CTV Can’t Out-Google Google, So It Should Play A Different Game
Connected television is projected to grow 14% in 2026, faster than the overall advertising market, according to IAB. But continued growth depends on proving that CTV delivers business results, not merely audiences and impressions.
CTV can solve two industry problems at once. First, it can prove the outcomes-based value of premium brand advertising, including its incremental effect on sales. Second, it can advance advertising measurement by making randomized experiments, the strongest evidence for causality, a routine campaign capability.
Search and social won with attribution. Retail media owns closed-loop marketing. CTV can claim the open ground: proving that premium brand advertising drives sales through credible causal evidence.
CTV is uniquely suited to do it. It combines the high-impact, high-attention, sight-sound-and-motion experience of television with digital delivery, addressability and the ability to vary exposure across households and geographies.
Brand advertising faces a credibility test
Brand advertising has always been tough to measure. Its effects accumulate over time and influence customers who never click. The resulting sales may occur through channels far removed from the original exposure.
Channels that produce immediate conversion reports give the appearance of more accountability. Premium video is then forced to compete for ad dollars using a performance-measurement framework designed for search and direct response.
Attaching more conversions to impressions does not solve the problem. Attribution can show that an exposure preceded a purchase, but not that it caused the purchase. Many people credited with a conversion would have bought anyway. This is correlation masquerading as causation.
The question that matters is incrementality: What happened specifically because of the advertising that otherwise would not have happened?
A different game from attribution
Google’s measurement advantage is inseparable from its role in capturing intent. Social platforms connect exposure, engagement and conversion within their own environments. Retail media networks link advertising to transactions within retailer systems.
CTV sits elsewhere in the customer journey. Viewers are in lean-back mode, engaged in moving images conveying stories or live events. Premium video ads build memory, preference and demand. Judging it primarily by clicks or attributed conversions understates how advertising in the medium works.
CTV cannot out-Google Google with more on-platform attribution. It should play a different game: causal credibility. No major channel owns the proposition of randomized experimentation for causal measurement. CTV can own the space of premium brand advertising routinely tested against total business outcomes using transparent randomized designs.
Why experimentation fits CTV
Randomized controlled trials create the most credible counterfactual. Some households or geographic areas receive advertising; others do not. Random assignment minimizes systematic differences between the groups, allowing differences in outcomes to be attributed to advertising with measurable statistical and scientific confidence.
CTV can support experiments at several levels. Addressable households can be randomly assigned to treatment and control groups. Geographic experiments can randomize delivery across designated market areas. As addressable delivery becomes more granular, ZIP codes can provide many more experimental units and greater power to detect modest effects.
Geo experiments are especially important because they can measure advertising’s effect on total company revenue. They do not require every exposure to be matched to an individual purchase. Advertisers can vary media across randomly assigned markets and observe changes in sales, store visits, leads, subscriptions or other outcomes. Advertisers can measure lift directly from their own CRM or transaction systems, or through sales-panel partners, using only ZIP code and transaction date. Incremental sales can then be reconciled with the lift finance teams see in the P&L, rather than confined to conversions visible within a platform’s targeted audience.
The United States has 210 DMAs, enough to support large-scale national cluster randomized trials in many circumstances. ZIP-level delivery could expand the available test units and improve statistical power. Together, households, DMAs and ZIP codes give CTV an unusual experimental range.
A position worth owning
Following the path of digital attribution would put CTV in a contest against platforms with deeper behavioral data and stronger closed systems. It would also reduce premium television advertising to the matches and conversions that happen to be observable.
By making randomized experimentation routine, CTV can strengthen the case for brand advertising precisely when that case is under pressure. It can show that sight, sound and motion do more than generate impressions or correlate with purchases. They cause measurable changes in the business. They can also create transparent standards for experimental design and ensure sellers are not the sole judges of evidence used to validate their own inventory.
Search and social can own attribution. Retail media can own the closed loop. CTV can own the credible connection between brand advertising and outcomes. No other medium is as naturally positioned to make that connection its defining measurement advantage.
“On TV & Video” is a column exploring opportunities and challenges in advanced TV and video.
Follow Central Control and AdExchanger on LinkedIn.
|
