reach the buyer. Wurl’s analysis found that the genre data reaching buyers can be nonstandardized, vague, and, in some cases, inconsistent with the content actually on screen.

To illustrate just how complicated things are for advertisers, we took a look at the genre signals during just two hours of prime-time ad requests. In those two hours, more than 3,600 different genre labels appeared. Often, those labels described similar content in different ways: for example, Drama could appear as “drama,” “genre_drama,” or “crime drama,” while Reality programming might surface as “reality,” “reality shows,” “reality & game show,” or “genre_reality_tv.” While this may be the result of a well-meaning effort to help advertisers find relevant content, on the flip side it places an operational burden on the ecosystem and risks limiting deal sizes.

On the opposite end of the genre spectrum, broad classifications were also common. 22% of genre-labeled bid requests carried at least one vague genre label, such as “entertainment” or “general variety.” More significantly, 15% of all requests offered buyers little to no genre information: 8% carried only vague labels, while another 7% had no genre at all.

More problematic, even when a more specific genre was available, it didn’t always align with the underlying content.

When compared to genre data indicated by FAST EPGs (Electronic Program Guides), 35% of ad inventory carried a declared genre that had nothing in common with the genre detected in the actual program. One reason is that genre is sometimes assigned broadly at the channel level, rather than the program level. In roughly half of the mismatched cases, the stream used a general classification, such as “entertainment,” while the other half carried a specific genre that was incorrect or too coarse.

For example, in some cases, programming Wurl identified as Crime, Mystery, and Thriller was labeled simply as “entertainment” in the bidstream. The same content can also carry different genre classifications depending on which supply partner sends the ad request, as each may apply its own taxonomy. A True Crime docuseries, for instance, could be classified as “documentary” by one partner, “crime” by another, and “drama” by a third – making the same program appear to buyers as three different types of content.

Top 30 genre labels by share of genre-labeled requests for one prime-time evening (1B+ labeled requests)

For buyers, these inconsistencies can make very different viewing environments look interchangeable – and make it harder to apply targeting strategies uniformly across CTV supply. A signal is only as actionable as its accuracy and consistency.

And, ultimately, genre itself only tells part of the story. A Drama can move between suspense, humor, and sadness within minutes. Similarly, a News program can shift from a serious story, to weather, to an uplifting segment. As contextual targeting becomes more sophisticated, understanding what’s happening within the content in real time the moment right before the ad break – not simply how the overall program is categorized – can give buyers a more precise picture of the environment surrounding an impression.

Buyers appear ready for that greater level of precision. In Wurl and AdExchanger’s research, real-time content analysis was the number-one capability respondents said would help advance contextual targeting in CTV, cited by 39%. Another 29% of respondents pointed to more granular content categorization.

Content intelligence solutions exist – so, what’s missing?

Advanced content intelligence solutions for CTV already exist, and they can address many of these gaps. Today’s tools can analyze what’s happening on screen in the moment, verify that content is authentic, and assess suitability and media quality with far more precision than a single genre label. But that intelligence is only useful when it can be activated at scale. 

Data providers need access to the underlying content, a way to match their signals to the corresponding inventory, and a path to pass those signals into the bidstream before the ad opportunity reaches a buyer. Without all three, intelligence can’t be applied, no matter how advanced it is. 

In other words, the challenge isn’t simply creating better content intelligence – it’s connecting it. For contextual strategies to work across CTV, content intelligence has to be available across enough programmatic supply to be actionable at scale. In a fragmented CTV ecosystem, achieving that scale can require hundreds of individual partnerships and integrations. That’s why interoperability is essential.

What matters most to buyers, and to us as an AI advertising and orchestration platform, is that any contextual strategy for buying CTV inventory is transparent, scalable, and drives performance.”

Zach Rosenstock

Director, Inventory Development

StackAdapt

Better CTV advertising decisions start with better content signals at scale

Closing that gap requires bringing content intelligence closer to the source. Because Wurl powers FAST channels upstream, we know what content is streaming immediately before an ad break and can connect that understanding to the corresponding ad opportunity. That creates an opportunity to enrich the bidstream with more accurate, granular content signals before the impression reaches a buyer.

That thinking is behind Wurl’s Content Intelligence Platform, launched in September 2026 to help bring advanced contextual intelligence to CTV at greater scale. Through one integration, DSPs can access more than 140 billion signal-rich CTV impressions each month across the majority of ad-supported streaming inventory, including signals from leading contextual data providers as well as Wurl’s own taxonomy.

