Signal vs. Static: How We Get to True CTV Transparency

Summary

---

Marketecture Podcast Episode Summary

Episode Title:

Signal vs. Static: How We Get to True CTV Transparency

Date: April 13, 2026
Host: Ari Paparo
Guests:

• Mario Diaz (CEO, Pure 39)
• David Nuremberg (SVP Digital, Intermedia Advertising)

---

Episode Overview

This episode features a live session from Marketecture’s Market Live event, with Mario Diaz and David Nuremberg exploring the persistent problem—and potential solutions—to transparency in Connected TV (CTV) advertising. They discuss the current “proxy” nature of CTV signals, data blind spots, why true content-level transparency is tough but essential, and practical steps for publishers and buyers. The tone is direct, data-driven, and focused on actionable insights for both buyers and sellers in the CTV ecosystem.

---

Key Discussion Points & Insights

1. The State of CTV Signals: Proxy Over Precision

• CTV has leaned heavily on proxy signals (e.g., app names, deal IDs) instead of true content-level data, which is a carryover from programmatic web advertising.
• Transparency and richness of signal in CTV inventory is lacking compared to linear TV, where buyers always knew what shows they were buying.
  • “CTV really is the next evolution of linear. Right. It's all television at the end of the day… But you really don't get much content level data in the bid.” (David, 02:57)

2. Why Lack of Signals Hurts Buyers

• Without robust, consistent content signals, algorithms can't optimize performance or reach. Buyers end up purchasing "bundled" or blind CTV inventory.
  • “Algorithms can't optimize and get deeper and deeper into the inventory. Right? …If they don't know what they're buying, then they can't get smarter.” (David, 04:10–04:29)
• There’s a myth among publishers that transparency enables "cherry-picking." In reality, buyers need scale and won't nitpick; they just want performance and reach.

3. The Data Reality: Blind Inventory & Its Implications

• Pure 39’s research found 60% of all CTV bid requests are blind, with only 33% carrying any meaningful content signal. (Mario, 05:08)
• A demonstration showed that the same movie (e.g., "Man of Steel") appears almost signal-less to DSPs, making targeting by genre or content nearly impossible.
  • “So there's actually maybe just one piece of genre or what's highlighted is there's a vast majority that is just blind or other. So it's the same piece of content, just viewed differently.” (Mario, 06:19)

4. Signal Variation Across Supply Paths

• Even when publishers want to be transparent, signal fidelity varies drastically based on SSP/partner integrations.
• David recounted posting public data on LinkedIn showing how different SSPs pass different amounts of data for the same publisher, often without the publisher realizing it.
  • “They were surprised…each of these different SSPs have different levels of controls…What they then did is…told them, all right, we want all our data shared…And this got fixed after the fact.” (David, 09:14)

5. Misleading Metrics & The Dangers of Blind Optimization

• Completion Rate (the percentage of ads watched to the end) can be completely misleading without content-level signals.
  • “If you're just measuring completion rate without a signal…you're basically commoditizing your buy.” (Mario, 10:10)
  • “I can tell you right now it's 100%...for a screensaver or a fireplace app.” (Mario, 10:48)
• Example: Documentary inventory was inflated due to mislabeled “yule log”/fireplace apps.

6. The Impact of UGC & Blind Inventory

• User Generated Content (UGC) is present and spikes unexpectedly in CTV, creating further inconsistency in campaign quality and measurement. (Mario, 11:24)

7. Lessons from YouTube & Spectrum

• YouTube provides more content-level transparency than most open CTV platforms. (David, 12:51)
• Spectrum rolled out 100% transparency, sending show-level data in every bid and saw dramatic improvements:
  • Targeting top shows outperformed all other tactics (even at higher CPMs).
  • “That tactic was the best performing tactic on the buy…It was by far the best performer and it had higher CPM too.” (David, 13:18–14:10)
  • Cherry-picking is a myth; buyers are incentivized to spend their full budgets, not restrict scale. (14:25)

8. The Path Forward: Increasing Signals for Everyone

• Increase in global program-level signals seen in Q4 (potentially due to educational/market pressure).
• Sports (e.g., NBA) is especially underlabeled and thus under-monetized in CTV due to lack of signals.
  • “Live stream content should really command a premium. But it's being bundled…which will then disincentivize you to pay more for the live stream content.” (David, 16:12)

