CTV Advertising Measurement and Attribution: Challenges and Solutions

Boasting a global audience close to two billion, CTV has surged in popularity in recent years, fundamentally altering content consumption across various markets. In the UK, for example, 32% of CTV viewers watch content via alternative devices, such as a smartphone or gaming console, with 70% of AVOD users watching ad-supported content every week. And it isn’t just audiences who are changing; public broadcasters in the market have also ventured into the connected TV space in recent months, highlighting the medium’s growing popularity.

In a gloomy economic climate, CTV has remained popular amongst advertisers and, considering the medium boasts a much higher ROI than other channels, this is hardly surprising. Similarly, reports that ads on streaming services are viewed more favourably amongst audiences than ads on linear TV can only bolster advertiser confidence, which is already proving robust: by 2024, global CTV advertising revenue is expected to hit USD$29.6bn (~£24bn) and to grow an additional USD$10bn (~£8.1bn) over the following four years. 

Nonetheless, CTV is not without its shortcomings; confusing acronyms, for example, have proven a particularly sharp thorn in advertisers’ sides. One of the channel’s most prominent flaws, however, is measurement. Measurement and attribution, in particular, prove a persistent challenge for advertisers hoping to leverage CTV to its full potential: while the medium certainly provides far richer insights into audience behaviour and preferences than linear TV, fragmentation and a lack of standardisation make it difficult for advertisers to deliver data-driven campaigns that maximise impact.

To better understand the challenges of measurement and attribution in CTV – and their potential solutions – we turned to industry experts for their thoughts:

Advertisers should invest in brand safety solutions

The main challenges of CTV measurement and attribution are limited visibility into brand safety, the need for a standardised measurement system, and the fragmented data landscape. These challenges can make it difficult for advertisers to compare the performance of their campaigns, track users across their CTV viewing journey, and ensure that their ads appear in safe and suitable environments.

A potential solution to this is investing in brand safety solutions. CTV is exciting territory but it isn’t brand suitable by default and risk content creates the potential for increased brand harm. Marketers, your brand identity is precious: protect it in real-time with a new level of data to inform CTV campaign strategy – without limiting scale. Another possible solution is collaboration between platforms and providers to develop standardised measurement solutions. This will make it easier for advertisers to compare the performance of their campaigns and attribute their results. Finally, advertisers can use first-party data to track users across their CTV viewing journey and attribute their actions to specific ads. This data can be collected from various sources, such as website visits, app usage, and loyalty programs.

Megan Reichelt, country manager, SEA at IAS

CTV must be viewed as complementary to linear TV

Measurement between walled gardens continues to be a core challenge in CTV and its incremental narrative. To combat this, CTV must be viewed as a complementary channel to linear TV by increasing unique big-screen impacts cost-efficiently. This can be achieved using automatic content recognition (ACR) data to plan linear TV reach extension by finding underexposed audiences and using third-party partners to validate incrementality.

Additionally, attribution is a function of measurement, meaning, it too faces similar fragmentation challenges. That being said, we’ve seen promising results that reinforce CTV’s place on the media plan in methods we’ve tested thus far e.g., direct response through QR codes and calls-to-action; capture pixels linking site visits to CTV ad exposure; and brand lift studies. All of which we apply A/B testing to at MiQ. The real challenge is linking it all together, something the whole industry needs to come together to solve.

Michael Lampard, product manager, Advanced TV, MiQ

Siloed platforms stymy precise data analysis

CTV measurement contends with challenges stemming from data opacity. Siloed platforms and limited visibility into key indicators of effectiveness, including contextual adjacency, hinder precise evaluation of CTV’s impacts. The lack of content signals, like genre, further complicates matters, making it difficult to accurately assess performance, draw comparisons with other mediums, and integrate CTV into omnichannel strategies.

An omnichannel media quality metric linked to full-funnel outcomes could solve some of these challenges by providing greater insight into placement-level quality, enabling meaningful cross-channel comparisons and smarter holistic investment decisions.

Marc Guldimann, founder & CEO, Adelaide Metrics

Education and understanding are critical

Sarah Lewis

A core challenge in CTV measurement has been education and understanding of the platform. CTV is ultimately TV advertising, but as it's digitally delivered, advertisers often expect the same measurement possibilities as digital advertising, which is just not the case. CTV measurement solutions have definitely progressed significantly in the past two years with brand uplift, attribution and incremental reach measurement more freely available but there is still a way to go. 

To truly progress, education is crucial so that advertisers understand the real possibilities on CTV and also the KPIs that are most meaningful. The old-world thinking of CTR KPIs just can’t be applied and even unique reach measurement should always be considered within the realm of the TV screen – a shared device with multiple viewers, unlike personal devices like mobile. If we do this, as well as standardising our KPIs and investing in CTV specific technology, then we can overcome the measurement hurdles that we’ve faced and unlock CTV’s full potential.

Sarah Lewis, global director CTV, ShowHeroes

We must avoid over-attribution

Fusing TV's potential with digital measurement capabilities, CTV is a new hot spot, where everybody wants to be. This evolution, however, has happened in a world that is fragmented, where multiple devices are used and the end of third-party cookies is on the horizon. 

CTV is a great opportunity, but it poses many challenges. We need to avoid CTV over-attribution by not assigning CTV achievements from other channels or media sources within our multichannel strategy. Transparency is mandatory when it comes to measuring the value of CTV and, right now, we're still far from it: MMM models are not as accurate as necessary and, although MMPs are working on multi-device solutions, and DSPs like TTD are announcing CTV measurement suites, these projects are in their early stages. Some companies have come up with attribution solutions that could solve this ecosystem's complexity, but will they? We must wait and see.

Agustín Jiménez SSP director at Tappx

Relying on IP-based methods can misrepresent viewership

The CTV landscape's evolution poses challenges in measurement and attribution, especially as digital advertising moves away from third-party cookies and other common stable identifiers. Relying heavily on IP-based methods can misrepresent viewership, especially in multi-member households. Regulatory frameworks like CCPA and GDPR underscore concerns with using IPs, and tools like Google's IP Protection mirror these concerns.

Furthermore, with so many different measurement approaches in the market, DSPs and SSPs often lean on third-party tools for insights. One of the most significant hurdles is achieving holistic cross-media measurement. As consumers diversify their viewing across devices, ensuring a unified brand message becomes increasingly complex. Adapting our strategies to this dynamic CTV environment is crucial. 

Mateusz Jędrocha, VP branding solutions, Adlook

Centralised buying is key

It is important that brands don’t fall into the trap of “CTV is digital, therefore it must be fully measurable from end to end”. Due to the nature of the product, its immediate measurability and attribution of performance is limited. 

The solution to this is to centralise the buying approach in a uniform manner. Custom algorithms provide brands the opportunity to do this, by building a plan which incorporates data from the real world and imbuing this into a custom algorithm to power campaigns, rather than just using what exists within a buying platform. By combining data with custom algorithms, campaigns can be supercharged with increased intelligence and deliver a consistent approach to the desired audience. Using this technology in conjunction with a control and exposed testing methodology, brands can establish the true value of their CTV activity and how to further optimise it for increased effectiveness.

David Newman, managing director, 59A