Let’s start with a theory I hear a lot, and honestly, I’ve said versions of it myself.
Streaming is where everyone is. CTV is “the future.” And yet it feels oddly hard to reach people there, because most of us go out of our way to avoid ads. We pay for ad-free tiers. We skip. We mute. We scroll. We do anything but sit politely through a pre-roll like it’s 2009.
Here’s the problem with that theory.
Most people are not doing that.
CTV is not getting harder to reach. It’s getting easier, because the economics of streaming are forcing the market in one direction: more ad-supported viewing.
Premium shows cost an absurd amount of money to make. When a single season can run into the hundreds of millions, the math stops working on subscriptions alone. And consumers are already feeling subscription fatigue. Nobody wants to pay for five or six platforms forever.
So what happens? Ad-supported tiers grow. Fast.
The better way to frame the challenge is not “can we reach people on CTV?” The real challenge is: how do we judge it fairly, measure it intelligently, and explain its value without forcing it into a conversion box it was never built to fit?
That is what this article is about.
The CTV reality check
CTV sits in a strange middle ground.
It looks like TV. It feels like TV. The content is premium and the attention is real. But because we buy it programmatically, we often expect it to behave like digital performance media.
That expectation breaks attribution before you even start.
Linear TV was never about immediate action. It was mass reach and broad influence. Programmatic display and search were built for measurable response. CTV borrows elements of both, which is why it can be so powerful and so misunderstood at the same time.
If you buy CTV expecting it to behave like paid search, you will be disappointed.
If you buy it as a premium driver that supports outcomes across the full journey, you are thinking about it correctly.
Why attribution gets messy fast
The programmatic brain wants a clean chain.
Ad served → click → site visit → conversion.
CTV does not naturally give you that chain. It was never designed to.
What CTV does well is start motion. It builds familiarity, compresses consideration, and increases the likelihood that someone later takes an action somewhere else. That “somewhere else” might be search. It might be direct. It might be a retailer. It might be a store visit. It might be a week later after the tenth exposure across other channels.
That is why trying to force CTV into last-click attribution turns it into a losing game.
You are holding it accountable for something it is not supposed to do.
The more useful mindset is this: CTV is an assist channel. It is not meant to be the final touch. It is meant to make the final touch easier.
What to measure instead
So if clicks disappear and attribution gets noisy, what do you track?
The answer is not one metric. It is a set of measurable proxies that tell you whether the channel is doing its job.
The most useful ones tend to fall into three buckets:
- Lift measurement.
Awareness lift. Consideration lift. Brand lift. These are not vanity metrics when they are tied to real movement and measured properly. They are the cleanest way to prove influence in a channel designed for influence.
- Incrementality, used correctly.
Incrementality can mean different things. Incremental reach. Incremental conversions. Incremental movement within the target audience. The key is agreeing on which one you are solving for before you run the test.
- Path signals that show momentum.
CTV often shows up in “secondary” movement: increases in branded search, increases in direct traffic, increases in site engagement, and downstream conversion rate improvements once other channels pick up the user.
One Trade Desk study put this into a simple, useful takeaway: when CTV is paired with other channels like display, audio, and digital out of home, persuasion and brand affinity increase meaningfully. That is what CTV is good at. It is the anchor that makes the rest of the mix work harder.
A quick word on “performance CTV”
There is a wave of vendors promising CTV as a pure acquisition channel.
Some of them do good work. Some of them are selling you a story.
The uncomfortable truth is that if a platform is acting like CTV is just another direct-response line item, and they are not transparent about how they do it, there is usually a trade happening somewhere. Often it is inventory quality. Often it is pricing. Often it is “underpriced supply” dressed up as certainty.
Brands do not buy TV because they want cheap remnant. They buy TV because they want premium adjacency, controlled audiences, and high-attention environments.
If someone is promising you CTV outcomes with no trade-offs and no visibility, ask what you are actually buying.
Why transparent channels get judged harder
Here is another dynamic that shows up constantly.