9. Industry Responsibility & Legal Barriers

• Signal fidelity is a choice made by both buyers and publishers.
• Buyers must demand content signals from the market and factor it into workflow and partner selection.
• Legal confusion, specifically around the Video Privacy Protection Act (VPPA), is a persistent blocker to content-level data sharing. The law is outdated and affects data sharing decisions. (David, 18:26)

10. Practical Next Steps for the Industry

• Buyers: Ask for and incorporate content signals, use reportable signals in campaign planning and optimization.
• Publishers: Work with partners and metadata companies, ensure they’re sharing rich signals, and understand the value proposition.
• Everyone: Industry groups (IAB, VAB, Tech Lab) must clarify and modernize legal frameworks like the VPPA.
  • “If you do an SPO, then you should start looking at how much signal per publisher is coming through their different supply paths. Especially in CTV, most publishers use maybe two to three primary SSPs. So it's really not too much work to look at it. You just got to do the work.” (David, 22:29)

---

Notable Quotes & Memorable Moments

• On the Proxy Problem:
  “The market has leaned on these proxies, such as apps becoming proxies for quality or deals becoming proxies for transparency.” (Mario, 01:33)

• On Data Blindness in CTV:
  “What you're looking at here is what a very large publisher looks like to your DSP partner…and there's no way to kind of prove that a publisher's inventory is valuable or not.” (Mario, 05:53)

• On Supply Path Optimization (SPO):
  “When we're thinking SPO, what we should be thinking about really is which supply paths are getting the richest signals coming through.” (David, 07:21)

• Cherry Picking Myth:
  “Media buyers need to spend their budget…they're going to pick as much inventory as they need to deliver that budget.” (David, 14:25)

• On Legal Barriers:
  “The VPPA…is a law that applies to cassette rentals that has been retrofitted for digital. And I think that alone is a big reason why we're being held back right now.” (David, 18:26)

• On Practical Steps:
  “If you do an SPO, then you should start looking at how much signal per publisher is coming through their different supply paths…it's really not too much work to look at it. You just got to do the work.” (David, 22:29)

---

Timestamps for Key Segments

| Time | Topic / Segment | |-------------|----------------------------------------------------------------------------------------------------------------| | 01:33 | Introduction to CTV signals and proxy problem (Mario) | | 02:57 | Why CTV transparency lags behind linear TV (David) | | 05:08 | Data: 60% of CTV inventory is blind (Mario) | | 09:14 | Variance in signal fidelity across different SSPs and fixing via publisher intervention (David) | | 10:10 | Why completion rate is an illusion without signals (Mario) | | 10:48 | The “yule log”/fireplace example skewing documentary campaign metrics (Mario) | | 13:18 | Spectrum case study: show-level data increases performance (David) | | 16:12 | Challenges labeling and pricing live sports in CTV (David) | | 17:26 | The frustration of measuring “premium” content without signals (David & Mario) | | 18:26 | The impact of the outmoded VPPA law on data transparency (David) | | 19:50 | Q&A: Are some DSPs more transparent than others? (David & Mario) | | 21:34 | Q&A: What can be done now despite legal issues? (David) | | 22:29 | Final advice: Buying with signals and supply path evaluation (David) |

---

Conclusion

The episode provides a candid, data-rich exploration of why CTV buying is still stuck in the “static” era—hampered by poor signal fidelity, system fragmentation, and outdated legal frameworks. Both buyers and sellers are encouraged to prioritize transparency, actively incorporate content-level data into planning and optimization, and push the industry (and lawmakers) for clarity and modernization.

Takeaway:
The next evolution in CTV requires everyone—buyers, publishers, vendors, and lawmakers—to break from proxies, demand authentic content signals, and do the operational work needed to build a transparent, high-performing CTV ecosystem.

Transcript

[A]

This episode is brought to you by indeed. Stop waiting around for the perfect candidate. Instead, use Indeed sponsored Jobs to find the right people with the right skills fast. It's a simple way to make sure your listing is the first candidate. C. According to Indeed data, Sponsored Jobs have four times more applicants than non sponsored jobs. So go build your dream team today with Indeed. Get a $75 sponsored job credit@ Indeed.com podcast. Terms and conditions apply.

