Can the Open Web Survive the AI Age? Kurt Donnell on Publisher Economics, Radical Transparency, and pubOS

3 days ago

72 min read

The open web has survived the rise of search, social media, programmatic advertising, privacy regulation, and the walled gardens. But does generative AI represent a fundamentally different kind of threat?

In this episode of Signal & Noise, Rio Longacre sits down one-on-one with Kurt Donnell, President & CEO of Freestar, while Brett House is away for the week.

Kurt brings an unusual perspective to the conversation: he began his career as an M&A attorney, moved into digital media and operations, helped take YogaWorks public, and eventually joined Freestar as its first CEO. Under his leadership, Freestar says it has grown sevenfold in revenue, expanded to roughly 185 employees across 15 countries, and returned more than $1 billion to publisher partners.

Rio and Kurt dig into a question that has become increasingly urgent for digital media: Is the publisher crisis still primarily a monetization problem—or has AI created a structural break in the economics of the open web?

Kurt explains why the historical bargain between publishers and platforms is changing. Search engines once consumed publisher content but sent meaningful traffic back. AI answer engines can increasingly provide the answer without the click. And while the impact isn't uniform—Kurt says larger publishers with strong brands, direct audiences, and genuine utility have generally held up better—SEO-dependent long-tail publishers have been hit particularly hard.

The conversation goes well beyond AI. Rio and Kurt unpack why original journalism remains essential, even in a world of podcasts, Substacks, creators, and AI-generated summaries. Kurt argues that someone still has to produce the underlying reporting—the “raw materials” that everyone else analyzes, summarizes, or synthesizes. Without a sustainable economic model for producing that original information, the entire information ecosystem eventually suffers.

They also examine the economics of programmatic advertising: supply-path duplication, unnecessary intermediaries, declining trust, the legacy “more ads equals more revenue” mentality, and why the open web remains dramatically harder for advertisers to understand than closed platforms. Kurt argues that one of the walled gardens’ biggest advantages isn't necessarily better content—it is simplicity, transparency, and the ability to clearly show advertisers what happens when they put a dollar into the system.

Kurt also explains Freestar’s pubOS, the company’s attempt to create a unified operating layer for publishers. Built around Freestar’s header-bidding technology, yield optimization, analytics and reporting, identity testing, third-party integrations, and managed services, pubOS is designed to reduce the enormous operational complexity publishers face across dozens of technologies and demand partners.

Rio and Kurt also explore:

Why publishers need to become better marketers of their own brands and audiences
Whether subscriptions, commerce, events, licensing, and other revenue streams can really replace advertising
Why email and newsletters may be one of the most valuable tools publishers still control
How first-party data and authenticated audiences could improve publisher economics
Why trust and verification remain obstacles to monetizing publisher signals
Whether cleaner supply paths can rebuild advertiser confidence
Why CPM and fill rate can both be misleading publisher metrics
The role of curation and first-party audience signals
Whether agentic trading could finally remove unnecessary layers from the programmatic supply chain
Why simply replacing DSPs and SSPs with agents still requires recreating much of the infrastructure those platforms provide
What publishers must own directly if they want to remain viable as discovery shifts away from traditional search

On agentic advertising, Kurt is cautiously optimistic. He believes direct communication between scaled buyer and seller systems could improve targeting, packaging, reporting, and transparency—but argues that the market is still very early and substantial infrastructure needs to be rebuilt before agentic trading operates at meaningful scale.

The episode closes with rapid-fire questions on the most misleading publisher metrics, the programmatic practice that should disappear, the most underrated publisher opportunity, what the open web should copy from the walled gardens, and decisions AI should never make autonomously.

This is a wide-ranging conversation about much more than publisher yield. It's about whether independent digital media can build a new economic model around direct relationships, trusted content, cleaner advertising infrastructure, better data, and dramatically less friction—before AI permanently changes how audiences discover information.

Featuring: Kurt Donnell, President & CEO, Freestar

Hosted by: Rio Longacre, Signal & Noise

#SignalAndNoise #KurtDonnell #Freestar #AdTech #Publishing #Publishers #OpenWeb #ProgrammaticAdvertising #Programmatic #DigitalMedia #ArtificialIntelligence #GenerativeAI #AgenticAI #AgenticAdvertising #PublisherMonetization #Media #Advertising #FirstPartyData #RetailMedia #HeaderBidding #SupplyPathOptimization #SPO #DigitalAdvertising #Journalism #MediaEconomics #pubOS

Read the full transcript below.

Rio Longacre (00:01.036)

Hi everybody, this is Rio Longacre with Signal and Noise Podcast, and I am here really by my well not by myself. We do have a an amazing guest I'm going to introduce in a minute, but my co-host Brett House is doing the college tour with his youngest son and is down in Charleston and I was happy to encourage him just to take the take the day off and enjoy his time with his family, eat some good southern food and enjoy his time down in down in the south. but it's great to see everybody. It is a Friday and it's

Kurt Donnell (00:09.091)

That's it.

Rio Longacre (00:30.71)

Another beautiful day here in Denver. And I'm here with a special guest, Kurt Donald, who's a CEO of Freestar, which is a publisher monetization and services company. we're thrilled to have Kurt here. He actually has a law degree from the University of Virginia Law School of Law and began his career. It's funny, we this is the s the second of these conversations in and actually the same in in a couple of weeks of people who went from becoming attorneys to getting into into advertising and ad tech. We actually talked to the

the CEO of Ibada the other day. and yeah, and he he he had a similar path. so not an MA, but he was an attorney at a White Shoe law firm. So so it's it's kind of an interesting story, Curtin. I think that's gonna be really really part of it. But a a little more about your background. You were the after leaving the legal profession, you moved into digital media and you worked at operations at is it pronounced She Media?

Kurt Donnell (01:03.607)

nice.

Kurt Donnell (01:29.836)

Yep, She Media, a women's lifestyle company that owned and operated media properties and represented some others.

Rio Longacre (01:37.187)

Cool. Then you helped take Yoga Works public, and then you joined Freestars as first CEO in 2019. And since then, Freestar has grown its revenue. I'm looking here, says sevenfold, which is which is pretty impressive. And it's it's expanded to roughly 185 people across 15 countries. you have several big acquisitions you've done, and you've helped return more than a billion dollars back to publishers, which is really amazing because I've written so much over the last few years. I've worked with publishers for

almost twenty years, Kurt. It's been a rough couple of decades for them. I think the publishing industry is literally half the size it was twenty five years ago, right when Craigslist first came into existence. It's just been one like one one one battle after another, right? To quip on that that recent movie. But it you know, it's it's it's been really challenging. So yeah, welcome to the show, Kurt. It's great to see you.

Kurt Donnell (02:07.63)

Yeah.

Kurt Donnell (02:27.982)

Thank you so much for having me, man. I appreciate it. It has been an interesting well, couple decades and certainly the last decade and a half I've been doing this, seeing the publisher game change. And obviously I don't think any of us had AI on the bingo card a few years ago, but as is everything in ad tech, we figured out and adapt every single day, week and month and you know, keep fighting the good fight.

Rio Longacre (02:50.69)

Well, it certainly is an industry that is never boring, right? The one constant is change.

Kurt Donnell (02:54.242)

Yeah, that is one thing. I I left for a couple of years. Yeah, I left for a couple of years and kind of swore it off. I'm like, this is stupid. I'm putting ads in the internet. And I realized I missed almost the chaos of it and the fact that it changes every three months and you've got new challenges. Like I've come to realize this about myself. I'm very bad at white collar factory work, whether it's doing like and A deals when you're a lawyer sounds all sexy and great. At the end of the day, particularly as an associate, you're just doing diligence and writing memos and everything and it can be the best deal in the world, but you're kind of just doing the same thing over and over again.

Even in my yoga works career, we were doing very bizarre, but a roll-up of yoga studios, building this big national brand. I don't know, people may be familiar with core power yoga. That's a little more on kind of the heated fitness side of yoga. We were building the more kind of holistic, authentic alternative to that. And even that, we went and closed something like, I don't know, eleven acquisitions in thirteen months, bought twenty some studios, and I'm sitting there, I'm like, Am I gonna go do my next one and the next one? Like I'm bad at white-collar factory work. So the chaos of ad tech, I guess, is good for me.

Rio Longacre (03:56.354)

That's a an amazing amount of acquisitions. It's about one per month. That must have kept you pretty busy as an I and I the MA attorney background really helped, right?

Kurt Donnell (04:04.716)

Yeah, I was running Corp Debas kinda thankfully it switched to well, I guess not to it'd switch over to the business side and part of my day job as a counsel of that company as well. So wearing a couple of different hats there. said it on a couple other podcasts and don't wanna beat people to death with it, but like the law firm was just not my calling in life. I'm so grateful in hindsight that I went and did it, but it's not the sexy thing you see, not that anybody thinks of MA attorneys on TV, but you have this idea that you're structuring these deals and you're working in these like

you know, all the things you're I think you said you're a New Yorker. You know, you think about that New York grinding all night, putting together these deals, like, no, you're proofreading for a living. You're going through diligence and data rooms and summarizing things. It's not like the sexy thing. The business people go make those deals. The lawyers just write it down. And that was the thing that I just had sort of missed in the the prep for the whole career. Nobody tells you what a lawyer really does. They just pat you on the head and they're like, That's great, son. Like go be a lawyer. It's good for you. Yeah. It's a little different than you see on TV or even make it out in like the movies.

