CTV Measurement: Key Challenges and Solutions for Marketers

The Importance of CTV Measurement for Marketers

In 2024, there’s a lot of excitement among marketers and brands around the promise of Connected TV (CTV) —  a medium that combines the creative capabilities of terrestrial broadcast with the addressable analytics of digital advertising. According to one recent study, 95% of advertisers planned to maintain or increase their share of spend on programmatic CTV in 2024. Compared to traditional TV, CTV is an order of magnitude cheaper, while allowing brands to laser focus their ad targeting on very specific audiences. As one insider recently opined, CTV’s “unique ability to measure audiences’ engagement and actions in response to TV advertising has allowed marketers to align the soft goals of brand-building with the hard goals of driving direct sales outcomes. In doing so, CTV bridges the gap between branding and performance.”

We believe CTV’s targeting capabilities make it a powerful awareness medium, but it needs to be said: When it comes to measurement and attribution, CTV is still in its relative infancy. The primary question we get from clients is: How do we know if our CTV ads are working? 

The answer changes depending on which CTV platform you’re using—and the various methods and standards can be difficult to follow, even for seasoned media buyers. With that in mind, we set out to create a transparent guide to CTV attribution methodologies: to explain how each of the major platforms measures performance, how they differ, and what the limitations are of each approach.  

Challenges In CTV Attribution and How To Overcome Them

Unlocking the true potential of CTV advertising hinges on one crucial factor: post-viewing measurement. With CTV’s sophisticated audience targeting, you’ll know who your ads reached. But how do you measure what actions the audience takes after viewing your ad? 

Attributing the impact of a CTV ad can be difficult, confusing, misleading, and frustrating. Unlike a straightforward website click, tracking a viewer's journey from ad exposure to conversion involves navigating a complex web of factors, including cross-device tracking, changing privacy regulations, different attribution models and windows per platform, audience overlap, user consent, and opt-in rates. Simply seeing an ad doesn't necessarily mean it directly influenced a purchase. This is where the challenge of CTV attribution arises. Advertisers need a reliable way to measure the true impact of their ads, filtering out extraneous influences and crediting only the ads that demonstrably drive results.

While the goal is clear, achieving it is no easy feat. The current CTV landscape is a patchwork of platforms, each with its own unique approach to measurement. Some platforms prioritize basic metrics like impressions and completion rates, while others leverage sophisticated attribution models. According to Digiday, increased privacy initiatives—such as the decline of Google’s third-party cookie—are leading to “signal loss,” which leaves “advertisers and their agency partners dependent on multiple sources of truth.” The result? “Market mix modeling as well as third-party vendors and platforms are often needed to fill in the gaps” on attribution. Understanding these diverse methodologies is crucial for advertisers, so that they can optimize their CTV campaigns for maximum effectiveness.

The crux of the problem is connecting a CTV viewer to actions they later make on another device, such as their phone or laptop. Most platforms augment their own data with signals from third-party companies including Innovid, Lucid, EDO, Samba TV, and Experian—each of which, in turn, has its own complex methodologies. (See our CTV Measurement Partner Guide, below.) 

For instance, Innovid uses anonymized user identifiers, such as IP addresses, device IDs, or cookies, to match users across different devices while maintaining privacy and compliance with data protection regulations. Once users interact with a CTV ad and later visit the advertiser’s website, Innovid tracks specific conversion events on the website. These events could include actions like making a purchase, signing up for a newsletter, or filling out a form.

One big challenge is that for the most in-depth analytics, most platforms require a high minimum spend or impression floor — which can prevent advertisers with smaller budgets from being able to test CTV placements. Making these studies more accessible would make this channel more approachable and easier to pitch to clients.

Below, we’ve broken down the methodologies of five major CTV platforms and how they measure outcomes with the audience that’s viewed a CTV ad.

How Is CTV Attribution Measured?

Hulu

For years, Disney has been consolidating audience data across its many platforms, seeking to unify its intelligence on the 218 million unique monthly visitors to its media sites, 100 million household-level IDs, and 160 million connected TV IDs. In addition to its first-party data, Hulu/Disney’s CTV platform partners with third parties such as LiveRamp, Experian and Samba TV to ensure reach among high value audiences and accuracy in cross-platform measurement. 