[B]

Hello, this is Eric Paparo. Today we're sharing a recording from our Market live event on March 10th and 11th. This is Mario Diaz, the chief executive officer of Pure 39, in conversation with David Nuremberg, the SVP of digital for Intermedia Advertising. The title is Signal versus How We get to True CTV Transparency. And it's about the signals that enable buying and selling programmatic advertising in ctv. So this should be an interesting one for those of you interested in data, ctv, et cetera. And Mario is always a very compelling speaker, so I hope you enjoy this.

[C]

Thank you. Thank you. I promised Ari I would take a picture of the crowd. Everybody wave or go on the post. Thanks. Well, we better be good at standing room only.

[D]

Hey, we brought the crowd.

[C]

Thanks everybody for joining us. My name is Mario Diaz. As David said, I really, really psychedel to talk to you guys about this. We're going to be talking about CTV signals and we're going to be going through as a promise to the market. We're going to be real, we're going to show data, we're going to have upfront conversation about it so that we can all walk away with some material on what to do because we're all in this together. Before we go in, it's really important to think about what has happened in CTV since most of us have been in this space for a while, consumers eyeballs, it was just typical playbook. The. The advertising followed but the infrastructure and the signals didn't. And what we have like moved into, if we think about it, is kind of like this proxy signal for CTV where the market has leaned on these proxies, such as apps becoming proxies for quality or deals becoming proxies for transparency. And the idea of premium, which is. Which is large part what you spend a lot of time on and talking to publishers about premium content, it was more about the supply path. So we're going to get into all this. But kind of like before we go, like, how do you think about this as a buyer in the community?

[D]

Yeah, I mean, I mean, to me, CTV really is the next evolution of linear. Right. It's all television at the end of the day, linear has always been bought at the show level. Right. Network, channel, show level. That's how you quantified the premium nature. You knew how many people are watching those shows, you would get all the Nielsen stats.

[C]

It's still like that now in linear.

[D]

It's still like that now. So performance media buying, you could tie an outcome to a specific show on the linear side. Now with ctv. Right. The beauty of ctv, what should be the beauty of it is that you get all this rich data in real time that you could, you should be able to decision on programmatically. Right. Unfortunately, the reality of the, of the moment really is that it's almost like we've devolved how we're buying from the linear. Linear, because you really don't get much content level data in the bit.

[C]

So all of these signals that you're seeing up here are traditionally still available and linear right now. But it's not what's happening in ctv.

[D]

Well, slowly but surely it's starting to happen, but not nearly as much as we'd like. Right. Because I mean, at the end of the day when we plan, right.

[C]

We.

[D]

We understand what content our audiences watch. Right. Then we want to go after that content. If we don't have the right signals in the bitstream.

[C]

Yeah.

[D]

Then our algorithms can't go after that content. Right, Right. And we're stuck buying these blind CTV inventory bundles, which either they work, which is great, or they don't, and then we pull budget. But the problem is without signal richness, algorithms can't optimize and get deeper and deeper into the inventory. Right?

[C]

Yeah.

[D]

So if they don't know what they're buying, then they can't get smarter.

[C]

So there's an interesting dynamic because a lot of publishers will question whether they should pass signals or be more transparent at the risk of some cherry picking. But it's actually the opposite. It's like you as a buyer, you can't.

[D]

It's like I'm not sitting, I'm not buying a programmatic campaign with 20,000 sites and then sitting there looking at each and every site trying to cherry pick those. Right. Ultimately, I want to buy, to go where the performance and the reaches. Right. And I trust my algorithms to do that at the moment. It's tough. Yeah.

[C]

