Welcome back to The Halo! In this issue we cover:
The Take: Where does CTV actually land?
From the Blog: Why Ads Manager does not tell the whole story on Meta, and how cross-channel halo changes how performance reads.
Discovery of the Week: $62.4M analyzed across one year: 2.21x Modeled ROAS, 96.6% average halo across all four quarters.
Prescient Voices: What changes when you measure CTV with the right commerce mix as your reference.
Three Things: Reads on omnichannel measurement, promotional effectiveness, and CTV incremental lift.
The last two issues looked at where revenue actually lands across channels, and how the saturation curve on your dashboard can mislead you. CTV makes both impossible to ignore.
We just published our CTV 2025-2026 Benchmark: $62.4M in tracked spend over a full year.
The headline is a 2.21x Modeled ROAS. But that number hides two readings of the same data:
• The biggest CTV spenders see 51.2% of revenue land on retail.
• The typical CTV brand sees 47.7% land on DTC.
Both are right, they just answer different questions.
Welcome to The Halo. Each week: one piece of measurement analysis, the best of the blog, one real customer result, and three reads worth your time.
Let's dive in.
01 · THE TAKE
Where does CTV revenue actually land?
The benchmark covers $62.4M across 48 brands over a period of 49 weeks, at a 2.21x Modeled ROAS.
In the opener, we mentioned that the biggest CTV spenders see:
• 51.2% of revenue land on retail.
• The typical CTV brand sees 47.7% land on DTC.
Those two numbers feel like they contradict each other. But they don't. They're the same dataset, weighted two different ways.
Weight by spend, and the picture tilts toward the biggest CTV advertisers, who tend to be retail-distributed. Retail dominates.
Weight by brand, and every advertiser counts equally. From that angle, the typical CTV brand is DTC-led.
If you're DTC-led and you read the spend-weighted number, you'll conclude your CTV halo should land on retail. The number isn't wrong. It just belongs to a different kind of business than yours.
THE FOUR ARCHETYPES
The benchmark identifies 4 commerce-mix archetypes by where CTV halo actually lands.
1. DTC-led. 92% of CTV halo lands on the brand's own ecommerce. The rest ends up on Amazon.
2. Amazon-led. Amazon is the primary destination, but CTV's lift there is weaker than for DTC-led brands. 72% of halo lands on Amazon, with DTC at a median of 18%.
3. Retail-led. The biggest spenders in the dataset, all retail-distributed. 79% of their CTV-generated revenue ends up in retail.
4. Balanced. No single channel dominates. CTV revenue spreads across destinations, with none crossing 50%.
If you're Amazon-led and you measure against the DTC median, or DTC-led and you measure against the spend-weighted retail number, you're asking the right question of the wrong benchmark. The wrong conclusion forms before the budget meeting, not in it.
WHAT THE DATA SAYS ABOUT THE CHANNEL
The benchmark summed up in one line:
CTV intensifies the channel where the customer already buys, instead of pushing them somewhere new. CTV has no sales surface of its own, so by design every dollar lands somewhere else. Scaling it without cross-channel visibility is flying blind.
CTV behaves the same way at peak. The values are bigger, but the pattern holds.
BFCM doesn't change the playbook. You need the same measurement setup you'd use the rest of the year: visibility into where the halo actually lands, not just what the channel reports on its own dashboard.
THE QUESTION WORTH ASKING
Before you use any CTV benchmark to make a decision, ask one question: do the brands in it share your commerce mix?
If the answer is "I don't know" or "probably not," use the benchmark for scale context, not as an efficiency reference.
The halo gap between archetypes is wide enough that a DTC-led brand reading a retail-weighted benchmark will draw a conclusion that doesn't apply to its business.
Download the benchmark for free →
02 · FROM THE BLOG
How to measure Meta effectively: why Ads Manager does not tell the whole story.
Ads Manager reports useful operational data, but it paints an incomplete picture of real performance.
• View-through conversions are unreliable.
• It misses cross-channel halo: branded search, direct traffic, sales on other channels that Meta exposure helped drive.
• Its attribution windows rarely match the real purchase cycle.
The post walks through why independent measurement surfaces Meta's true contribution. It echoes today's Take from a different angle: even inside one channel, what platform data shows you depends on the question you're asking.
03 · PRESCIENT VOICES
Someone sees your CTV ad. Doesn't click. Later, they Google your brand and buy on Amazon. That sale? Your CTV ad drove it. But Meta, Google, and your MTA tool can't see it.
04 · THREE THINGS WORTH READING
1. The biggest challenges of omnichannel marketing (and how brands are solving them)
Once you sell through physical retail or Amazon, omnichannel strategy stops being optional. Your customers already move across those channels. The post walks through three problems brands run into: data fragmentation across platforms, attribution silos that can't be reconciled, and org structures that reward isolated-channel performance over joint impact.
2. What promotional effectiveness really measures, and where teams get it wrong
Most brands measure promo success by sales lift during the campaign window. That misses cannibalization across products and channels, the halo that extends past the window, and the blind spots baked into platform data itself. The lesson applies directly to reading aggregated benchmarks without brand-level context.
3. Leading tools for measuring incremental lift from CTV ads
CTV doesn't produce clicks. Viewers search for the brand later, visit the site directly, or buy on Amazon. The post centers on BrüMate: native platforms ranked CTV last, but once Amazon sales and halo effects came in, it turned into the top-performing channel.
The Halo
Exclusive insights, every week.
Subscribe to The Halo for sharper marketing thinking.