Editorial Team We were all sold on the precision and reach of Connected TV (CTV) advertising, but what we got is a mess of media fragmentation that makes real measurement almost impossible. The industry isn’t at a crossroads. It’s stuck in a maze of walled gardens, where every platform has its own ruler. This makes true cross-platform attribution a fantasy.
Key Takeaways
- You’ve got to invest in a unified measurement platform that can actually ingest data from all your different CTV sources to get a single, coherent view of campaign performance.
- The whole industry has to agree on common identity resolution standards to get past the broken, device-centric measurement we’re stuck with in this fragmented market.
- As a media buyer, you have to demand more transparency from CTV publishers about their data methodologies and how they define their audience segments if you want any hope of accurate targeting and reporting.
- You need to start experimenting with incrementality testing to prove CTV’s actual value, because relying only on last-touch attribution is a recipe for getting your budget cut.
The Measurement Mirage in a Fragmented Field
CTV advertising’s potential is huge, but the industry’s rush to cash in has turned standardized measurement into an afterthought. Every major streaming service, smart TV maker, and ad server works inside its own data silo with its own measurement rules. When Nielsen reports there are hundreds of streaming services, each with its own audience and viewing patterns, trying to aggregate performance data is an impossible task. Imagine you’re running a campaign across Roku, Samsung TV Plus, and Hulu. Each one will hand you a report with its own metrics: impressions, video completion rates, maybe even some made-up engagement score. What they won’t give you is a single, coherent view of how those impressions influenced one person’s journey or how they stack up against each other. This isn’t just an annoyance. It actively torpedoes your budget allocation and strategic planning.
I see this play out in media planning meetings all the time. A client gets excited about the high completion rates from one platform while another shows lower numbers, and the first instinct is to immediately shift the budget. But without knowing the methodology behind those numbers (are they measuring the same “completion”? are they filtering bots the same way?), you’re just making reactive moves on bad data that lead to poor results. The absence of a universal ID for households across these platforms means we’re just guessing at reach and frequency, constantly over-counting our audience and burning ad spend. We have to get past simply reporting what each platform spits out and start demanding solutions that actually connect the dots into a picture of audience exposure and response. The argument that proprietary data gives a competitive edge just creates a fog of war for the advertisers footing the bill.
Identity Resolution: The Linchpin of Coherent Measurement
Getting out of this measurement swamp starts with solid, privacy-compliant identity resolution. The old cookie-based tracking we used on the web is basically useless in the CTV space. Instead, we’re left with a patchwork of IP addresses, device IDs, and household graphs, and none of them offer a deterministic, scalable, cross-platform solution. This makes it incredibly hard to attribute conversions, especially for anything lower in the funnel. How are you supposed to connect the dots if a CTV ad leads to someone walking into a store or visiting your website on their phone?
We’ve seen attempts to fix this, from companies like LiveRamp with its Authenticated Traffic Solution (ATS) to the push for universal IDs, but nobody can agree on anything. Every big player, from the broadcasters to the ad tech vendors, is trying to build their own ID, which just makes the fragmentation worse. This is a political and economic problem, not a technical one we can’t solve. Everyone is scared to share data or adopt a standard that might hurt their competitive position. Meanwhile, advertisers are paying the price. We need a real, collaborative effort to build an open, interoperable identity framework that respects privacy but allows for accurate measurement. Without that, any conversation about real ROI in CTV is just guesswork. Trying to plan a CTV campaign right now is like trying to navigate Atlanta traffic without any street signs or a working GPS.
Demanding Transparency and Data Standardization
The other huge problem is the complete lack of transparency in how anything is measured or reported. Advertisers get handed aggregated numbers with almost no context for how they were calculated. What exactly is a “view”? How are you filtering for invalid traffic and ad fraud? What specific audience segments are you targeting, and how are those defined from one platform to the next? These aren’t small details. They’re everything when it comes to evaluating if a campaign worked. A “completed view” on one platform could mean the ad was watched 100% of the way through, while on another it might be 75% or even less, all based on their own internal rules. This makes any direct comparison a total fiction and creates an atmosphere of distrust.
Advertisers have to get more aggressive and demand more granularity and standardization from their CTV partners. That means pushing for common definitions for basic metrics like viewability, completion rates, and audience demos. It means requiring access to raw, anonymized impression logs wherever you can get them, so you can do your own verification and analysis. Groups like the Interactive Advertising Bureau (IAB) have been developing standards, but adoption is spotty and there’s no real enforcement. The responsibility falls on media buyers to make these demands a non-negotiable part of the deal. If we don’t, we’re just accepting whatever we’re given instead of getting what we actually need to make smart decisions. We need to flip the power dynamic, because we’re the customers and we deserve to know what we’re buying.
Beyond Last-Touch: Embracing Incrementality
The fragmented mess in CTV also makes last-touch attribution models, which were already questionable, pretty much useless. In a world where people are bouncing between their TV, phone, and laptop all day, giving 100% of the credit to the final touchpoint is a massive oversimplification that ignores what CTV is good at. A CTV ad is great for building brand awareness or initial interest, which might lead to a search or a direct website visit on a phone hours or days later. If you only measure that last click, the CTV impression gets zero credit for its contribution.
This means we have to switch to smarter measurement, specifically incrementality testing. This is where you run a controlled experiment, showing CTV ads to one audience segment while holding them back from a control group. By comparing the lift in conversions or other KPIs between the two, you get a much more accurate read on the real, incremental value your CTV spend is generating. Yes, it’s logistically harder to set up across all these different platforms, but the insights you get are infinitely more valuable than just chasing last-click vanity metrics. Agencies with good data science teams are already doing this, using models that can isolate the actual impact of CTV in a bigger media mix. It’s the harder path, but it’s the only one that leads to real accountability and smart spending.
The way CTV measurement works now isn’t sustainable for any advertiser who cares about return on investment. The industry has to get out of its proprietary silos and start using open, standardized, and transparent measurement. Demand better from your partners, invest in real attribution models, and push hard for a unified identity framework that works for everyone and respects privacy. That’s the only way CTV will ever live up to what we were told it could be.
Why is CTV advertising measurement so challenging?
The challenge is media fragmentation. You have tons of streaming services and devices operating in their own little worlds, each with its own data and measurement rules, which makes cross-platform attribution and counting unique users a nightmare.
What is “identity resolution” in the context of CTV?
In CTV, identity resolution is about connecting the dots, linking data points from different devices and platforms back to a single household or user. This lets you see the whole customer journey and get an accurate count of reach and frequency.
How does a lack of transparency impact CTV advertisers?
A lack of transparency means advertisers are in the dark about how key metrics like a “view” or “completion” are even defined by different platforms. This leads to inconsistent reports, makes it impossible to compare performance, and causes you to waste ad spend.
What is incrementality testing and why is it important for CTV?
Incrementality testing compares a group that saw your CTV ads to a control group that didn’t. This lets you measure the true lift in sales or other goals that came only from the ad campaign. It’s important because last-touch attribution completely misses the upper-funnel impact that CTV is great at.
What steps can advertisers take to improve their CTV measurement?
To improve CTV measurement, you need to demand more transparency and standard metrics from publishers, put money into unified measurement platforms, look into privacy-safe identity resolution tools, and use incrementality testing to figure out what your campaign is actually worth.
