Everyone’s talking about Connected TV advertising as the “next big thing.” But here’s what they’re missing: CTV isn’t just another advertising channel-it’s forcing us to completely reimagine how we measure marketing effectiveness.
The uncomfortable truth? The industry is still trying to cram CTV into frameworks designed for 1960s broadcast television.
The Measurement Paradox Nobody Wants to Address
Here’s what keeps me up at night: CTV advertising represents the first time in history where we have too much data while simultaneously having no standardized way to measure what actually matters.
The Multi-Identity Problem
Traditional digital advertising trained us to worship the individual user ID. Facebook’s pixel, Google’s cookies, your retargeting campaigns-they all rely on persistent identifiers that follow people around the internet.
CTV demolished this model overnight.
One television set isn’t one person. It’s a household. Sometimes it’s a college dorm. Sometimes it’s a waiting room. The IP address tells you there’s a Roku at 123 Main Street, but it doesn’t tell you if the person watching your skincare ad is a 35-year-old woman interested in anti-aging products or her 14-year-old son scrolling TikTok while Netflix plays in the background.
Here’s the controversial take: This isn’t a bug-it’s a feature we’re too stubborn to appreciate.
The advertising industry spent the last 15 years obsessing over hyper-targeting, convincing ourselves that reaching the “perfect” individual at the “perfect” moment was the pinnacle of advertising sophistication. CTV is forcing us to remember something the masters of advertising knew intuitively: context and creativity matter more than precision targeting.
The Attribution Theater We’re All Performing
Every week, I see another agency pitch deck claiming they can “fully attribute” CTV advertising results. It’s mostly theater.
The dirty secret of CTV attribution is that we’re using probabilistic modeling that would make a Vegas oddsmaker blush, then presenting it to clients as gospel truth. We’re matching TV viewing data with mobile location data with purchase data with IP addresses, running it through algorithms, adding a confidence score, and calling it “attribution.”
But here’s what’s actually happening: We’re recreating the same fundamental limitations of traditional TV advertising while charging digital advertising prices.
The Three Attribution Lies We Tell Ourselves
Lie #1: “We can track conversions from CTV ads”
What we actually mean: We can see correlation between CTV ad exposure and conversion events. We cannot definitively prove causation in most cases. Your attribution model is making educated guesses based on statistical probability.
Lie #2: “CTV provides the targeting of digital with the impact of television”
What’s really happening: CTV provides the approximate household-level targeting of direct mail with the approximate reach of local cable, measured with the approximate accuracy of a well-designed survey. It’s powerful, but it’s not magic.
Lie #3: “Our CTV campaigns are fully measurable”
The reality: We’re measuring impressions (maybe), estimating attention (definitely), guessing at attribution (absolutely), and hoping our control groups are actually controlled (they’re not).
The Uncomfortable Truth About CTV Creative
Here’s an angle the creative agencies don’t want you thinking about: Most CTV advertising creative is catastrophically bad because we’re optimizing for the wrong engagement metric.
The industry has trained itself to believe that “completion rate” is the holy grail of CTV creative performance. If someone watches your entire 30-second ad instead of hitting “skip,” you won-right?
Wrong.
Consider this: The highest completion rates on CTV ads come from one of two scenarios:
- The viewer is literally not in the room (they walked away to grab a snack)
- The viewer is on their phone and not actually watching
The ads that get “completed” most often are the ads being ignored most successfully.
This is the exact opposite of effective advertising.
Meanwhile, the ads that generate the most skips might actually be working-they’re provocative enough that viewers make an active decision to engage with the content by removing it. That’s attention. That’s processing. That’s the viewer’s brain actually acknowledging your message exists.
But we’ve built our optimization models backward. We’re rewarding the wallpaper and punishing the art.
The Fragmentation Crisis We’re Ignoring
There are currently over 200 CTV platforms, each with different ad specs, different targeting capabilities, different measurement standards, and different definitions of what constitutes an “impression.”
Let that sink in: We have less standardization in CTV advertising than we had in broadcast television in 1975.
A 30-second ad on Hulu is bought differently, served differently, measured differently, and priced differently than the “same” ad on Peacock, which is different from Roku, which is different from Samsung TV Plus, which is different from Pluto TV.