Rio Longacre (05:00.747)

Funny.

Yeah, that's funny. That's exactly what Brian Leach from Ibata said. He said that, you know, he didn't realize fully realize like what the work would actually be when looking at what's a billable hour. Well, a lot of it's reviewing contracts, reviewing documents, you know, litigating, doing all sorts of things. And, you know, some people love that. I mean, I I have a couple of friends who like they you like

Kurt Donnell (05:08.419)

Yeah.

Rio Longacre (05:24.809)

You they're they're my age and they've been doing it for twenty-five, thirty years and you know, they they'll keep doing it until they retire, right? But I think for I mean certainly I I think my personality type that would have driven me crazy. And it's funny what you say about ad tech as well. So I got out for a few years too. I w actually went into management consulting after working in media and ad tech for a while and then was in more like just strategy consulting with helping a lot of life sciences companies with everything from you know pro you know, product product launch and you know

fi fixing operating models and working on commercial contracting things. Some of which some of it is related to marketing and advertising, but it but most of it wasn't. So I did that for a number of years and well just like you found myself right back in it. So it is hard to get out.

Kurt Donnell (06:04.43)

I do think like eighty or ninety percent of all businesses are the same, whether you're selling widgets or ad tech or yoga, but the challenges that come at you are different. And is it just like repeating the same thing over and over again and our changes in ad tech are wild every three months? I mean, we may get to it later today, but what was a focus on made for advertising MFA stuff a couple of years ago to duplication and now UGC is the hot thing? Like you're solving a new problem all the time.

While it's frustrating, it does keep you on your toes and engaged and being creative. And that is, I guess, the part of it that I still love despite it being, you know, an interesting marketplace.

Rio Longacre (06:42.119)

So well one thing we wanted to definitely dig into today, if we're going to look at what is the theme for this episode going to be, you know, we've talked a ton about publishers on this podcast. we have fact we had PubEx on a couple weeks ago talking about their publisher monetization stuff. We've had Lupas Kalis on, we've had who talking about you know the challenges of the news media. We've had Jessica Hogue, who's the chief data officer hearst on before. we've had Steph Laser who's gonna talk about publisher ad tag and kind of return like the replatforming publishers going through so we've had a lot of really great discussions.

About this. And I think this one's gonna be no exception, Kurt. So our thesis here would be that the publisher problem has moved just beyond, let's say, ad optimization, right? That like better yield cannot really fully offset, especially now, because you you figuring out how to the yield problem was the last maybe 10 years. Like, okay, this is like how do publishers, how do we set up better yield management platforms to figure out like what's actually going on? And obviously, MFAs there were lot of reasons why yield was going down or a challenge.

But I think with what's going AI right now with the collapse of referral traffic. And I've had publisher clients tell me that they're already seeing 25, 30, 35, 40% reductions in overall traffic. And with that, obviously the monetization of that traffic. So it's not really, it's more than just a yield problem now. It's okay, like how with AI systems consuming content without consistently returning audiences to the places that produce the content, right, in the first place, with this fragment of technology, with like

Maybe some of them are starting to have advertising businesses, but generally these are closed platforms that are keeping people in there. What's the answer? And and I I think you you're probably gonna answer the answer is PubOS, which is Freestar's unified operating system for publishers, which for the AI age, which I'm excited to learn a bit more about here, beyond what what you have publicly available on the websites. maybe maybe we start there. Let's first question: Are we fundamentally looking at a different cycle for digital publishing?

What we experienced like the last, let's say, 10 years with the focus on yields. Is this a structural break in the economic model? Thoughts on that.

Kurt Donnell (08:41.934)

I think there's been yeah, there has been a structural break, whether it was stated or not. there was a quid pro quo in the past. Google gets to scrape your stuff, they send you back users, they'll even help you monetize those. And so the deal worked, right? Google's never owed anybody a business model, despite what is thought. And so I think that is the change right now is publishers didn't have to necessarily think of themselves as marketers, which is interesting because they sell advertising. They didn't necessarily think of themselves as marketers. The smart ones did and built newsletters and a real relationship.

Some of them were able to translate that into events or e-commerce, but a lot of them just sort of published their content, played the SEO games, got something back. So I do think there's obviously been a fundamental shift in that. And you're seeing that in who's getting traffic. you look at the smaller sites in the internet, the blogs and everything. fortunately, I think blogging is back to being more of a hobby than a business at this point in time. You're just not getting the referral traffic.

And that's because there's certain things that AI is great at. It's great at planning a five day Disney adventure. Like what's the best thing, you know, what are the best things to do in five days with kids that are five, seven, and nine? Like it spits out a pretty darn good answer there and you don't necessarily need to go

Rio Longacre (09:48.954)

And by the way, Kurt, sorry to interrupt for a second. I you know, ha being someone who runs a blog, totally agree. I mean, you're not getting much referral traffic. You have to generate your own it it's your own traffic. You it's a great place to put out content, but it's it is a tough place if you're if you're looking to get traffic back from the Walt Gardens. So yeah, I totally agree and keep going.

Kurt Donnell (10:05.538)

Yeah, I mean, I think it it's important and we need the content creation. I'll certainly get we'll talk more about that today, but that in and of itself is tough to make the cornerstone of your business if it is a publishing business strictly doing that that doesn't have a very specific utility for somebody. we've really carved out our niche with mid-sized enterprise publishers, the Reuters and Associated Press and Fortunes and Boston Globes and Smithsonian UGCs or excuse me, UFCs of the world that have a very specific brand and people go there.

For a specific thing, whether that's breaking news, or even we have sites where there's sort of scheduling sites, if you will. This is where I go schedule my potlucks for my kids. This is the place I play casual games. There's a reason that people go there. They have a relationship with that site for one reason or the other. Like car reviews, I go to a very specific site, just whenever I'm looking for a new car, I tend to go there. Hopefully there'll be a client soon. So I'm not gonna say the name yet and jinx anything, but you know, that is what I do because that's what I've been doing for 20 years.

And those brands built an affinity one way or the other through trust, I think is a piece of it. Through giving me utility, there's other sites that their newsletters are so valuable to me every day. They're giving me insights that I'm going to go back and do that. And so I think it really comes down to a lot of way the model shifted that you have to be a marketer yourself if you're a content creator, because the business model's just not going to come to you. Facebook and Twitter killed referrals, what was that, two, three years ago? And Google's done the same now.

Rio Longacre (11:28.181)

Yep. Yeah, they've slowly like started to like well send I mean, I guess this goes back maybe over ten years where like they were there there was a deal they'd send you some referral traffic, but it was always debatable, okay, how much are looking at the value of the content and how their how Facebook and you know, Meta and and Google are monetizing that, there's always some argument about okay, like the tr referral traffic they're giving back, how much value are they safetying off? Are they commensurate? I I I I had publisher clients who would say

you know, we did some studies where it was like it was in the billions of dollars a year and the the delta between how they're monetizing and what they're sending back. So it was always an unequal relationship with the Walled Gardens, but I do agree with you. There always was a quid pro quo. We're sending you something back and and that's clearly ended. And I imagine for those for the aggregators, God, it's gotta be I mean I mean, if if normal news publishers have seen twenty five to forty percent reductions, which is things anecdotally I've been hearing, for some of these so some of these sites are aggregators, there must be more than that. It must be catastrophic, I'm guessing, right?

Kurt Donnell (12:03.591)

yeah.

Kurt Donnell (12:28.738)

I think so. We I mean obviously I could see we represent about a thousand sites, you know, four or five hundred publishers themselves. It varies. Different categories have gotten hit much harder and there's still cyclicality. We work with a lot of news sites. I'm a big fan of journalism and I think it's a really cornerstone to a democracy and having, you know, unique thought and not the echo chambers that you do get in the social media platforms. So really proud of what we do in the news space, particularly, but that's cyclical. Like this March, obviously we had a war pop up, like things are gonna be up and down.

Rio Longacre (12:39.328)

Okay.

Kurt Donnell (12:57.912)

So it's sometimes tough to track exactly like what news sites are doing. Is that relation related? Is that breaking news? Whatever. But when you look at other sites, the ones that always had a good direct population going there, they had a lot of direct traffic, those have held up pretty well. Search has, of course, gotten decimated across the board. There are a couple of things that have popped up and actually done a little bit better. I think we've all seen the proliferation of Reddit being the top thing, which is

Sort of antithetical to this whole we don't want UGC thing that a lot of the buy side's doing, but then Reddit's like the number one thing. I don't know. Very random, probably conversation for another day there, but definitely a lot of shifts in the marketplace. I do think it's gonna be very tough. And listen, if you're sitting there, you're gonna go start a new digital media publication right now, probably not. That's tough to do. So figuring out different business models that have multifacets to them, I think is really important for folks. And it's not necessarily

subscriptions. I think people have been subscriptioned out a little bit. And so making sure you're building that relationship with the publisher by giving or excuse me, the audience by giving value. I really think email is a key part of that is important.