Transparent channels get punished for being honest.
If a platform gives you log-level detail, placement transparency, and a clear path, you can see the gaps. You can see what is not working. You can see where waste lives. That invites scrutiny.
Walled gardens often get the opposite treatment. We accept promises because the alternative is missing out. We trust them because they are big and because their data advantage feels unchallengeable.
That is not rational, but it is common.
The implication is simple: if you can measure something clearly, do not let that become a reason it gets held to impossible standards. Use the transparency to improve, not to self-sabotage.
Where waste actually comes from in CTV
If you want to find waste in CTV, start here.
Publisher-direct only buying creates frequency blindness.
Consumers do not watch one platform. They bounce across Netflix, Hulu, Peacock, Paramount, and whatever else they are on this month. If you are buying direct in silos, you can easily over-serve the same person without realizing it.
We have all had the experience of seeing the same ad so many times we start to resent the brand. That is not a targeting win. That is waste.
Programmatic waste often looks like vanity optimization.
Completion rate is a perfect example. On CTV, completion is often high by default. Optimizing to it can become a way to feel productive without proving impact.
CTV is valuable because it is addressable and measurable. If you are not using that measurability to control frequency, prove lift, and understand outcomes, you are leaving the best part of the channel on the table.
First-party data matters more than ever, and the first use is suppression
Here is a hot take that should not be a hot take.
If you have first-party data, the first thing you should do with it is suppress the people you should not be paying to reach.
If someone already bought the product, do not keep advertising it to them. If the product is a one-time sign-up, do not chase recent converters with the same message. If someone is in an ineligible segment, stop wasting spend on them.
This is the baseline. It is the cheapest win in media.
From there, first-party data becomes fuel. Seed audiences built from your highest-value customers. Lookalikes. Segmentation. Overlap reporting. Better measurement. Better modeling.
And for categories that struggle with first-party data, the strategy shifts to partnerships. Retailers, marketplaces, and platforms often hold the closest proxy to the audience you want.
If you are a CPG brand, most of your “customer data” does not live on your website. It lives where people buy. The winners will treat that as a planning input, not an inconvenience.
CTV creative: relevance beats polish
Another misconception we should retire.
You do not need a Super Bowl spot to run CTV.
And you also should not run something that looks like it was made on a kindergarten laptop.
The middle ground is not about production budget. It is about relevance.
AI has democratized production. It is easier than ever to create a decent-looking video quickly. That is a good thing. But polish is not the job. Relevance is.
The best CTV creative is the creative that feels aligned to the viewer, the moment, and the content environment. If it is perfectly produced but not for the right person, it is still wasted.
If it is slightly rough but highly relevant, it will often outperform what the brand teams thought was “premium.”
AI is changing programmatic, but it should not replace judgment
AI is already showing up in programmatic buying in three places: forecasting, optimization, and analysis.
It is good at sifting through huge amounts of data and surfacing patterns. It is good at saying “something changed” and “this is likely what happens next.” That is real value.
Where it struggles is where marketing still lives: emotion, intuition, disruption, and taste.
If everyone uses AI the same way, we risk homogenization. Same campaigns, same targeting, same creative, same plateau.
The right mental model is co-pilot. Use AI to handle the scale and the signals. Keep humans responsible for what to test next, what to say, and how to stay distinctive.
The future of CTV is challenging, but it’s also the opportunity
CTV is evolving fast. The ecosystem is fragmenting. Measurement is hard. Clients need education. Expectations need resetting.
It is not easy.
But it is also the clearest direction of travel in media.
The future of CTV is intentional and outcome-driven. The more we get disciplined about what it is supposed to do, how we measure it, and how we connect it to business impact, the more confident the channel becomes.
The goal is not to prove CTV works by forcing it into last-click metrics.
The goal is to judge it by the job it was hired to do.
When you do that, CTV stops being a mystery line on the plan and starts being what it actually is: a premium engine that makes the entire mix perform better.