So let's get into some of the data because we did promise to get into real stuff. So company pure thriving. We released research today. We'll have a download link or find me afterwards. I'll show you where to get it? How many of you on the sell side of CTV or just in general buy side? Okay, that's just a really, really good blend. So you probably know this when you're doing ctv. What we've looked at is the entire ecosystem and found that 60% of all calls are basically blind. 30, the next 33% have some level of program signal. And this is what really creates the problem. But to really break this down, and I don't know how many of you are in the weeds, but this was our promise to show data. What you're looking at here is what a very large publisher looks like to your DSP partner. And what you're looking at here is what is basically blind. So this publisher has roughly 60% of their inventory that has zero signal on it. So what happens is you have this commoditization of inventory and there's no way to kind of prove that a publisher's inventory is valuable or not where this breaks down even further. So stay with me for a second on this. What you're looking at here is a movie, right? One movie. Man of Steel, Superman, if you've seen it, we've done work to authenticate what that movie is in the midstream and all the richness of the metadata associated with it. This is just a sample. The reality of who are all the buyers over there? I'm talking to you. So if you wanted to find action and adventure and service, it's nearly impossible to find with consistency because on the right, you're seeing what an actual DSP sees. So there's actually maybe just one piece of genre or what's highlighted is there's a vast majority that is just blind or other. So it's the same piece of content, just viewed differently. How does that impact you guys?

[D]

Well, I mean, it even goes beyond that. Right. So like the same publisher will have different levels of signals coming through, different SSPs. Right?

[C]

Yeah.

[D]

So like we want to talk SPO, but when we're thinking SPO, what we should be thinking about really is which supply paths are getting the richest signals coming through. So my algos can make the best decisions. Right. Versus just chasing efficiency, which I think is the primary thing that people are going after right now. Yeah, now people would not be so focused on cheap CPMs and efficiency if they actually knew what they were buying and could value inventory accordingly.

[C]

Because a lot of the inventory all looks the same.

[D]

Exactly. It all looks the same.

[C]

Yeah.

[D]

And at the end of the day, in an AI first world, and I'm going to try not to say AI more than we promised everybody.

[C]

We wouldn't say AI, dude. At least I did.

[D]

The algorithms look at each and every single impression and predict the likelihood based on historical data and all the other metadata signals in the bitstream, how likely it will convert and then it'll bid up or down based on that probability. If it doesn't have any data to look at, everything's going to be priced the same.

[C]

You're just going to look at audience signal. So we want to break down. You talked a little bit about supply pass, but just to break this down a little bit further. And again, this is all data that we've been looking at with David across Pier 39. This is one publisher and how they pass CTV signals across multiple different exchanges. And it varies with how much they pass. But what is the interesting part is that same publisher has basically two blocks of inventory. There's one set of inventory that carries a signal and then there's another set of inventory that does not carry a signal. Same publisher. And what's even more striking is for those of you on the buy side that are working with DSPs, that same inventory is going to look different in how you're buying based on which partner that you have. So to the point of supply path optimization, you've done a lot of work with what you had posted the other.

[D]

Yeah. So I actually posted this data on LinkedIn a few months ago. And this is one publisher, the same publisher, through three different SSPs. And there's just a big degree of difference in the fidelity of the data, of the content level data that they're passing through each ssp. Now, this publisher specifically actually wants to be transparent. They want to pass back as much data as possible. They actually reached out to me after I posted this and they were surprised that there was such a degree of difference in how much they were passing. And what they didn't realize was that each of these different SSPs have different levels of controls and different settings that they need to unlock in order to get those signals going. So what they then did is they went back to each of those SSPs and told them, all right, we want all our data shared out as far as content goes. And this got fixed after the fact.

[C]

All right, we're gonna jump ahead a little bit cause we're running on time. This is mine, actually. All right, so how many people have optimized or been asked to optimize or have been asked to evaluate complet rate as a metric? A lot of people for ctv. Okay. That without a Signal is basically an illusion. Again, some of the research that we're, that you'll be able to download later, if you're just measuring completion rate without a signal of what that inventory was, you're, you're basically commoditizing your buy.

[D]

What do you guys think? A screensaver or a fireplace app? What, what do you think the completion rate on that will be?

[C]

I can tell you right now it's 100% from the work that we did and the inventory is out there. In fact, in Q4 there was a very large percentage of documentary inventory that was eating up campaigns that we were measuring. And when we got into it, it was actually a mislabeling of yule log. So pretty sure yule log was not documentary last time I checked. But I mean, I don't know how wood burns. How wood burns maybe, I don't know. But anyway, so where this actually materializes even more. And this is why it's so important to work with signals and supply paths because there are things that are possibly eating budget or eating spend away from quality content. What you're looking at here is the last few months of traffic for what we labeled UGC in ctv. Yes, it's out there. What's fascinating about this is it's relatively consistent and then a couple days it spikes. Well, because it spikes, it commoditizes the impressions and you're running on things that you may not want. Now maybe you're targeting UGC or what have you, but these are the types of signals that are coming into the stream and with your DSP partners that you need to pay attention to. Yeah.