Rio Longacre (14:05.48)

So what if you're able to give us any numbers related to the decline referral traffic, I'm I don't know if you're able to because it could be client conf confidentiality, but what are you seeing with some of your clients? Is it is it what I'm seeing twenty five, thirty, forty percent? Is it more than that? Is it less than that? Is it is it impossible to have let's say an average across? What are you seeing?

Kurt Donnell (14:27.936)

The average across isn't possible. It's so segmented. What I will say, we did some cohort analysis from our sites. And like I said, we tend to work with larger, more utility and nature sites. And utility does take a lot of different flavors there. Those sites actually held up reasonably well. And we did a study from Q3 of excuse me, Q4 of 2023 to the end of last year. What did we see? There was a massive dip in May of 2024 when all the AI stuff happened and it was rocky.

But on the whole, that group of sites was actually up four or five percent in traffic by the end of last year. It's eroded a little bit more this year, but it hasn't been as catastrophic in certain pieces of the industry as you hear about. What we talked about earlier, the bottom part and the smaller sites have gotten nuked because they were all SEO traffic to begin with, whether it was travel or cooking or some of those things. So on the whole, I think the numbers you're referencing are probably the case, but there's certain segments that have been hit as hard. I would say the larger sites have been more sustainable.

And that's tended to be our clients. So we're a little bit of an outlier in that regard.

Rio Longacre (15:33.098)

So the long tail sites have been punished, which makes sense. It was all all mostly SEO traffic before. Yeah. Okay.

Kurt Donnell (15:37.465)

Just unfortunately, I mean, they were so heavily reliant on social. I mean, years and years ago, back in my She Media days, we almost bought a company that was almost all social related. And I remember talking to their CEO, like, we've got the greatest relationship with Facebook. They would never do anything. They would we, you know, we work with them all the time. And I was talking to my CEO. I'm like, this isn't good, man. They built their business on somebody else's platform. It can go away. And no shit. Three months later, Facebook just one day just cut them off and they were out of business and had to fire a hundred plus people.

Three months after that.

Rio Longacre (16:08.138)

Yeah. Yeah. Yeah, I worked with a data company that I I won't name them now, but same thing happened. Like mo a lot of their a lot of their models are based on face on basically i signals they were able to get from Facebook and then Facebook facing all those privacy problems, locked everything down. I completely eviscerated them. You went they went through pretty big layoffs. I mean that I guess that's the the the lesson learned is don't dep don't don't base your business or one company or one partner, right? Or one vendor. So it can be

Kurt Donnell (16:34.966)

Yeah, I mean don't build your don't build your business on somebody else's platform if you can avoid it, man. That's always a tough one and you see it even some like social media influencers or something where all of a sudden they get shadow banned or whatever, like they've just lost their livelihood. You gotta have your eggs in a few more baskets if you can.

Rio Longacre (16:37.322)

Be challenging.

Rio Longacre (16:52.256)

Yeah, it makes sense. So let me let me ask you a question, maybe play devil's advocate for a second before we kind of dive into how how like I'd like to definitely talk a little bit about your background too. before we dive into that, like looking at the topic itself, there's and I've heard this a lot, like I said, we had Lupus Gallas on a couple of times. We've talked in depth about the importance of the news media. And in fact, we had we had Vanessa Otero from Ad Fontus. We talked about the the media bias chart, which I think is pretty cool.

Kurt Donnell (17:15.512)

Yeah. They're great.

Rio Longacre (17:18.858)

But I guess the yeah, she she was fantastic. That was one of our our best rated episodes, actually, just to shows what a what a what a great reputation she has. But looking at the I just yeah. she's fan yeah, she's fantastic. I mean, I I'm not yeah, I know her pretty well. I I can you not not to say we're we're we're best friends, but I do know her we w she's also based in Denver, which is pretty cool. In fact, she and her she and her wife have a Brazilian restaurant that's awesome that I've been to a c a couple of times. It's really, really all it's really awesome food, yeah.

Kurt Donnell (17:26.734)

She's an awesome person. If you get the chance to hang out with her, she's awesome too, for for what it's worth.

Kurt Donnell (17:43.81)

Really? I did not know that.

Rio Longacre (17:47.347)

It is worth checking out. She's a yeah, I agree. She's a really good person. But but looking at the news media itself, I mean, I like I believe we need a we need a strong news media and a and a healthy press. And I, you know, Sarah's the important part. I'm not gonna, but just just to play devil's advocate, with the rise of new media and let's say the rise in distrust in the in the mainstream news media, like.

Yeah, sure. We've seen the news media get pummeled, right? I mean, they've really suffered because of a lot of these problems, and they continue to, right? And you know, it's not worked out. Like like what what is their pl you know, they people are now going more to I mean, I get a more of my news from different substacks that I I follow or subscribe to, from blogs. I listen to probably ten, twelve hours of podcasts. Granted I'm like I like podcasts, obviously I run one, but it's how like

Is it as important with the fracturing of media, the the many different channels, the growing distrust of mainstream media? I mean, I don't I I read stuff, I take everything with a grain of salt now, even if I had to read New York Times, which is obviously a very, you know, it's a very reliable source of information. I mean, I think a lot of their headlines and even the choices of what they cover tend to, you know, reflect a certain point of view. So do we really need how like how would you argue

Why would you argue we actually need to worry about the mainstream news media when clearly the like the the new media alternative media seems to be kind of filling in the gaps? Thoughts on that?

Kurt Donnell (19:10.626)

I think a lot of the new media stuff is regurgitating or taking their spin on what you read in real news media. Like the new media stuff is not there on the ground covering wars wherever they may be, or the international events and they're not in the rooms. They're not getting the leaks or whatever or breaking the news that Trump was smuggled out in a catering cart. Like there's a lot of things that synthesize other data. Like unless we're going to get it is real. I mean, definitely real. Like ridiculous. Also very stupid to like publicize this because then you can never do it again.

Rio Longacre (19:32.256)

That's a true story apparently too, isn't it?

That's wild.

Kurt Donnell (19:40.406)

Right? Like I don't know. John Oliver made that point the other night, which is pretty pretty salient. Like you probably shouldn't give away all your tricks. But anyways, I think there's an element of

Rio Longacre (19:42.346)

Yeah, one and done.

Rio Longacre (19:50.88)

Well someone must have leaked that though, 'cause it's so absurd, right? Someone so somebody admin is probably I can't believe this guy did this and it's like this like I and call c called us, Okay, this is off the record, but this this this jerk was was in a catering card or whatever. So it's funny.

Kurt Donnell (20:01.686)

Yeah. And and by the way, left like fifty or a hundred people on the plane that was targeted to be blown up, right? Like it's ridiculous.

Rio Longacre (20:11.175)

Yeah, yeah, he sh he showed who he cared about, right?

Kurt Donnell (20:13.726)

Yeah, it's terrible. But an aside there, Yeah, the new media versus the original. So I it I think it ties into a conversation we'll surely have about AI at some point. It wouldn't be a podcast that we talked about it. Like at some point we have to have the raw materials for anything. And I think the people, the real journalists that are doing the boot boots on the ground, that are funding this, making the things happen is the raw materials we need for.

Rio Longacre (20:16.423)

That's crazy, but

But looking but looking at like the v the r the Yeah, it keep going.

Kurt Donnell (20:39.832)

synthesis, whether it's on podcasts or on other new media things or your substacks, they're typically discussing something that a let me like proper big J journalism company wrote, reported on, did something that got it out there. And then there's a a lot of discourse around that and analysis and all of those things. So I think you need that. That's the same thing for AI too. If we stop publishing content on the web, AI becomes pretty freaking useless. Like it needs the raw materials here.

Rio Longacre (20:54.245)

Usually they break the story, yeah.

Kurt Donnell (21:09.534)

We've got the the incentives are still sort of screwed up because they're still scraping for free and not paying. I don't see that changing imminently, but at some point it has to, because people quit producing the raw materials, right? And then the whole supply chain breaks down. So when it's the new media or discussions or substacks or whatever that may be, you need the original source to even have anything to talk about. Same thing for AI. AI is not going to be out there. I'll keep using covering wars because it's a pretty extreme example, but like AI is not going to be out in the field reporting on things. So I think

That is a piece that maybe gets lost of what the mainstream media is doing in many ways, or at least the decent ones, while they may have a bit of a bias, they are bringing us things that are unique and different perspectives, but very real of what's happening in the world. And if you don't have that, you know, where do you have to go from? AI can't cover the Olympics. You need the people on the ground doing those things. So there is a place for it forever. And we all can pick and choose what we like. I think you are probably in the top.

X percentile of people that are news gatherers and you probably go diff get different sources and opinions. You know, many people won't, but I think it's important that we have the opportunity to do that and fund that and not have it s censored, like where it's demonetized because it has a word in there. Like the Celtics blew a lead last night, or like the Celtics blew up last night. Should we be blocking that? No, like context matters here too. So I just really advocate for let's keep a feasible ad supported media.