Hulu also analyzes dozens of campaigns to build industry-specific benchmarks for reach, frequency, and impressions—so if you are able to measure lift and outcomes, you’ll have a guide against which to measure your progress.

To measure what happens after audiences see an ad, Hulu leverages its relationships with external partners to provide full-funnel measurement solutions. In line with most platforms, access to these studies is generally predicated on hitting campaign minimum buys, and for accurate results, there is a recommended impression minimum of 15-20M. Here’s what Hulu is able to measure:

• Brand Lift. For brands interested in direct campaign impact and quantifying how CTV ads influenced consumer behavior, Hulu uses Lucid (see below) to measure brand and ad awareness, as well as purchase intent.
• Tune-In. Hulu Attribution, Hulu’s baked-in analytics platform, uses Automatic Content Recognition (ACR) — a technology that identifies and captures content being played on a TV screen—or in-house viewership data to identify content being consumed.
• Search. Hulu uses Entertainment Data Oracle (EDO), a third-party company that uses advanced data analytics and proprietary methodologies to track and measure the impact of TV ads on search behavior, helping brands understand how their ads drive consumer interest and  engagement.  
• Online Web Conversion. Hulu uses a variety of market-leading platforms — including Innovid, Samba TV, and Experian—to measure conversions via sign-up, purchases, and app downloads. These platforms all do the background magic of connecting ad exposure to viewers who took a specific action (such as visiting a website or downloading an app).  

An advantage Hulu offers over its competitors is that Innovid’s web-conversion studies are now available through the self-service Disney Campaign Manager (formerly known as Hulu Ads Manager). Having the study available via self-service makes reporting easier and less time-consuming for the advertiser. Innovid can provide KPIs including cost per response, total web response, and a response rate index. The study requires a minimum $40,000 campaign buy, an eight-week flight minimum, and at least 50,000 monthly website visitors. Plan ahead: a 2-3 week lead time is required to set up the study.  

Amazon Prime Video

Since the launch of Prime Video’s advertising tier in early 2024, the volume of Prime Video users has surpassed estimates, exceeding 200 million monthly viewers — placing it second highest among streaming services by subscriber base. (That puts it behind Netflix, with 247 million users, and ahead of Disney+, with 150 million users.) Prime Video taps into the Amazon Ads ecosystem to offer access to first-party solutions like Amazon Brand Lift, with additional support from third-party providers including Nielsen, Lucid, Innovid, and Kantar to help brands contextualize reach and frequency, quantify brand lift, and measure in-app conversions based on campaign objectives.

While we can measure Prime Video ads in terms of completed views and CPM, the real differentiator comes when we examine the impact on post-view actions. In our tests, Prime Video ads amplified the success of the brand’s Amazon Sponsored ads by 5x. We also found that after seeing Prime Video ads, viewers were more aware of and favorable of the brand, and expressed more intent to purchase. And Prime Video ads delivered incremental sales of the brand’s products on Amazon’s sales platform. In our tests, more than half of sales attributed to Prime Video ads came from new-to-brand users, showing that Prime Video wasn’t just targeting existing brand shoppers. 

Prime Video’s main differentiator — access to Amazon’s streaming and shopping data — is precisely what makes it difficult to compare it with competitors like Hulu. While Hulu has broader first-party targeting capabilities, Prime Video offers a vast array of targeting via Amazon’s user base across In-Market, Lifestyle, and Prime Video Retargeting segments. 

It’s important to underscore that when we looked only at basic CTV metrics like spend, impressions, video completions, and CPM, the results we observed between Hulu and Prime Video were fairly even. But if you’re only looking at those metrics you’ll miss the differentiators that make CTV such a potential game-changer. For instance, when we dove deeper to look at how likely viewers in a target audience were to visit the brand’s site within seven days of viewing an ad, the results between Hulu and Prime Video were very different — with significant advantages in Prime Video’s favor.