[D]

So you need the signals to make good decisions, right? Yeah, especially in the outcome. First world, we want to talk about performance. Ctv. I truly don't believe that we'll be able to realize the full potential of CTV and actually be a performance channel unless we go fully transparent and start being able to decision off the content. The reason Meta and Google are able to drive performance at the end of the day is because they understand every single inch of real estate and they're able to price it accordingly and predict how likely it is to convert against their, their, their clientele.

[C]

So you're a buyer of both. How does YouTube and CTV compare?

[D]

I mean, YouTube, you do get content level data in YouTube. That's the, that's kind of the dirty secret. We want to knock on Google and I'm the first person to knock on Google. I love knocking on Google, but they're actually more transparent on the YouTube side of things than we currently are in the open web CTV ecosystem, which I think is a very low bar which we should be able to meet.

[C]

Yeah, but you've done a lot of work with, you've done a lot of work with publishers directly, Right? So maybe talk a little bit about the Spectrum work that you've done.

[D]

Yeah, so I know there's some people from Spectrum in the room, so this was actually really exciting. So Spectrum was the first publisher to roll out 100% transparency in the bid stream on every single impression. So they're passing back full show level data on every single impression. So early on when they started doing that, we put together a case study. So we buy Linear for our clients, so we know what shows work really well on linear. So we built a 50 show inclusion list of the top 50 shows that we know have performed well for this brand and created an inclusion list on CTV and bought just those shows. And that tactic was the best performing tactic on the buy. You beat out all the different adjustable targeting tactics. It beat out the lookalike targeting tactics. It was by far the best performer and it had higher CPM too.

[C]

So better performing, higher cpm, better performance.

[D]

Performance and a higher cpm.

[C]

You weren't cherry picking though one show here or there, right?

[D]

No, because we have a budget to spend, so it's not realistic.

[C]

So is that a reality or a. There's that cherry picking a myth?

[D]

I mean, I think the incentive for media buyers is they got to spend their budget.

[C]

They got to spend their budget to

[D]

make their company money at the end of the day. So they're going to, they're going to pick as much inventory as they need to deliver that budget.

[C]

And like one show would be a direct IO business anyway.

[D]

Yes.

[C]

Yeah, yeah, yeah. All right, so where is this all heading? This is super interesting. So what you're looking at here is global traffic for program level signals. You can say it's in the open market, but it's basically the whole market. The way that we see data is all traffic signals from all DSPs. Maybe this was because of all your efforts in Q4 of asking publishers to work and work on things a little bit better on program level, but we're definitely seeing a lot of improvement. So in October of last year there was definitely a big jump in program level signals and we're continuing to see that. And I think one of the things that's helping is that we're all basically in this together. It's not just a tech company or a publisher or a platform or a buyer. It's like if we want a healthy ecosystem, you kind of need to know what you're buying, when you're buying, and to be able to optimize towards it. An interesting work on program levels for NBA content. This is a really interesting use case because sports is like the de facto CTV environment that everybody is looking for. But at the end of the day, if that's not labeled out into the market, it will look exactly like the UGC content or the fake content or whatever it may be. And that hurts everybody into the ecosystem. You've actually done a lot of.

[D]

Yeah. I mean, my challenge right now, what I'm seeing is there's really not that much sports inventory. Right. Even less livestream. So that live stream content should really command a premium. But it's being bundled right now with a lot of adjacent content like commentary, SportsCenter, your pregame, your postgame, that when that is bundled together, one, you can't really tell how the live stream performs versus everything else.

[C]

Yeah.

[D]

Which will then disincentivize you to pay more for the live stream content. So, like, again, we knock the cheap CPMs and the race to the bottom in the industry, but I think we're kind of from the publisher and the tech side, we're kind of bringing it on ourselves by hiding the content.

[C]

Well, it feels like CTV is a little bit different than the normal race to the bottom and maybe web and app where it's premium content. There's premium content out there and a lot of the content you just can't find and you don't know how to monitor it.

