Understanding CTV data and measurement

Aaron Kaplan

Aaron Kaplan

Associate Director, Planning

1st September 2022

~ 5 min read

Last month, we kicked off our connected TV (CTV) blog series by diving into how CTV has developed into the expansive and continuously growing paid media channel that it is now. 

In our previous blog, we’ve spoken to one core part of the upcoming evolution of CTV - the shift of walled-garden video on-demand (VOD) services, like Netflix and Disney+, towards an ad-supported ecosystem.

The challenges of CTV data and measurement

However, the essential question marketers should be asking themselves will remain the same - how do we measure and define value from this media activity? Data and measurement in the CTV space will be the key to success, as these new entrants to the field are shaking up the traditional, somewhat out-dated, field of viewership measurement that is deeply rooted in the world of linear TV. Some of the main issues marketers have had with CTV measurement has been the inconsistent, fragmented reporting across various CTV media, as well as the general lack of congruence with additional channels, such as linear TV and other walled-garden environments.

Another common gripe with CTV is concern over frequency. However, research conducted by Innovid and ANA has demonstrated that this frequency issue is highly exaggerated, with their study revealing an average frequency of 4.6 across CTV campaigns. Generally, frequency caps can be instated to fully avoid any issue.

How CTV is measured

Now let’s take a step back and examine how CTV is typically measured, as well as what standard pieces of measurement are offered across the board of partner direct (Hulu, Peacock, WarnerMedia, etc.) and programmatic CTV buying (Innovid, VideoAmp, SambaTV, SamsungAds etc.). 

The standard metrics that any CTV video partner will offer generally span across reporting metrics and viewing dimensions. 

 Reporting metrics 

  • Impressions
  • CPM
  • Video completions
  • Frequency
  • Total reach 
  • Unique (household de-duplicated) reach

Viewing dimensions

  • Platform
  • Program
  • Viewing device
  • Time of viewing
  • Type of viewer (ex. light, medium, or heavy viewer based on daily viewing habits)

Data on specific program and genre viewing allows for a unique view of the specific interests and habits of an individual, providing a new route for optimization and audience insight/targeting based on these consumption habits.

Within CTV campaign flights, advertisers are able to optimize across inventory, audiences, placements, and creative. While dependent on the campaign objective, advertisers will typically optimize towards and set CTV KPIs based on:

  • ideo completion rate (VCR): Optimization towards where the viewer is exposed to the full video message, indicating engagement.
  • Cost per view (CPV) or Cost per completed view (CPCV): Optimization towards the most cost efficient inventory.
  • Cost per thousand reached (CPMR): Passive optimization towards the cost per one thousand households reached; an efficiency measure for awareness based campaigns.
  • If a brand allows tracking pixels to be implemented on their website, visits and actions taken on site can also be measured to optimize towards cost per landing page visit (CPLV) or cost per action (CPA).

Access to CTV data

Now taking a step forward, what are the types of unique data that CTV programmatic vendors have access to?

When it comes to programmatic CTV vendors (aggregators/resellers of media owners’ CTV inventory), measurement technology is shaping the space through first-party data or third-party data licensing. The two main types of data sources can be classified under:

  • Automatic content recognition (ACR): The recording of viewership on SmartTVs through tech built into SmartTVs that matches either audio- or video-based patterns to a content reference library
    • Examples) Samsung, Vizio, LG
  • Set-top box: The passive collection of viewership data, leveraging set-top boxes (cable boxes) that are connected to TVs in a home
    • Examples) Comcast, Dish, Frontier 

Some CTV suppliers have access to particular combinations of set-top box and SmartTV brand data. Vizio alone has more than 18 million ACR-enabled smart TVs in the market. With these large data sets of households, these suppliers are then able to model the full US TV viewing population.

This TV viewership data fuels unique approaches to CTV advertising. Through a combination of sources, CTV programmatic vendors are able to identify what, where, how, and how long unique households have been exposed to particular TV content, including specific advertising spots.

Leveraging CTV data for unique audience targeting

Beyond your standard programmatic types of targeting, such as demographic, content, third-party segment based, etc. (that are all valuable in their own right), leveraging this type of viewership data opens up a world of exciting audience targeting opportunities and combinations. Common targeting routes include:

  • Reach extension: Extend off of an advertiser’s Linear TV buying by removing those exposed to your spot on LinearTV from your audience targeting pool. This will provide incremental household reach across your target audience, reaching any hard-to-reach cord-cutters and more. 
  • Competitive conquesting: Target those who have been exposed to a competitor's spot in order to counter their message with your own, fight for share of voice (SOV), and piggyback off of their buying approach.
  • Content targeting/retargeting: Capturing an audience that tunes into an event on TV so you can activate against that same audience on digital. 

Cross-channel measurement

A handful of CTV tech suppliers (i.e. VideoAmp and SambaTV) have expanded to also measure particular digital cross-channel activations through the tagging and tracking of other partner and channel media activations. For example, let’s say you have a video campaign running programmatic CTV as well as video across Hulu, Facebook, and Conde Nast publications - all of these have the potential to be measured under the same hood. You will therefore be able to report on cross-channel reach and frequency (as well as outcome measurements), allowing you to examine specific duplication or overlap across your media partners, and more.

This enables both valuable reporting on an expanded picture of success across media campaigns as well as audience optimization opportunities. As this capability expands and becomes more widely adopted across the industry, the standards and potential for CTV will further expand, especially as advertisers’ fears of siloed or inconsistent measurement across buying is assuaged.

In the next blog of this series, we will cover a raging battle between the traditional world of TV measurement and new digital measurement disruptors, which may change the fabric of how TV buying is transacted. If you’d like to learn more about CTV advertising or would like to explore ways to optimise your CTV data and measurement strategy, please get in touch with our team.

Sign up to Croud’s Digital Digest

Get Croud's monthly newsletter, which is packed with the latest news from Croud from across the globe, along with updates and commentary around the latest developments in the digital marketing space.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.