Roku

The advantage of Roku lies in the nature of its platform: an ecosystem of streaming-player devices and smart TVs with greater visibility into audience and viewership. Roku’s secret sauce includes RIDA, the Roku ID for Advertising, “a device identifier available to Roku publishers for development and marketing purposes,” which “enables ad targeting and frequency capping on the Roku platform.” Augmented with inputs from third-party sources such as TapAd, LiveRamp, and Acxiom, as well as Roku’s own OneView BidStream platform, Roku’s data set includes hundreds of attributes—from browsing history to app installs and TV exposure. And Roku boasts that RIDA provides an advantage over other platforms that rely solely on IP addresses, since an estimated 30% of IP addresses churn each month. The bottom line: Roku’s selling point is that they’ve “built our measurement framework on a foundation of verified identity, consisting of direct relationships with streamers.” 

• Roku’s RIDA stays persistent and is linked in a privacy-compliant manner to registration data. Currently, Roku claims 80 million active accounts.
• RIDA provides IP address and device ID for advertising targeting, frequency capping, and campaign measurements.
• Automated Content Recognition (ACR) is built directly into the devices built by Roku TV Manufacturing partners. For opted-in viewers, Roku can detect what the audience is watching in real-time across devices and platforms through access to a rich database of descriptive metadata. The quality of source data that’s powering ACR affects every aspect of its output: targeting, attribution, research and analytics. Using data that is representative of the in-market population is the key to gaining insights and reaching a relevant target audience based on TV viewership. 

Additionally, to measure reach measurement, brand resonance, and sales attribution, Roku works with a suite of vetted third-party vendors including Nielsen, VideoAmp, and Kroger Precision Marketing.

NBCU

NBC Universal uses Innovid to track web traffic. With Innovid, a pixel is attached into the creative; the client will then be sent a tag to add to all pages on their website in order to track the response data through Innovid. The tag on the client’s website will capture all website traffic—regardless of whether the visitor was served the client’s ad—and then matches it against the IP addresses of anyone who saw the spot and went to the website within a 7 day attribution window. If the client has a mobile app, they may also be able to track app activity through their mobile measurement partner. Typically, in order to access measurement studies, NBCU requires a minimum buy of $30,000 for a single market or $50,000 for multi-market campaigns, with minimum impressions of 1 million for a single market or 3 million for multi-market, with a flight of 1 to 3 months. 

Netflix

Netflix’s advertising-supported tier launched in November, 2022, and is now estimated to include 40 million global monthly active users. But its ad capabilities lag well behind the rest of the field, with AdWeek noting that Netflix “still lacks technical capabilities that are table stakes for most of its competitors, like measurement, advanced targeting, and the ability to buy inventory in biddable programmatic contexts.” One agency source likened buying ads on Netflix in 2024 to “buying ads on a website circa 2005.” 

In late 2023, Netflix noted that it was “working diligently to bring more measurement capabilities to our advertisers globally, as we look to partner internationally with third-party providers to enable campaign verification in 2024.” In a Q2 2024 earnings call, Netflix co-CEO Greg Peters acknowledged that “the biggest negative feedback we get is that we aren’t there right now,” while MarketingDive described the platform’s progress on the scale of “crawl, walk, run” as “inching nearer to the ‘walk’ portion of that road map.”

That said, Netflix has announced plans to launch its own in-house ad serving technology by 2025—and has announced deals with demand-side platforms and supply-side platforms including the Trade Desk, Google, and Magnite. Currently, Netflix uses Microsoft Advertising’s owned-and-managed platform as its exclusive technology and sales partner. The platform offers post-view attribution through Microsoft 1P solutions, by mapping Netflix profile IDs to Microsoft unique IDs to determine if a user associated with a pixel fire can be connected to an ad exposure. The Netflix team notes that while its ads are not clickable on any device, “QR code-enabled creatives are a new alpha feature of [Netflix], and we plan to support ‘post-click/post-scan’ attribution in the future." While third-party measurement is not currently supported on Netflix, they do use third-party solutions for non-CTV measurement.