Rio Longacre (22:16.349)

Yeah, that's a really good point.

Kurt Donnell (22:35.782)

news media portion of the landscape viable. Like it we need this as a society.

Rio Longacre (22:41.193)

Yeah. Yeah. But it's a really good point. And I think for the listeners who don't realize even just looking at New York Times, who I just mentioned before, they have I'm looking here, seventy five foreign correspondents worldwide. It's not a huge number, but that's a good number, right? The fact that they have seventy five people on the ground in these countries who know who know who have contacts locally, who are gathering information. You can't just do that sitting in like you know, sitting in like a building in New York City, right? You need people on the ground. I mean, so the value of that's expensive running you know.

Paying all those salaries and flights and everything else, right? So that and that costs a lot of money to do it. And now you could maybe argue that you know the way that they're assigning to far s Farid correspondents could be could be better. I think, like, for example, they have more of them in in Israel than they have in all of in all of Africa, which is kind of which is kind of outrageous, I think, which is probably why they publish so much news on this one on one on one little place. But the reality is they do have 75 people.

Stationed different in different locations, which is hugely valuable in terms of understanding. So and that costs money. So I think that's a really good point about the value of them doing the actual research, doing investigative journalism and reporting and actually breaking these stories. Because you're right. I mean, most of us in the in the potosphere, hey yo, ha have I broken a couple stories since Sigland Noise has been been in existence? Sure, there's a couple, you know, but I wouldn't say they've been earth changing. But in ad tech, I think they've been kind of cool. But but the reality is most of the news is coming from mainstream news media.

Who still have the apparatuses in order to do it. That's a really good point.

Kurt Donnell (24:08.238)

Something people don't think about, like just even the internet and ads and everything. Like you have all of the world's information available for free. I think we've done a very bad job of explaining that to users. Like, this is the value exchange here, guys. You see some ads or you click on our affiliate links or whatever that is. I give you the world's information for free. We've just done a shit job. Like you wouldn't expect the newspaper to just show up if you didn't pay for it, or Netflix to turn on if you didn't pay for it or anything. But people just expect the internet to be free. Like it's a weird.

Rio Longacre (24:09.406)

Now Kurt.

Kurt Donnell (24:36.386)

dynamic that we somehow missed telling that story on very well.

Rio Longacre (24:40.221)

Well yes, interesting too, 'cause your phone your phone's not free, your electricity's not free, right? I mean, obviously you pay for cable, but you know, and generally speaking, people believe the internet, you know, is free because the government funded it, right? And then advertising supports has always supported most of the content. I think people who don't know the ad business or the publishing business really don't realize that. They think all this stuff should be free. And before the internet we you had you had to pay to get a paper delivered to your house every day, right? Whether it's the Times or Wall Street Journal or the local paper, right?

And a lot of those have obviously gone away. So no, it is so so I y I think you're right. That's a perception that is hard to dispel, but it's true.

Kurt Donnell (25:16.364)

And people will make that exchange. It's just weird in the open web, particularly. People make that exchange all the time. They'll do ad supported Netflix or ad supported I don't know, Paramount Plus or ad whatever they are. And they make that conscious decision. It's just not they don't think about it. They're just like, I hate ads on the internet. It's a weird dynamic. Anyways.

Rio Longacre (25:17.598)

So Kurt

Rio Longacre (25:36.32)

Yeah, well, people like to complain. But Kurt, looking at your background, you began an and A law, you moved in digital media, and you helped take yoga works public. What part of that, I mean, I can think for ad tech is an unconventional path, right? But granted, people in ad tech, it's so funny. Like like we've had pods where we realized, well, everyone in this episode is from had a liberal arts degree or like has a non technical degree. But then we've had like some very technical people too. But I think but I what I love about advertising in ad tech is you do get people from all over. But I would say it's kind of an unconventional path going from attorney into this. Like

What about that background kinda help you really h helped you as you as you took on this role? 'Cause it sounds like it was a it was very important if I'm not yeah, if I'm reading between the lines here.

Kurt Donnell (26:16.974)

Yeah, I think it is the opportunity to see so many things. Whenever you're in professional services, you're in consulting, lawyers, accountants, you get the opportunity to see so many different business models and you get to meet so many different operators and the way that they did things, particularly MA law, you're doing a lot of diligence and like, ooh, they made that mistake, I would never do that. Or I really like the way they did that. It gives you perspective on how different people do things in this sort of MBA in different business models in a way that's tough to get. Like you don't have the chance to go work in

200 different businesses, but I had to go figure out how to understand the due diligence on 200 different businesses. And so that was a real gift that that gave me that I think you can get in many professional services, but if you kind of don't know what you're gonna you want to do or be when you grow up, I would actually recommend professional services a good place to start. You get a foundation of whether it's accounting, finance, management, consulting, lawyer, those are very useful skills that you can put to work and always find a job. But you also get just get to see a bunch of things and meet some interesting people.

And frankly, see the pitfalls that you want to avoid going forward. So that was one of the biggest things I would say being a lawyer was valuable with. The flip side of that is when you're negotiating something from more of a commercial side, I could always be using, you know, have the legal hat on one side, business hat on the other side, and kind of be negotiating them in parallel, not in like a bad way, but I could be thinking about like I want to make I don't want to step on this landmine. So let's talk about it this way and then this, and you can end up in a better place together.

It certainly has helped me when I'm doing MA deals where I can do the deal myself to some extent, but also then hold counter like counsel accountable or say, All right, guys, that's theoretical. That's never gonna happen. We're doing this. I make a business call on it, but it's informed with a legal background to it. So like anything in life, it's kind of breadth. We're all just a collection of our past experiences. And I got to have a a lot of past experiences in those, you know, seven or eight formative years in a law firm.

Rio Longacre (28:12.317)

It's interesting you say that. So I'm actually married to an attorney who does not practice the law. I mean, you know, so my wife, she's been in-house at a couple places. She worked at a firm for a while, but most of her career, she whether she's she's had business development roles, she's had like more corporate some c corporate communications roles. She had a couple places where she was doing more like I think government and community affairs or relations. And she said every single one of those roles, being an attorney or having a legal background was incredibly useful. In fact, a couple of them hired her specifically because of that.

Right, but she does not miss practicing law, but she actually has said she couldn't have had this career without having a legal background. So I think I think it's interesting and she's told me I think it's a some credible number of attorneys who passed the bar, who've obviously went to law school, and are still have kept their licenses valid, actually don't practice and I think it's almost fifty percent if I'm not mistaken.

Kurt Donnell (29:02.83)

It's I've I've talked a lot of people out of going to law school, but I what I would say is you should go for the right reason. If you want to go to law school, I really recommend people go be a paralegal to understand what a lawyer does. The flip side of that is law school, while it's not exactly a trade school, it does teach you how to think, sort of that Socratic method and the analysis of problems and everything. And not that you can't get it elsewhere, but it is a very systemic way to learn how to think. And that is very, very useful.

What I will say, everybody's like, if you get a law degree, you can go do anything. That is true after you get your first non law lawyer job. The first one's really hard to get. After that, everybody's like, my god, it's such a benefit. But why would they hire a lawyer to be a finance guy when they can just hire finance guy? Like

Rio Longacre (29:42.111)

Okay, right. So okay, so so so once you're pigeonhol so once you're pigeonholed as a lawyer, getting your first non non laugher job is hard, but once you've done that, you then have a try Okay, that's interesting. Makes sense. Yeah. Well I think you could say the same for

Kurt Donnell (29:51.532)

And it's like, my God, you're a lawyer too. That's the greatest thing ever. Like, okay.

Rio Longacre (29:56.575)

Well I think you can stay the same for any career, right? If you've been doing something long enough, right, it's it's it's it's hot difficult to switch. But once you switch, you realize, okay, like to your point earlier, many of the same business issues come up, regardless of whether you're consulting or an agency or work at a tech firm, like you're seeing a lot of the same business things, but having that that that knowledge from your from your background being able to apply to these different situations, they kind of like they don't repeat themselves, but they rhyme, right? So I think so I think that's super helpful.

Kurt Donnell (30:21.368)

Yeah. I think interesting another as I think about another one is if you don't know what to do, go be a salesperson. Selling is freaking hard, but once you've done sales and understood whether it's onboarding or anything like that, that can make you a great ops person or even a finance person because you've seen the front end of the whole thing and then how do you make the back end work better? That's another one that I'd probably say. Either go selling or go try to consult or not consult, but professional services. Those would be if you're lost in listening to this coming out of college one way or the other.

Go do one of those things.

Rio Longacre (30:53.726)

Yeah. I mean, I think plus with sales, if you are a good seller, you don't have to be the best seller in the world, but if you are a decent seller, you can always find a place to go. I mean, revenue generating roles are are the first to hire, the last to be less to be let go. I mean, a company's can't run without revenue. So if you can bring in revenue, right, in some way, if you have a skill to do that and you know how, it's it's I think it's a lot easier for those people to consistently find work. So I think that's a really good point.