This isn’t a temporary inconvenience we’ll solve with better technology. This is a fundamental restructuring of how television advertising works, and it’s creating a two-tiered system:
Tier 1: Massive brands with eight-figure CTV budgets who can afford to work with aggregation platforms, employ full-time CTV specialists, and run sophisticated attribution studies.
Tier 2: Everyone else, who’s buying CTV ads through resellers, getting opaque reporting, and essentially playing a very expensive guessing game.
The gap between these tiers is widening, not shrinking.
The Attention Economy Meets the Distraction Device
Here’s the existential crisis at the heart of CTV advertising: Television used to be the lean-back medium. Now it’s the background medium.
The average CTV viewer is second-screening 87% of the time. That means your beautiful brand film, your clever product demo, your emotional storytelling-it’s competing with Instagram, work emails, and online shopping.
But here’s where it gets interesting: This might actually make CTV advertising more valuable, not less.
The Ambient Branding Hypothesis
I’m going to propose something controversial: The future of effective CTV advertising isn’t about demanding attention-it’s about earning ambient awareness.
Think about how radio advertising evolved. The most effective radio ads aren’t the ones that make you stop what you’re doing. They’re the jingles that worm into your subconscious while you’re driving, the brand mentions that register peripherally while you’re cooking dinner, the sonic branding that becomes part of your environmental soundtrack.
CTV advertising is heading toward the same model, except we’re too busy trying to recreate the Mad Men era to notice.
The brands that will win in CTV aren’t the ones creating “breakthrough” creative that demands full attention. They’re the ones creating consistent, recognizable, ambient brand presence that registers even when the viewer is only 30% engaged with the screen.
This requires completely different creative strategy:
- Sonic branding that works without visual attention
- Visual branding that works in peripheral vision
- Message simplification that conveys value in 3 seconds, not 30
- Repetition strategies that build familiarity without causing annoyance
The Fraudulent Paradise
Let’s talk about the topic everyone whispers about but nobody wants to address publicly: CTV ad fraud is substantially worse than anyone’s willing to admit.
Server-side ad insertion (SSAI) has made CTV advertising significantly harder to fraud-proof than traditional digital advertising. When ads are stitched directly into the video stream at the server level, traditional fraud detection methods become nearly useless.
Here’s what’s actually happening in the CTV fraud ecosystem:
Fraudulent App Spoofing: Fraudsters create fake CTV apps that claim to be premium inventory. These apps generate millions of “impressions” from bots, virtual machines, or device farms.
SSAI Exploitation: Because the ads are inserted server-side, bot detection pixels can’t be fired effectively. The impression happens in the video stream itself, making it look completely legitimate.
Residential Proxy Networks: Sophisticated fraud operations route bot traffic through residential IP addresses, making it appear as if real households are streaming content.
The industry estimates suggest 10-15% of CTV ad spend is fraudulent. I believe the real number is closer to 25-30%, based on patterns I’ve seen analyzing cross-platform campaign data.
The verification vendors are selling certainty they cannot actually provide. The platforms are incentivized to underreport the problem. And advertisers are hesitant to look too closely because they’ve already committed their budgets.
The Pricing Inefficiency Costing You Millions
Here’s an angle I’ve never seen properly explored: CTV advertising has created the most inefficient pricing market in modern advertising history.
The same impression-literally the same ad shown to the same device at the same time-can vary in price by 400-600% depending on which buying platform you use, which sales channel you access, and which intermediary is involved.
Consider this supply chain:
- Content owner (NBC, Paramount, etc.)
- Streaming platform (Peacock, Paramount+)
- SSP (supply-side platform)
- Ad exchange
- DSP (demand-side platform)
- Trading desk
- Agency
- Advertiser
Each layer is taking a cut. By the time that impression reaches the advertiser, you might be paying $45 CPM for inventory that the content owner is selling for $8 CPM.
The spread isn’t value-add. It’s structural inefficiency masquerading as expertise.
The brands that figure out how to collapse this supply chain-by going direct to platforms, building in-house CTV capabilities, or leveraging programmatic guaranteed deals-will have a 3-5 year advantage over competitors who continue buying through traditional agency models.