The Future of CTV Measurement

The allure of Connected TV is undeniable. Its precision targeting and potential for measurable impact have captivated advertisers. Yet, beneath the surface of this promising medium lies a complex web of measurement challenges. While platforms like Hulu, Amazon Prime, Roku, and NBCU offer varying degrees of insight, a standardized, transparent, and accessible approach to CTV attribution remains elusive. As the industry grapples with privacy regulations and the evolving digital landscape, the need for robust, independent measurement solutions becomes increasingly critical. Until then, advertisers must navigate a labyrinth of metrics and methodologies, relying on a combination of platform data, third-party verification, and a healthy dose of skepticism. While the path to definitive proof of return on investment is still under construction, CTV remains a compelling opportunity for brands willing to navigate its complexities with the guidance of experienced partners.

Appendix: CTV Measurement Third Party Partners Guide

Innovid: A leader in CTV measurement, providing advertisers with a comprehensive view of their campaign performance across the complex CTV landscape. The platform empowers advertisers to make data-driven decisions, optimize their CTV campaigns, and measure the true impact of their advertising efforts.

• Comprehensive Data Collection: Innovid gathers data from various sources including ACR data, smart TVs, and set-top boxes, providing a holistic picture of viewership and ad exposure. 
• Advanced Analytics: By analyzing this data, Innovid delivers in-depth insights into ad performance, audience behavior, and campaign effectiveness.
• Cross-Platform Measurement: Innovid's measurement capabilities extend beyond CTV, enabling advertisers to compare and contrast performance across linear TV and digital video.
• Real-Time Optimization: Innovid's platform allows for real-time adjustments to campaigns based on performance metrics, ensuring maximum ROI. 

Samba TV: Renowned for its advanced CTV measurement capabilities, Samba TV leverages a unique approach centered around Automatic Content Recognition (ACR) technology. By combining ACR with a large-scale panel and advanced analytics, Samba TV offers advertisers a powerful tool to measure the impact of their CTV campaigns and optimize their media investments.

• ACR-Powered Measurement: Unlike traditional panel-based systems, Samba TV relies on ACR data to measure viewership. This method offers several advantages, including the ability to capture all content, including streaming services, without relying on set-top box data. 
• Large-Scale Data: Samba TV boasts a massive panel of smart TVs, providing a representative sample of the US population. This scale enables accurate measurement and insights.    
• Cross-Platform Measurement: Samba TV offers a comprehensive view of audience behavior across linear TV and CTV, allowing for holistic campaign evaluation.    
• Advanced Analytics: Beyond basic metrics, Samba TV provides in-depth analytics, including audience insights, ad effectiveness, and campaign optimization recommendations.
• Identity Graph: Samba TV's identity graph connects devices and households, enabling deduplicated reach and frequency measurement across screens.    

Experian: Experian's role in CTV measurement is to support other measurement platforms by providing essential data and services that improve the accuracy and effectiveness of campaigns.

• Data Enrichment: Experian can enrich first-party data with demographic, lifestyle, and financial information, providing a deeper understanding of the CTV audience.
• Fraud Prevention: Experian's fraud prevention solutions can help advertisers protect their CTV campaigns from ad fraud and invalid traffic.
• Identity Verification: By verifying consumer identities, Experian can help improve the accuracy of audience targeting and measurement.
• Clean Data: Experian's data cleansing capabilities can help ensure the quality of data used in CTV measurement, reducing errors and improving campaign performance.

Lucid Impact Measurement: a platform designed to measure the impact of advertising on brand metrics. While Lucid is effective at measuring brand lift, it doesn't provide the same level of detail on audience measurement, ad exposure, or reach as platforms like Samba TV or Innovid.

• Integration with Ad Servers: Lucid integrates with CTV ad servers to track ad exposure.
• Survey Distribution: After ad exposure, a targeted subset of viewers is invited to participate in a survey.
• Brand Lift Measurement: The survey measures changes in brand awareness, purchase intent, ad recall, and other relevant metrics.
• Data Analysis: Lucid's platform analyzes the survey data to determine the lift generated by the CTV campaign.

LiveRamp: Plays a crucial role in CTV measurement as a data connectivity platform, connecting the dots between data, identity, and measurement. It doesn't directly measure ad performance or audience behavior, but it provides the foundation for other tools and platforms to do so effectively.