Kurt Donnell (31:02.016)

Always.

Kurt Donnell (31:07.278)

Yeah.

Rio Longacre (31:20.124)

but looking back at Freestar, so it's growing from twenty I'm reading here twenty five to hundred and eighty-five employees, sevenfold revenue growth. what do you think like ha was there an inflection point? Has it been steady growth? How would you describe it?

Kurt Donnell (31:34.895)

It's been really steady over the years. There's obviously been bumps in the road, you know, periods of super hyper growth, periods where we have to rethink things a little bit, but it's been a pretty steady run for the last. This is my eighth year doing this since we got here. It's been a lot of fun, a lot of work. And as we talked about earlier, no lack of curveballs thrown at us by the industry. But I think that's been a part of why we've been able to do this. We've got a lot of resilience built into our culture. We're very willing to change.

We really try not to get stuck in our old ways and with that have stayed adaptable and built a workforce that's very unique. We've got people in fifteen different countries around the world. we were a hundred percent or not hundred percent remote, probably seventy percent remote going into COVID. So the move to a hundred percent remote was super easy for us. It also made us very scalable, both from a time zone coverage standpoint, as you can imagine, running a business.

Different geos have different cost structures. So you can be very efficient from that standpoint, being more international as well. And so that's been, I would say, a unique benefit of what we have done that made you know COVID easier for us and then scaling long term easier as well. yeah, unique, unique business culture is something that we really, really value. And I think being remote, you have to be hyper, hyper focused on that 'cause you don't have the water cooler moments or the in person all hands. So we've really leaned in on that to help continue growing.

Rio Longacre (33:00.85)

Were you always focused on publishers or is that more recent? was that the initial business plan?

Kurt Donnell (33:08.036)

there was a pivot before I got here. So I was a hired gun CEO, came in a couple years into the journey. Company's 11 years old now. I came in at year two and a half, something like that. So it's always been the intent was publisher focused. So the two co-founders had kind of found their way into a website that had millions of visitors, had no idea what to do on monetization, overpromised, under delivered by all sorts of folks in the space that could help them, and went just sort of figured it out, realized they were pretty good at the monetization side of things.

Found what we know is header bidding now, quite ubiquitous. By then it was brand new. Found that early and started playing around, figured it out. Like, hey, we're on to something, grew the one website, went out and raised a very small friends and family round from like our founder's literal mom and sister are still on the cap table. to go buy some websites, juice the revenue, flip them like real estate. That was the original model here. It was supposed to be like a fix and flip situation. About a year into that deal, realized they were awful at content and running the editorial sites.

Still pretty decent on the monetization side. So pivoted to what is our current business model now. our company's name and its our number one core value is we're Publisher First Inc. Publisher First is our number one core value in really what we do. We have seen over the years so many times the publishers get taken advantage of. Advertisers are always sort of king of the hill here. And so we wanted to not in sort of a support the underdog way, but like let's represent a piece of the industry that is crucial to the whole thing. It is the full supply side and give

even mid sized publishers, the operate oper operations of an enterprise level publisher. So it has been publisher focused from the jump.

Rio Longacre (34:44.242)

Looking at the root causes of some of the problems publishers are experiencing now and have been experiencing for the last, you know, let's say the last few years, I mean, I'm gonna list out a few here. Could like weaker ad demand, declining referral and search traffic, platform dependence, you know, on you know, Google Meta, et cetera, rising tech costs, or let's say lack of technical expertise, audience fragmentation that I mentioned earlier. Like, which ones do you think are the most important or of those? Have I missed any? Like

What are the things that you're seeing w what what are what are the root causes? I guess let's start there.

Kurt Donnell (35:19.372)

I think you hit on most of them. Obviously AI's been the biggest curveball on everything we talked about earlier from a search standpoint. That has just fundamentally changed. Discovery in general has changed over the last three years. Facebook and Twitter, as we talked about earlier, you know, they kind of pulled the plug on sending things off site, kinda after Elon bought Twitter, like their X, you know, that fundamentally changed. Facebook really started keeping things inside of its own walls. So we felt, you know, a bit of a cliff. Not cliff, but I mean something we had to deal with then.

Obviously AI took that to another level. So that's fundamental to the whole thing. When you sell, you know, spots and dots to eyeballs, you need the traffic to be there. So that's certainly that element of things. I think from the demand side, we've seen hiccups. Obviously, when COVID happened, demand fell off instantly. CPMs were down 50%, if not more, overnight. Rebounded, you know, different advertisers came into the market, but then we saw some shifts in traffic after that.

where people were going back outside and then you know dollars reallocated. I think the demand side of the business is interesting. There's different platforms, CTV obviously rising significantly, grabbing a chunk of the dollars. I often say, you know, it's not really the demand and the supply side of this industry they're fighting each other. It's the open internet versus the walled gardens. Walled gardens take increasingly large shares of dollars. And so we need to really work hard to keep the open internet as a very viable, valuable part of any media buy. But

You know, fighting those dollars is a piece of the whole thing, of course. And then just the operational elements of things. there's so many different players in the space from identity and privacy to different bidders, different technology, client side, server side, this, that, and the other thing. It is tough. That's where a freestar and we've got some good competitors in space, but that is sort of our Pub OS model. As you plug in this one thing, we will take care of all the complexity of 50 different

Vendors, the procurement, the legal side of things, the collection of payments, the discrepancies, yada yada yada. I can go on for for days on all the things they have to deal with, like really simplifying that for a publisher that's probably trying to do more with the same or the same with less on their side from a headcount standpoint. Let's be the easy button. So that's where we kind of come in here. And I think all the things you said are real challenges for a publisher. Not necessarily the death knell in any one of them, but it does feel like you're kind of getting poked all day long.

Rio Longacre (37:40.958)

So I'd love to unpack like exactly what PubOS does and how it how it handles all those specific activities you just mentioned. But before we do that, maybe like I mean, having been a consultant working publisher for many years, I've had so many meetings with publishers where there's discussions about building, you know, whether it's like social commerce putting commerce to social commerce in or publishers hold ho holding events or

You mentioned email, having newsletters or video or licensing and you know, maybe like licensing of data companies. There's been so many talks about publishers diversifying their revenue streams, building these businesses. How many of these are genuine businesses? Or is that the wrong question? Is it okay that's more like looking at things holistically about how like the the s like the the total the sum of the parts being more you know, the the the sum being more than each in than the each individual part put together, right? So is there a

Like how would you look at it? Like how would like when you're talking to publishers, what would be your advice to them?

Kurt Donnell (38:41.558)

I think it really depends on what your sort of affinity group looks like. I think different publishers can do different things. There's very, very few publishers that can demand a subscription. It was, you know, supposed to be the savior of all this. And we're gonna have micropayments and this, that, and the other thing. Sure, you read all about that. Probably talked to some of your clients about it, what, two, three, four years ago? How many of them are actually still doing it right now? Relatively few. As I mentioned earlier, I think people got sort of subscription to out, and there's not been

Rio Longacre (39:06.725)

Micropayments, I don't know. I don't know if anyone's actually doing it now. Yeah.

Kurt Donnell (39:10.51)

Yeah, and there's not been a really elegant solution to do that, in all honesty. We're working on a couple of things, but it's more like gated content than micro payments. People will pay with their eyeballs and their time probably a little more than they will open their wallet, unless there was truly something I'll pay a dime to read this site, you know, read this page. It just that hasn't happened yet. So subscriptions are tough outside of that. I think brands that have built an affinity around a specific topic or something, they have the opportunity to do an event strategy.

I think there's things you look at review sites or something like that, obviously. Really nice tie-in there to affiliate commerce opportunities there. So I think it really depends on what your site is, what your audience is, and what is the affinity and like what is one swim lane over. I've always said in our own business and other things, like go be the best at the thing you can be the best at. And then only once you've done that, move one swim lane over. The people try to boil the ocean and do everything, like it's just never gonna work very well because

If it's like a feature for you, but it's somebody else's whole company, they're always gonna do a better job if it's their whole company. Like they just will spend more time, energy, and effort to get it right than you will if it's too far afield. So if you've got a loyal community around, I don't know, some interest group, you're probably ripe to do events. Like start doing local events there. You can build those things up. Or if it is, like I said, a review site or something like that, commerce makes sense for you. A few can pull off subscriptions, but it's gotta be pretty darn good content.

and very unique and differentiated.

Rio Longacre (40:41.927)

Yeah. It's interesting what you said about micropayments. I remember when I heard about the concept, it was probably five or six years ago. I got incr I mean, I I was always a bear on crypto. I always thought it was just a ridiculous idea. It provided like it was like the proverbial like a consultant with a hammer running around looking for nails. I certainly thought NFTs were the most idiotic thing I'd ever heard of in my life. And I was when the like the board eight yacht club things crashed down to the basically they're the word. But I knew people who were sh who were pump shilling those and excited and said it was gonna be the next best best thing and I just thought it was

Kurt Donnell (41:02.467)

You and me both.