By connecting content intelligence directly to premium streaming inventory and enriching bid requests in real time, DSPs can get more from their existing contextual investments while advertisers gain a clearer understanding of the environments they’re buying. For publishers, richer signals can also provide new ways to demonstrate the value of their content and differentiate inventory beyond broad categories or audience reach alone.

Ultimately, richer intelligence has to translate into results. In Wurl and AdExchanger’s research, 64% of respondents cited proven performance as important when evaluating new strategies for their next campaign – and there is growing evidence that better content intelligence can translate into better campaign performance. Wurl’s research with TVision found that aligning ads with the emotional context of a scene drove a 2.4x increase in attention. Real-world campaigns have also shown how more granular contextual targeting can influence engagement and business outcomes across the full funnel. The implication is important: understanding what’s happening on screen isn’t just about giving buyers greater transparency – it can also help them identify the moments and environments where their advertising may be more effective.

Advanced contextual targeting already exists. What’s been missing in CTV is the footprint to make it truly impactful. By bringing richer signals to significantly more streaming inventory, we’re giving DSPs the ability to get more from the investments they’ve already made, while helping advertisers reach highly engaged audiences in the moments where their messaging can have the greatest impact.

Ria Madrid

Vice President, Global Advertising

Wurl

With richer signals available at scale, buyers can start putting that intelligence to work in more sophisticated ways. Through partners like KERV and Peer39, that can include everything from object-level recognition and precise ad-break targeting to pre-bid contextual, suitability, fraud, and media quality signals.

Ultimately, the opportunity is to make what’s actually happening on screen more visible, accurate, and actionable for everyone buying and selling CTV.

Key takeaways

• Ad-supported streaming is entering a more mature phase of growth: More households are watching streaming, while viewing time remains steady – with daily HOV up nearly 12% year over year. As CTV matures, the opportunity is increasingly shifting toward creating more value from the audiences and inventory already available.
• Basic content transparency remains inconsistent: The overwhelming majority (89%) of ad requests lack a usable program name, while just under half (45%) had no channel name. Buyers are often making buying decisions with limited visibility into the content behind the impression.
• Declared metadata can diverge from the content itself: On inventory where Wurl could compare bidstream metadata with its own contextual analysis, more than a third (35%) carried a genre that did not align with the program. Better contextual decisioning starts with signals grounded in what is actually on screen.
• Better content alignment can drive stronger outcomes: Richer content signals don’t just give buyers more transparency – they can help improve performance. Wurl research with TVision found that aligning ads with the emotional context of a scene drove a 2.4x increase in attention, while real-world campaigns have demonstrated the value of more granular contextual targeting across engagement and business outcomes.
• Content intelligence can help buyers better understand their CTV investment: CTV already has sophisticated contextual technologies like those offered through Wurl Advertising. The next step is making those capabilities available across enough inventory to allow scalable campaign activation. Connecting real-time content understanding with corresponding inventory and bidstream enrichment can give buyers a clearer picture of each opportunity – and create a more transparent, intelligent CTV marketplace.

Methodology

This report is based on Wurl’s aggregated data across major ad-supported streaming platforms, with additional modeling to highlight trends for advertisers. All findings are intended to illustrate broader patterns in streaming viewership and advertising opportunities, rather than represent census-level market measurements.

Streaming growth analysis:

• The data represents aggregated ad-supported streaming usage across leading platforms.
• All analyses in the report use U.S.-only data, with the exception of the HOV metric, which uses global viewing data to maintain consistency with prior analyses.
• Analysis and comparisons for HOV cover January through August 2025 and January through August 2026.

Bidstream metadata analysis: 

• Bidstream analysis and comparisons cover a two-hour window of a single prime-time evening in August 2026. While results may vary across different time periods, our team repeated the analysis using additional windows one week and one month later, and found that the results remained largely consistent, with only minor variations.
• Channel- and program-name missing data figures distinguish between fields containing no value and fields containing either no value or placeholder values such as “unknown.”
• Genre analysis covered approximately 1.09 billion genre-labeled ad requests. Individual requests may contain multiple genre values, so genre shares are not mutually exclusive.
• Genre accuracy was analyzed on Wurl-distributed channels where Wurl’s contextual technology could observe and independently classify the program actually airing. Declared bidstream genre values were then compared with genres detected by Wurl’s contextual model.

Make CTV your next performance
channel for growth