The advertising industry continues to buy connected TV the wrong way. We have carried over one-to-one programmatic logic from display and online video into a channel that behaves nothing like them. CTV is not a personal screen; it is a household screen.
Treating it like a digital channel has led to inefficiency, waste and false precision. A logged-in profile on a streaming app does not tell you who is watching. In a shared environment like a living room, targeting a female 25-54 on Disney+ could just as easily serve an impression to her child. Yet many buyers keep applying audience data as if CTV were a mobile app.
This is one of the reasons CTV is considered expensive. Third-party segments shrink available supply and force buyers to compete for the same small pool of impressions. CPMs rise. Scale disappears. And the data being applied often lacks relevance because it does not reflect the shared nature of household viewing.
Another issue that gets little attention: Geo delivery patterns always look the same. Activate a third-party segment, and you will see the same markets dominate delivery: New York, Los Angeles, Chicago, Miami, Houston. The system defaults to where scale is easiest to find, not where the audience that matters actually lives.
There is a better way.
From false precision to real performance
We need to stop chasing impressions and start chasing impact. Start with your customer file. Identify the ZIP codes where your customers overindex. Use census data to understand the socio-demographic makeup of those areas. Unlike third-party audience segments, which are black boxes built from unverifiable sources, ZIP code data and census information are self-declared and real.
These data points offer a clear, proven proxy for finding and scaling into the right households. From there, you can identify lookalike markets that share those same characteristics and represent meaningful growth opportunities.
There is a related problem tied to this legacy mindset. Because we are still applying old programmatic principles, buyers are conditioned to believe they need massive scale from a wide diversity of publishers. That thinking leads to reliance on inventory aggregators, ad tech resellers and unnecessary middlemen. In reality, the vast majority of CTV viewership sits with no more than 20 premium publishers, OEMs and vMVPDs.
The most efficient and effective path is to form direct relationships with these partners. Cut out the ad tech middle layer. Doing so drives down CPMs, reduces fraud and inventory misdeclaration risk and builds long-term leverage.
As spend grows with preferred partners, buyers benefit from stronger pricing and better access. There is far more value in building a direct relationship with a trusted legacy company like Disney, NBCU or Paramount than in transacting through a five-year-old venture-funded reseller that may not exist five years from now.
CTV should be planned like linear: context-driven, geography-informed and designed for households, not individual IDs.
The tools are there. What it takes now is the courage to break from outdated norms and the willingness to think differently about what CTV can and should be.
“On TV & Video” is a column exploring opportunities and challenges in advanced TV and video.
Follow InterMedia Advertising and AdExchanger on LinkedIn.
For more articles featuring David Nyurenberg, click here.