Rio Longacre (41:11.877)

idiotic, but clever on the same page there. But when I heard of like micropayments is a really cool use case for publishers, but I think the problem is that how many people actually have like a wallet installed in their browser and actually have like actually you know, in ac can actually transact with micropayments, right? Even if they and granted a stable coin is actually for those who don't understand a stablecoin is actually pegged to the dollar and they ha they actually have like some kind of f assets that are equivalent to the stable coins they're issuing. So like within the world of crypto

I I can see stable coins having some potential value 'cause they're pegged they're pegged to an actual currency. And for publishers, if you get people to do that, wow, that'd be incredible. But like no one's doing it.

Kurt Donnell (41:52.963)

I mean, how many people really have crypto that are gonna go read the New York Times to keep world picking on them? My mom doesn't, my dad doesn't, whatever that is. You'd have to have something that's so ubiquitous from a mutual login, shared infrastructure, micro payments. You need something like like ASCAP almost, which is for those that aren't familiar, it's like the music licensing thing. It's a hundred years old. It's how royalties get paid. So if you have X number of streams on Spotify, it all gets paid in there. They distribute that. Spotify probably helps there, but

across all these different things. It ties into micro payments, but also AI licensing. Until you have sort of a trade organization that's neutral, that everybody can buy into, not buy into, but you know, shake hands, sign the paperwork. They will handle the accounting on all this stuff. It's tough to do. Like a micro payment to five different websites doesn't matter. It has to be something that I'm like, okay, pay a dime here, dime here, dime here, but it's completely unrelated ownership. It's just tough to do. As we continue to scale and

You know, maybe we can put together sort of an industry coalition between ourselves, some of our competitors and things to bring this to life, but the infrastructure is way harder. The concept sounds great. I think people might do it if it was genuinely that easy of like unlock this for a dime. Okay. But it just is harder to actually bring to practice than it appears.

Rio Longacre (43:11.005)

Another thing you mentioned was the declining CPMs. You mentioned it going up and down a little bit. It's I you know I we talked to Permitive here, I think it was earlier in the year, and they had put out the study showing that open exchange CPMs for publish for me publishers were under a dollar, which is just absolutely bananas, right? And when you look at the at the tech tax that comes out from all the ad tech intermediaries, so if a CPM is under a dollar and let's say the publishers are only getting thirty nine cents on the dollar anyway, I mean it's just an incredibly

small amount of money they're getting from fr from that. I mean, so I guess the question I have related to that would be if if that's su been such a bad business, let's say dis display, putting banners, like using programmatic and exchanges in order to serve up ads has been such a cruddy business for publishers, which I think generally it has been. Why have they not gotten better? Like why did they take this path? Why, you know, let's say fifteen, twenty years ago, they

Their whole businesses were about selling, you know, belly bands and space ads and different inventory and like package deals, right? That were that they that were great for advertisers. They could sell them for a lot of money, right? And I get it, like the print business started dying. But it seems like that expertise never translated over to the digital side and and like they relied predominantly on these automation and exchanges. Was it because of lack of digital skills? Was it because of it was easy money? Was it just

Was it just the game is rigged against them by by ad tech and by the Walt Garden's thoughts?

Kurt Donnell (44:44.214)

A, I don't think CPMs are quite as bad as you said. It's not thirty percent of a dollar. I mean, there are certainly many dollar CPMs, but that's after everybody's taken their fees. So there's that. I think being more thoughtful about formats and there was a more is more theory for a while where I need to make more money, I'm gonna put more ads on there. You get ads to content ratios that look r ridiculous. Therefore you devalued the inventory because just the sheer number of ads, you lose attention, all of those things. So it was this sort of bad self-fulfilling prophecy.

In the best way, I think there's been kind of a flight to quality movement in the last year, year and a half. Folks like Jounce Media talking more about that, Kamara and everybody making more noise about like what is an appropriate ad layout for this. And we've seen that where you take away two smaller ad units, you put in one video unit, for instance, and you've actually reduced the number of ads on the page, but higher RPMs are driven by video versus display. So I think it's kind of a balance. I think the pendulum swung.

too far where it was spam. I think the MFA sites frankly hurt quite a bit where there was all of these sites of slideshows and a bunch of crap that no human really meant to be on, but you'd end up there one way or the other, you'd click and like that drove down the value of that inventory, which then spread a little bit to some other things that really were quality and driving the outcomes. but there were and are a lot of hops. I think the supply path has gotten much, much cleaner at this point in time. I think we're moving toward a world where that becomes more important.

Less duplications, certainly less MFA, less fraud, all of those things. And I've said this publicly, written about it on Sell Cider and things. Like I think we need to figure out how to make the open web look more like the Walled Gardens. You put a dollar in the top, you see how many dollars come out the bottom. The problem is the Wall Garden can do that in a closed loop system. And you know, the open web often looks like a 40 row spreadsheet. And that's the stuff we need to fix. I think a gentic buying, I've often you said it earlier, but like AI has been a hammer looking for a nail sometimes.

In many ways. I think the agentic buying can be a piece of that of just getting the reporting better there. but I think it's cleaning up the industry in general and probably cutting out some middlemen. Like can't pretend like everybody's adding enough value to justify their fees, right?

Rio Longacre (46:53.36)

Okay.

Rio Longacre (46:57.22)

Yeah, so y yeah, and I was not saying all average CPMs or publishers were under dollars, open exchange, which is only a a small piece of it, right? I think it was it is, you know, that was by permitted data reporters under a buck, but been obviously private marketplaces and and programmatic guaranteed deals and other types of and certainly reservation media, you know, much higher and C T V. I mean like you some Netflix inventory is going fifty fifty dollars C PM. I think that's fallen a little bit now, it's now in the twenties, but you know, certainly CPMs can be a lot higher just the pay

Kurt Donnell (47:00.97)

Yeah. Yeah. Yeah.

Rio Longacre (47:24.508)

depending on the type of the type of deals that are being structured, but I but for sure it's been you know, it's been it's been a yeah. But you know it is interesting too. You like I know MFA got a lot better for a while, which I think was good, but I think AI is starting to make it worse again with like the proliferation and then supply path. I mean like you know, I've had people we've had we had Dr. Fu on here and you know he would probably he would probably say otherwise that supply path is just as bad as ever and there you know there were I think eighty percent of bids are duplicate at this point still, you know, which is which is kind which is kind of crazy. But

Kurt Donnell (47:27.371)

Audience.

Kurt Donnell (47:35.416)

While

Rio Longacre (47:54.471)

You mentioned agentic trading. does agentic trading, and for those who haven't listened to the pre past few episodes, agentic trading is the I guess the proposition that a buy-site agent representing either a you know a brand or an agency buying behalf of a brand could negotiate directly with a sell site agent, right? representing a publisher, whether inventory has been, you know, who's able to communicate with an order management assistant or OMS, could then represent that inventory and actually figure out.

Some way to transact. TBD, what types of deals that would be, whether it'll be, I mean, I think if you ask, I mean, Pub PubEx is on here last week and he was saying how he thinks like it could it could actually be for a lot a lot of current programmatic could be flipped over to that. But then we interviewed Patrick Patrick from Booster a couple months ago and he felt it was more, let's say, you know, more of the reservation type media and direct sold deals would probably were probably a better fit. But so I think it's TBD because it's early. But what are your thoughts on agentic trading? Is it is it is it real?

already is it going to be real and does this is this an amazing opportunity for publishers or are they gonna get screwed again?

Kurt Donnell (49:00.014)

I think it's certainly not there yet by any means. I think I've got more fingers on my hands than dollars that have really gone through real deal agentic trading. I mean there've been tests and everything that have been run. I do think it's very interesting from the targeting and packaging. Sorry?

Rio Longacre (49:09.062)

Yep. Have you done tests?

Rio Longacre (49:15.11)

Have you done some agentic training tests? Have you done any agentic training tests for clients?

Kurt Donnell (49:15.576)

What'd you say? We've started working on it a little bit in the background. We've taken a little bit of a I don't say wait and see, but monitoring approach to find out what is real or what are sort of PR stunts to some extent. So we're working on some things right now that hopefully we can announce in the next 60 to 90 days. I'm hopeful on that, that is a little more meaningful than we did something for a press release. we try not to do things that are just hyperbolic that way. I do think it's going to be interesting.

Rio Longacre (49:29.788)

So there's a lot of that, yeah.

Kurt Donnell (49:44.12)

But it's probably going to take a rewiring of the whole system. I don't know that we need a gentic buying to go from an agency to a DSP to an SSP to us. Those are all pretty darn automated systems right now. Programmatic, in and of itself, is an automated high frequency like trading system that does a lot of this stuff. To really rethink it, it has to be something that is different where it's buying directly from a holding company, probably to a freestar, where it is, you know, a good group of inventory.

Doing these things. There still needs to be human human in the loop, like any AI things, but there is, you know, the famous PWC study that you lose 50 cents of every dollar that comes in the top of the machine. There's certainly margin to be captured that could be shared between the advertiser and the publisher, probably with more transparency. And so I do think that matters. And in some ways we've seen that happening. Obviously, the DSP and SSP layers seem to be converging. You know, Trade Desk buys directly from us, Magnet other folks sell directly to advertisers. Like we're already seeing one layer become that.

