Some 60% of U.S. households now own a Connected TV (CTV) according to eMarketer. CTV consists of Smart TVs, connected devices like Roku and Apple TV, as well as gaming consoles like Xbox and PlayStation. CTV usage has grown over 300% in the past year and now accounts for more prime-time streaming of TV and movies than DVRs, and twice as much as desktop and mobile consumption combined. However, anyone who has watched ad-supported CTV apps from broadcast and cable networks or “premium” publishers has seen the same poorly targeted commercials over and over again. This happens because even though 160 million Americans are streaming content on CTV each month, CTV accounts for barely 1% of TV ad dollars. Why aren’t brands moving to CTV in droves when it’s no secret where viewers have shifted their time and attention? There are three roadblocks preventing the inevitable shift of tens of billions of linear TV and desktop ad dollars to CTV, all of which ultimately tie back to gaps in measurement: 1. TV and Web measurement incumbents have failed to bring forward a viable OTT measurement “currency” to validate audience size and makeup. 2. Legacy segmentation and targeting solutions are paralyzed by their dependence on cookies and device id’s. There are no cookies or device IDs on CTV. 3. There is no marketing attribution (i.e., brands and agencies have no way to analyze the efficacy). Why is there is no measurement currency for CTV when creating one is mission-critical to the future of TV? So much of the value of CTV lies in the ability to engage viewers at a highly granular level with the precision and immediacy characteristic of digital marketing. It follows that a true CTV measurement currency – one able to unlock the full value of the medium – must provide near-census level measurement and near-instantaneous reporting. Real-time, census-based CTV audience tracking presents major challenges to a measurement industry establishment that has long relied on sample-based approaches. Proposed solutions often borrow from legacy tools and techniques, resulting in complex, sub-optimal offerings. An example is deploying SDK’s (software development kits). However, these require frequent updates, and must be customized to each OTT network/publisher for each of the myriad devices that stream CTV content. All that data must then be reassembled on the back end to tally viewers for each network. Census-level measurement is only half the battle. In a world that’s more about buying the audience than buying the content (exposures are guaranteed on CTV), advertisers need to know the composition of an audience to buy into a program. Demographics are mere table stakes for digital media like CTV – purchase, preference, and other consumer characteristics are often required to guide digital buys. The current way to gather such information in the digital world is by dropping cookies on a PC or tracking mobile device IDs. These techniques are non-starters for CTV, as the entire CTV ecosystem is cookie-less, and CTV devices do not carry IDs. Simply put, audience segments based on cookies or device IDs cannot be bought on CTV. Attribution presents another stumbling block for traditional audience measurement. As with all digital media, “attribution” is much more than a report card delivered after the fact. It’s about a continuous feed of KPI’s used to monitor campaign performance and initiate course corrections in real time. That means having the ability to link CTV ad exposure to activities like browsing or purchasing that occur on other devices or outside the digital realm entirely. That’s impossible if your CTV measurement solution exists in a vacuum. A panel-based approach could connect CTV advertising to other influences and behaviors, but relying on post-event sample-based reporting is no way to go through life in a digital world. What’s the answer? A true measurement currency for CTV will have to be real-time and census-based, with rich profiling information and linkages to desired outcomes like shopping and buying. Easy enough, right? In truth, that depends on where you’re coming from. Adapting legacy systems to the task will be a tractor pull – slow and arduous. Fresh approaches are needed if Connected TV is to have the measurement system needed to fully realize its potential, and viewers spared the mind-numbing effects of commercial overexposure.