The Creative Opportunity Everyone’s Missing
While everyone’s obsessing over targeting and measurement, there’s a massive creative opportunity sitting in plain sight: CTV is the only digital advertising medium where long-form creative actually outperforms short-form.
In every other digital channel, shorter is better. Instagram stories, TikTok videos, Facebook feed ads-they all reward brevity. But CTV flips the script.
Our testing at Sagum has shown that 60-second CTV ads consistently outperform 15-second and 30-second versions across every meaningful metric: brand recall, purchase intent, message comprehension, and even completion rates (when the creative is actually engaging).
Why? Because CTV viewers are in a lean-back entertainment mindset. They’re not scrolling. They’re not jumping between apps. They’re settled in for a viewing session. They’ll tolerate-and even appreciate-longer creative that actually tells a story, explains a value proposition, or creates an emotional connection.
But here’s the catch: Only about 5% of brands are taking advantage of this.
Most CTV creative is just repurposed 30-second TV spots that were designed for broadcast interruption, not streaming engagement. They’re boring, generic, and optimized for a medium that no longer exists.
The CTV Creative Framework That Actually Works
Based on millions in CTV spend and hundreds of creative tests, here’s what we’ve learned:
0-3 seconds: Sonic and visual branding that works even if the viewer is looking at their phone
3-8 seconds: Problem identification or pattern interruption that earns active attention
8-45 seconds: Value proposition, product demonstration, or emotional storytelling
45-60 seconds: Clear call-to-action and memory encoding reinforcement
This structure acknowledges the reality of partial attention while rewarding full attention. It works in the background while inviting foreground engagement. It’s ambient and active.
The Platform Wars and What They Mean for You
Here’s the strategic angle most analyses miss: The CTV platform wars aren’t about streaming content-they’re about advertising control.
Every major CTV platform is simultaneously:
- A content distributor
- An advertising network
- A data marketplace
- A measurement provider
This vertical integration creates massive conflicts of interest that directly impact your advertising effectiveness.
When Roku tells you their advertising platform delivered great results, remember: Roku controls the content, the ad serving, the measurement, and the reporting. They’re the player, the referee, and the scorekeeper.
The strategic implication: Never rely on platform-provided measurement alone. Always insist on third-party verification, always run holdout tests, and always be skeptical of results that seem too good to be true.
The Regulatory Tsunami on the Horizon
Here’s what’s keeping the smart people in CTV awake at night: The regulatory framework that’s about to hit CTV advertising will be more disruptive than GDPR and iOS 14 combined.
CTV advertising currently operates in a regulatory gray zone. It uses television-style content regulations (looser) while employing digital-style data collection practices (much more invasive). This won’t last.
The coming regulatory changes will likely include:
- Strict limitations on household-level targeting
- Mandatory opt-in for advertising data collection
- Significant restrictions on cross-device tracking
- Required disclosure of data sharing practices
- New privacy standards specific to CTV environments
Brands that are building their CTV strategies on current data collection practices are building on sand.
The smart move? Start testing privacy-first CTV approaches now. Develop creative that works without behavioral targeting. Build measurement systems that don’t rely on persistent identifiers. Treat current data capabilities as a temporary advantage, not a permanent feature.
The Counter-Intuitive CTV Strategy
After analyzing millions in CTV spend across dozens of industries, here’s the strategic approach that consistently outperforms:
Treat CTV like brand building, not performance marketing.
I know-every CTV platform is selling you on “TV-like reach with digital-like targeting and measurement.” But the brands getting real ROI from CTV are the ones ignoring that pitch.
They’re using CTV for:
- Mass awareness at efficient CPMs (not micro-targeted precision)
- Emotional brand storytelling (not direct response creative)
- Long-term brand building (not immediate conversion optimization)
- Complementary brand presence (not standalone campaign performance)
When measured against performance marketing KPIs, CTV looks expensive and inefficient. When measured against brand building objectives, CTV is one of the most cost-effective channels available.
The brands that figure this out will win. The brands that keep trying to make CTV into Facebook Ads for TV will waste millions.