Does that layer get skipped? I don't know. Do they become the clearinghouse of the agents? You know, we'll have to figure that out. I do think it has great promise, though, from an inventory targeting, reporting, interoperability standpoint that could take what I mentioned earlier, the 40-row spreadsheet and turn it into something that's more meaningful and easier to digest, much like the wall gardens are. So that's where I think it's going. It's got a ways to go because you got to connect pipes from people that have not historically had pipes running between them. And that's the new.

sort of infrastructure that needs to be built.

Rio Longacre (51:15.854)

In an agentic trading scenario, like what can you explain how f like what would Freestar's role be? How would you work with the different players? Like have you thought that through it all? Can you elaborate?

Kurt Donnell (51:25.751)

Yeah, I mean I think and I'm not saying anything bad about SSPs or anything, but it's a lot of the ways that the SSPs aggregated supply in the past.

Rio Longacre (51:32.124)

So okay, we we w y you could say stuff about SSPs. There are some SSPs that y that that don't add a lot of value. I think we can acknowledge that.

Kurt Donnell (51:38.682)

Well, I have good friends. I was actually talking to two good friends at a couple of them that they're doing a great job. But if you just think about the historical view of SSPs where they were the aggregator of the supply side and DSPs are aggregator of demand, right? In an interesting way, the SSPs didn't love the Freestar model and a couple of our competitors' model, because we kind of took care of that. They subsequently flipped to realize we're their best customers because they don't have to go chase down, you know, five hundred publishers and deal with that side. So we've kind of built this ecosystem where we play our part, they play theirs.

where I think that we got value is the scale. And so let's just say it's the what, six big holding companies. If they build a buyer agent, they have to work with it's I don't know the la there was a Jumce report last week. I think it's like the top six or seven what they call sales houses, what we are ad management companies represent like eighty percent of the managed inventory out there. Numbers are wrong, but like it's a large majority of the whole thing. So if you just think about who can interact

with each other at scale. It's the six or seven big holding companies, the six or seven big companies that look like us. You know, there's a pretty easy path you can connect the dots on there. And I think that's probably where it gets. There are a lot of things that SSPs and D SPs do from a targeting standpoint that needs to get built. And that's why I don't think that the agentix stuff is like quite there yet. You have to rebuild a lot of lessons that were learned over a decade, decade and a half of building what is truly an impressive machine, buying in milliseconds with targeting and reporting to some extent, but

AI does make it, you know, certainly much faster to iterate on those things, perhaps create new plumbing.

Rio Longacre (53:09.744)

Yeah, it's interesting too. I've worked with a few publisher clients as well as advertiser clients on on different scenarios here. And and you're right. If you want to do agentic trading at any scale, you kind of have to recreate many of the things that like for example on on the on the supply side, like you you need to aggregate the inventory. You need to be able to communicate with OMS, be able to send orders to GAM. That's what SSPs are for, right? So like you're kind of recreating but and then on the demand side you kind of have to do the same thing. See if so if you let's say you you you you say, Okay, we're gonna bypass

Kurt Donnell (53:32.217)

Yeah.

Rio Longacre (53:39.1)

Current ad tech, that just means you're bypassing the current existing players. But a lot of functionality needs to be recreated. Maybe it gets recreated differently. Maybe it gets recreated with a different cost structure. Maybe different people recreate it. Like, or maybe there's one, you know, there's maybe you're right. It ends up being like as supply and demand kind of cr compressed together, decision kind of floats in the middle. You know, may you know, may I don't know, may maybe it looks maybe it looks different, but all those those all that functionality still kind of needs to exist because those things were not created for no reason. It's a good point.

Kurt Donnell (54:06.347)

Completely. I mean, they really do things. Like as much as maybe they'd get some shit for it. Like they do things. You could argue the fees are wrong. I think some of the tangential things where there's add-ons for this, that, and the other thing, brand safety, all that stuff. I do think you can clean that stuff up with agentic buying of like I can give you a, you know, farm to table impression I promise is a real person. There's no stuffed IDs, there's no this, that, or the other thing. And you can buy my like organic inventory package if you come directly to me. When it goes through other paths.

you know, there's going to be somebody that's gonna try to extract something out of that, or yeah, we'll just shift this and, you know, it's both Kurt and Rio. Like we'll send it two different ways, whatever that is, package it up, because the more I sell, you know, the more fees I take. I think you can cut some of that out, obviously the less middlemen. We'll see how that shakes out. But I do I do think it's promising that you could leverage the agentic tech to do some of those things.

Rio Longacre (54:58.555)

So you you refer to on your website PubOS as an operating system. It's a term that's been coming out a lot. It's a good one. I use it a lot too, but it you could some people argue it's maybe overused. And like my understanding is it combines your freestyle proprietary technology with some third-party products as well as a managed services layer as well. Is PubOS generally like what is it doing that publishers like they're not gonna be able to create on their own? Love to dig dig in and unpack that a little bit.

And then like what other partners do publishers need in order to r to let's say begin to run scenarios like you're describing?

Kurt Donnell (55:32.24)

Sure. So what I would say is kind of what I mentioned. It's sort of the easy button. You install this one piece of software. We obviously use the operating system model as much because it's, you know, explain it. It's very explanatory, but people get it. Like I run iOS and I have the App Store. I run Android and I got the Play Store. Like it just is an easy thing to do. So you install this one thing, easy access to everything. A bunch of it is built in. Our proprietary

flooring technology and dynamic timeouts and a lot of the things that we do are all baked in there. It is, of course, based around our core header bidding platform. We've got our identity testing framework. All of those things are woven in and out. And it's easy button pushes 90% of the time, 99% of the time it's our team doing on the back end. We do have a suite of third-party solutions too that you can kind of have one click access to or preferred pricing and everything. We've made that all interoperable. One thing Freestar's always done is

Maybe slightly different is take a real best of breed approach. We have built a lot of our own tools, but also work with multiple video players. It's not just our own. We never built our own CMP because different publishers have different needs, what they want there. And so we've been, I don't say a pipes company in any way, shape, or form. We've got a lot of proprietary tech, but there's certain things, sort of to my point of if somebody else has built a company or on something that's a feature for me, let's work with them. That's probably gonna be the best. And then when that doesn't work, we can swap something in and out. And so that interoperability is kind of the core.

Of this, of you've got this base technology, unified reporting and analytics. We've made these things interoperable, one touch access to it, so you don't have to go negotiate your own terms. We've heard so many times from publishers, they've got really a brilliant ad ops person or team in-house. They can never test anything because they can't get it through procurement or legal. Even if they wanted to own it themselves, that's fine. Hit the easy button with us. Let's try it for three months. Okay, this makes sense. If it's something that

For whatever reason you have a burning desire to manage the discrepancies with and want to do it, okay, go do the deal yourself then. But really making that seamless for folks ties right into reporting and kind of take away the pain in the ass factor of doing things has been the goal. So it's really ease, you know, our value prop is easy. We increase your revenue, we decrease your costs, and we take away the hassles. And that's what really is wrapped up in this Pub OS.

Rio Longacre (57:41.307)

So did

So where does this run? Does is are these tags they install on their, let's say on their website, on their CMS? Is this run as a separate kind of application? Do they did it is it do they run installed locally? How how is the how does this work and how how would a publisher get this installed?

Kurt Donnell (57:58.85)

Yeah, so it's run like any other header bidding wrapper. You install it on the page. We've got a couple different ways. You can do it depending on the complexity of the setup. There's sort of super easy ways. There's more detailed ways. And we work very closely with people. We've always prided ourselves on kind of more custom solutions and white glove service. And that's how we've ended up working with some of the biggest brands in the world and excited about some new announcements we have come in there too. but it's set up like any other header bidding. Put the, you know, code in your page, determine where you want the ads to run, and then everything else kind of happens in the back end on our side. You've got

A unified analytics dashboard to log into. just launching this week, actually. I'm excited about it. A new section of insights that are powered by third parties like Adomic and Gomera, where you get insights into your advertisers and quality of your content. Ad Fontes, we've part of our marketplace too. So able to bring in third parties that give you unique insights for your business. Does a casual gaming site need to understand exactly what their users are doing, or that probably not as much? Is a lifestyle person where it can help guide what content.

they are going to create and what those affinities look like and how we could do curated deals around them. Yeah. So try to kind of tailor the solutions to what that individual publisher needs. And that's been, you know, the benefit of bringing in some of these interesting third parties.

Rio Longacre (59:11.407)

Makes sense. So it installs as a header bitty wrapper, essentially. Okay, cool. Yeah. Makes sense. So

Kurt Donnell (59:14.679)

Yeah, installs just like any header bridding wrapper, but then there's a lot of bells and whistles that you don't have to do anything for. And that's kind of the the one install, you know, you put Windows on the machine and everything else kind of works has been the goal.

