Many brands buy Connected TV like digital advertising when it should be bought and measured like TV.
Common CTV myths around cost, frequency, targeting, and measurement are quietly wasting advertiser budgets
The brands winning in CTV use AI-powered buying, layered measurement, and run linear and CTV together for maximum impact.
Nearly every US household has a smart TV today. Americans spend 25% of their waking hours in front of those screens. And Connected TV ad spend keeps climbing.
So why are so many brands buying CTV the same way they buy digital ads?
At Shoptalk Spring 2026, Marketing Architects Chief Media Officer Catherine Walstad broke down the five biggest myths holding CTV advertisers back, and what smarter brands are doing instead.
Sure, Connected TV has a lower barrier to entry. You don't need hundreds of thousands of dollars to launch a campaign. But accessibility and efficiency are not the same thing.
When linear TV is bought well, CPMs can be significantly lower than CTV. Add in the tech and targeting fees layered on top of CTV buys through a traditional DSP, and the cost gap widens fast. Five hundred dollars can technically get a CTV campaign live. But for most brands, it won't move the needle.
The awareness, credibility and reach that make TV worth doing require real investment. Don’t confuse "easy to start" with "cheap to scale.”
This one makes sense if you're coming from performance marketing. Repetition works in search. In email. In retargeting. On TV, it turns into the ad that makes people dive for the remote.
Marketing Architects data has found that the first TV impression does the most work. After that, response declines. Multiple impressions can still drive positive ROI and support brand recall. But excessive frequency wastes budget and can even irritate the people you're trying to reach. We all know how it feels to see the same ad over, and over, and over.
Frequency in CTV is a documented problem. It needs to be actively managed and monitored, not left to run.
Everyone wants to tell their CEO the brand is running on the most well-known streaming platforms. There's real pressure to be on Netflix, Hulu, and Disney+. These platforms charge high CPMs for a reason.
But prestige and performance aren't the same thing either. There's no evidence yet that premium platforms drive more valuable customers or stronger business impact. What limiting buys to a handful of marquee platforms actually does is raise costs and shrink reach.
Quality of placement matters. Without the right supply-side controls in place, buys can spill into gaming apps, sketchy placements and environments that have nothing to do with the lean-back TV experience you thought you were buying.
But equating a platform's cultural cachet with its advertising effectiveness is a leap the data doesn't support.
This may be the most damaging myth. And it's easy to see why marketers fall for it. CTV feels digital. So CTV targeting should work like digital targeting. Clean, accurate, and deterministic.
It isn’t.
Most CTV targeting relies on IP addresses. IP addresses don't equal people. On their own, they're nearly useless. To work at all, they need to be mapped to a consumer profile using first or third-party data, both of which can be expensive and very error prone. In one real case, IP matching came in at 13% accuracy. Truth Set reported CTV advertisers are on track to waste $7 billion this year on bad identity data.
Smarter targeting skips the guesswork. Rather than relying on third-party data, it accesses media based on what's most predictive of viewer engagement, reaching new audiences without the IP matching errors. For Marketing Architects clients, this approach has delivered a 2X performance gain over traditional third-party targeting.
CTV is delivered through apps and devices, so marketers often assume it's easy to measure. That it avoids linear TV’s notorious measurement challenges.
But ask five people how the same CTV campaign is performing, and you'll get five different answers. Weak test design, misaligned control groups, overly long attribution windows, and unsupported incrementality claims are all common. Catherine shared an example where four different measurement approaches on the same streaming campaign produced wildly different results.
No single metric tells the full story. A layered measurement system that combines incrementality testing, IP matching, econometrics, and brand studies builds a more complete picture.
It's TV. Consumers already see it that way. Whatever appears on the screen counts. That means CTV should be bought and measured like TV, not a display ad that accidentally found its way onto a 65-inch screen.
Doing that well at scale requires AI. Today's fragmented TV marketplace requires analyzing billions of data points across pricing, inventory availability, and viewing behavior. No human team can do that manually. Annika, Marketing Architects' media-buying AI, identifies opportunity across the full TV landscape and predicts performance before a buy is made. The result is more efficient spend and broader reach.
Linear TV still matters, too. Three in four marketers agree that CTV and linear work better together than apart. Linear builds the broad base of reach and awareness. CTV helps convert that awareness into customers. Counting linear out means leaving scale on the table.
83% of marketers plan to grow their CTV investments over the next few years. The brands that win won't be the ones spending the most. They'll be the ones spending smarter.
Want to go deeper on what the TV landscape looks like in 2026? Listen to The Marketing Architects Podcast for more.