[D]

Well, it's premium content. It's by default, it's not cheap. Right. So it's already at a higher price point than all the other channels for the most part. So the fact that it's so tough to validate the premiumness of that content, I mean, it drives me nuts. Because every sales pitch with a publisher, it's all about the content. Right.

[C]

So what happens? So everything's about the content in the sales pitch. You run the campaign and then after that, and then.

[D]

Yeah. And then when it comes time for the transaction, you have no idea what you're getting.

[C]

And now no way to measure it.

[D]

Well, there's ways to measure it, but not granularly.

[C]

So in your eyes, signal fidelity is kind of a choice right now.

[D]

It's a choice. It's a choice by both the publishers and the media buyers. Right. The media buyers need to understand the value of it and they need to ask and demand of it from the Publishers. Otherwise the publishers are going to hide it as much as they can for various reasons. On the publisher side, I truly do believe that. I think Spectrum has proven it out that when you do share content out, your inventory can command a higher premium because people know what they're getting and then they're able to decision off that accordingly, especially in the performance world.

[C]

So now what? How does everybody participate from here? You want to go first?

[D]

Well, I think the media buyers need to start asking for content data. They just need to keep hitting, beating that drum. I mean, in my combos with the major publishers, they're all doing what they can and pushing on the teams, the powers that be, to try to get it to be, try to share it in some manner or form. I think they realize the value of it. On the other hand, the one big blocker that just continues to be in our way is the vp, the vpa. And that's a law that hasn't been updated since like the early 90s. So I know the IAB, the tech lab, all the industry groups, the VAB, maybe we should all come together and ask for some clarification on the VPPA and get that updated because it's a law that applies to cassette rentals that has been retrofitted for digital. And I think that alone is a big reason why we're being held back right now.

[C]

Yeah, so we had an internal bet as to whether we would finish all of our content in time. The marketing team from my company is not here, but they owe me a lot of money on that. We have two minutes for questions. That's maybe one. While we're doing that, if anybody wants to download the research that was used today, you can take a picture. Ben Lewis, Pontiac Intelligence so you mentioned that different DSPs see the metadata differently. Are some more transparent than others?

[D]

Yeah,

[C]

I mean it's kind of just a fact. Right. So one publisher may. So you have a publisher that sends various signals out with each bid call that may go to one exchange a little bit differently than another exchange. And then where it gets a little bit complicated is on the DSP side, some are taking all the signal and making it actionable, some are not. Kind of depends on know where they are.

[D]

Well, yeah, it's as simple as some DSPs just don't have the reporting fields built into their out of the box reporting to surface those signals. If within the log files they'll exist. But most, a lot of media buyers don't have the capability to pull those log files.

[C]

Yeah, and there's two part and there's two parts of it. There's one part which is like the activation side of like can you action off of it? And then there's the reporting side. And those are most of the times on platforms. Those are two different products.

[D]

I will say Pontiac does a great job at sharing.

[C]

We got one more. You've talked about the problem. Aside from legislation, what can we do about it? Why are you guys on the stage?

[D]

Well, so like Amazon, right? Amazon shares full show level data for Prime Video in Amazon DSP at an aggregate level, right? So the vppa, the problem is when you link the show down to an individual identifier on an impression. But there's nothing wrong with providing aggregate reporting because that can't be. You can't identify people by that. And if Amazon now is doing that in real time within their dsp, then I think everyone can do that because they're the most privacy conscious company of them all.

[C]

Just to add onto that, if you're on the buy side, it's a commitment to using this type of signal in the workflow that you have. Ask questions about who provides the most of it and the reporting piece of it and start to build the practice around it. Because I'd say probably 90% of stuff in CTV outside of Direct buys is some geode audience Z4 scoring on all the different inventory that looks blind and you're just looking at audience scoring. So that's number one. If you're on the pub side, evaluate your partners that are passing in authentication services and metadata companies and get over the myth you're signaling to the market. A signal that people like David multiplied by 10,000 can use for optimizations and to push more spend to your supply.

[D]

And if you do an spo, then you should start looking at how much signal per publisher is coming through their different supply paths. Especially in CTV, most publishers use maybe two to three primary SSPs. So it's really not too much work to look at it. You just got to do the work.

[C]

Amazing.

[D]

Thank you guys.

[C]

Thanks everybody. David Nuremberg, Mario.