• Data Onboarding: LiveRamp helps advertisers and media owners onboard their first-party data, such as CRM data or website traffic, into a unified, privacy-safe format.    
• Identity Resolution: It matches and connects data across different platforms and devices, creating a comprehensive view of the audience.    
• Data Activation: LiveRamp enables the activation of targeted CTV campaigns by connecting audience segments to media buying platforms.
• Measurement Integration: By providing a consistent identity across platforms, LiveRamp facilitates the integration of measurement data from different sources. This allows for a more accurate and holistic view of campaign performance.  

Kantar: A global leader in data, insights, and consulting, it combines traditional research methodologies with advanced data analytics to deliver comprehensive insights into the CTV landscape. 

• Audience Measurement: Kantar provides insights into the CTV audience, including demographics, viewing habits, and consumption patterns. This data helps advertisers understand their target audience better.
• Media Planning and Optimization: By analyzing viewership data, Kantar assists media planners in optimizing CTV ad placements and achieving the desired reach and frequency.
• Ad Effectiveness: Kantar offers tools and methodologies to measure the effectiveness of CTV campaigns, including brand lift, ad recall, and purchase intent.
• Cross-Platform Measurement: Kantar provides a holistic view of media consumption across TV, digital, and social platforms, enabling advertisers to assess the overall impact of their campaigns.

iSpot.tv: a leading player in the CTV measurement landscape, offering a unified measurement platform that addresses the challenges of measuring ad performance across both linear TV and streaming platforms. However, it may not offer the same level of granularity in audience measurement as a platform focused solely on CTV, like Samba TV. And while it offers a strong foundation for campaign measurement, it might not be the best choice for in-depth brand lift analysis like Lucid Impact Measurement.

• Unified Measurement: iSpot.tv provides a single platform to understand campaign performance across linear TV and CTV. This eliminates the need for siloed reporting and allows for a holistic view of campaign reach and effectiveness.    
• Impression Measurement: iSpot utilizes various methods, including proprietary ad catalog data, airing data, and a smart TV panel, to capture ad impressions accurately across screens.  
• Demographic Targeting: Impressions are then mapped to the precise demographic makeup of the exposed audience, providing valuable insights into who is seeing the ads.    
• Attribution Modeling: iSpot goes beyond basic measurement by offering attribution modeling tools to understand the impact of CTV ads on conversions and other campaign goals.
• Real-Time Reporting: A comprehensive dashboard allows advertisers to monitor campaign performance in real-time and make adjustments as needed.   

Nielsen One: From the company that set the industry standard in TV measurement comes a unified approach to understanding audience behavior across linear TV, streaming, and digital platforms. For CTV specifically, Nielsen One provides a robust suite of measurement solutions. However, due to its comprehensive nature, Nielsen One can be complex to navigate and understand. And the cost can be prohibitive: accessing Nielsen One data often involves significant investment.

• Audience Measurement: Nielsen One provides detailed audience demographics and viewing habits for CTV content, enabling advertisers to understand their target audience.
• Cross-Platform Measurement: It offers a unified view of audience behavior across linear TV and CTV, allowing for a holistic understanding of media consumption.
• Ad Exposure Measurement: Nielsen One measures ad exposure on CTV platforms, providing insights into ad reach and frequency.
• Attribution Modeling: By linking ad exposure to consumer behavior, Nielsen One helps advertisers understand the impact of CTV ads on sales and brand metrics.

VideoAmp: a strong platform for advertisers seeking to optimize their CTV campaigns based on data-driven insights. Its focus on cross-platform measurement and ad optimization makes it a valuable tool for maximizing ROI.

• Audience Segmentation: VideoAmp utilizes advanced data analytics to segment audiences based on various factors such as demographics, viewing behavior, and purchase intent.
• Cross-Platform Measurement: The platform enables advertisers to measure campaign performance across linear TV and CTV, providing a unified view of audience reach and impact.
• Ad Optimization: VideoAmp offers tools to optimize ad delivery in real-time based on performance metrics,ensuring maximum ROI.
• Attribution Modeling: By analyzing data from various sources, VideoAmp helps advertisers understand the impact of CTV ads on consumer behavior and sales.