Rio Longacre (59:26.107)

Okay. You mentioned supply path. that that's been you know certainly I I I mean there's been a lot of effort to clean up supply path, this certainly make it more transparent. I think that's been good progress. But from from what I've seen, I wouldn't I I certainly wouldn't argue that the problem's been solved. Is that something you work with your with your publisher clients on? And the reason why I'm bringing this up is there I think there was always this belief among publishers. Well, in order to get higher CPMs and boost yield.

we should work through, you know, with many, many SSPs and that'll give us the better chance of having, you know, of winning more, of getting higher high having higher bids win for our inventory. And I don't know if it's worked out so well, right? I think a lot of these, you know, you'll see you'll, you know, you do these projects where you see there's multiple hops and they're getting marked up and it's it's lots of murkiness in supply paths. So what are you seeing in that space? And then is would using your your technology help help a publisher potentially clean that up?

Kurt Donnell (01:00:24.531)

Absolutely. We handle that all on their behalf. I mean, that's part of the the ease here in taking away the headaches. They don't have to think about that. They don't have to think of what's going client server side, both all of those types of things. Understanding that. And then the advocacy piece of the whole thing. I'm on the programmatic governance council inside of the IAB that's really focused on this and putting out some programmatic was best practice, but really standard practices. What's anticipated and what's duplication versus replication? Is it o it's okay to fan things out? It's not okay to duplicate them.

just for the sake of driving demand density because you sent it to two different data centers. Like what's kind of inbounds and out of bounds? I think it's been an evolving space. For better or worse, much like ad layouts, historically more has been more. If you duplicated something, you get, you know, got an extra percent. If you triplicated it, you got another half a percent and all of those things. And you know, there was the big blow up about transaction IDs a year and a half ago. I'm a big believer in the more we clean this up,

Rio Longacre (01:01:19.675)

I remember that, yeah. That was big news for about a week or so, yeah, I remember that. Yeah.

Kurt Donnell (01:01:23.895)

Yeah, people were furious for about two weeks and then it moved on to the next thing. I forget what was happening. It's probably a Google Monopoly or something next week we talked about, but I I think if we can get that cleaned up and I don't I'm not saying that the Wall Gardens have done a great job because they're wonderful places, what they've done is they made it easy to buy and understand your outcomes. And I think a piece of this is a more unified ecosystem where there is more transparency. And our industry loves to talk shit about our industry more than any I've ever seen. Like

We love like airing our dirty laundry, like, well, this thing and that thing, and you Trade Desk calls out this and somebody else calls out that. Probably isn't helping anything, but if we can genuinely clean it up, show the outcomes, make it easier. And I think supply path is one of those elements. I don't think most advertisers could tell you what a duplicated triplicated supply path actually looks like. They just don't like the idea of it. So let's clean up a little bit of our own perception issue and the PR around it by doing the right thing, which is better, but

There's also the the PR of the whole thing too that's important.

Rio Longacre (01:02:27.962)

Yeah, this industry's so funny. So I remember like it's and it's so like nape the navel grazing is like absolutely incredible, right? And like the specificity of it. So it's funny, I I I was getting my hair cut a couple months ago, and my my barber, he he knows I run a podcast, right? So one time he's like, you what, I've never listened to your podcast. Like, give me so I I told him, he's like, cool, I'm gonna listen to it. So I go in about a month later for my neck my next cut, right? He was like, you know what? I didn't understand a word of that.

I didn't realize like and he listened to like the it was an episode with like chalice AI we're talking about containerization and like just decisioning and it was just it was very nerdy, right? I realized this like only people in ad tech would get this stuff. It's it's it's so it's so crazy and weird and and like and like and specialized, but but we but you we think it's important so you know we can have like arguments on Twitter and or X and LinkedIn for two weeks about transaction IDs, right?

Kurt Donnell (01:03:09.692)

yeah.

Rio Longacre (01:03:21.06)

And then I'll move on to like the next thing it'll be about. There'll be a there'll be a big food fight about like curation or something like that, right? The next week it's so funny. never never ceases to make me laugh.

Kurt Donnell (01:03:28.099)

Yeah. But I mean the bat like it's interesting in our weird little tiny world. When you think about how like few people really, really know ad tech, there's like there's probably more NFL players than people like truly like crazy programmatic experts, which is weird to think about. But what we I think forget about when we're having these fairly technical and academic debates about things like what's the headline out of that? There's a bunch of fraud in ad tech and a bunch of duplication, therefore it's bad and I shouldn't buy it.

Like w we are having what seems like this great academic debate, which is, in all honesty, not necessarily fraud or anything like that. Like could we make it a little more efficient? Sure. Does it use less server costs and everything? Sure. Is it like hurting the w world or the outcomes horribly? Probably not. But that is, you know, to the lay person what comes out of that, right? So we do have a bit of a our own PR problem that we create.

Rio Longacre (01:04:19.268)

Yeah, but it is interesting you say that. So like in generally I think you're right. Like us complaining and carping about transaction IDs is kind of absurd, probably to serve someone not in ad tech, right? Or any any these discussions. But there are some real wor world consequences. I mean, certainly the if if twenty five, thirty percent of traffic is is IVT and in and non humans, right? And publishers are, you know, getting getting hosed out of that, or if you know, AI like Goo Google the walled gardens are are st are

Using publisher content and not sending traffic back, publishers, and they it's impacting the news media, right? It's impacting our ability to get news and potentially impacting our democracy. I'm not suggesting most ad tech problems are like are that are that important in big picture, but I think some of some of them do have real world consequences, you know, which is really interesting. But I but actually, so Kurt, I had an interesting conversation with with an SSP yesterday, and his statement to me, I thought it was interesting. I'd like to get your take on it. So he told me he believes the open web

is contracting. Before I give you my opinion, I'll love your thoughts on that.

Kurt Donnell (01:05:25.145)

Think portions of it are. Like I said earlier, I think the long tail of the internet that was and my wife wrote a blog. That's how she ended up in her business influencer marketing, which is now turned to social channels, but started out as a blogger. I'm not throwing bloggers under the bus. I used to work for a company that bought Blog Her, which was the biggest network of blogs. Like I believe in it, but that as a business model, I think is tough because for better or worse, a lot of it was commoditized content with a personal spin on it. And if you had a relationship with the person.

Great, but a pumpkin pie recipe is a pumpkin pie recipe at some point. Five days of Disney is five days at Disney. Like that good, bad, or ugly, those are things that AI is very good at doing. And so I think that part of the internet is shrinking. I think that's I'll go back to my favorite word here, utility, because it provided value to me as a user. I go read my car review here and I go get my sports news on ESPN. Like from as much as I spend like my time thinking about the internet, I don't like use it a ton, but

Lord knows when I'm getting sports news. I'm going to ESPN.com to read it in my, you know, 10 minutes of downtime because they provide me value. I get certain newsletters every day. I am a New York Times subscriber, for instance, and like they feed me information that's useful to me, built a relationship, and so they give me utility. And so I think that part of the internet will continue to thrive. It's a different internet than it was. And, you know, one hand, that sucks for the creator, certainly that had built a business doing this. And that's

Awful. I think there are different means. There's companies that will help you do content syndication. One of our big publishers, I don't know if they do more revenue through syndication than they do with us, but it is a big, meaningful piece of their business. So figuring out your own distribution channel at times can be as important as anything else. So I think content creators need to figure that out. My wife's a great example. I don't think she's written a blog post in 10 years, but Instagram does well for her. So I think it's finding your niche to express whatever thing you're doing.

And that naturally is gonna mean some portion of the internet does go away. That's just the truth of the matter. What do you think?

Rio Longacre (01:07:22.19)

Yeah. Yeah. I think it's interesting. Yeah. So my opinion, I guess I would say up until AI, I think that there is a business model problem more than it like in which was about like the traditional news publishing business business, the transition to digital was very rocky, right? And like they weren't able to create the same I mean it it it shrank by fifty percent. Just shows that like they were near not able to adjust to this. Now part of the reason why they had that problem was the Walled Gardens were siphoning off some money.

Craigslist eviscerated th 25 to 40% of the class revenue which came from classified ads, which you could maybe argue is a dumb business. They never should have relied on it so much anyway. I mean, that's be my argument, right? It was like it was destined to be to be destroyed because it was so idiotic. But, you know, it is what it is. It was a big part of their business up until around 98, 99, 2000, right? And then, but I think that the publishers had a business model problem as they tried to as they tried to adjust to this. But I think that.

Kurt Donnell (01:07:58.019)

Yeah. Sure.

Rio Longacre (01:08:17.182)

But I think there was revenue there, right? I th I just think that the publishers fumbled it, right? But I think what's happening with AI now is actually it a potential existential threat to the open web because like people just aren't going to these sites. It's not like news publishers can't monetize them. People just aren't going. They're there's they're staying in these answer engines, they're staying in the in it in the chat interfaces. I I I'll I'll have a couple tabs open now. I used to have 20, 30 tabs open at all time, right? That's so that's a huge difference just to my personal behavior. And I and I'm like a

Kurt Donnell (01:08:30.671)

Sure. Yeah.

Rio Longacre (01:08:45.922)

News junkie as as you said before. That's an accurate description. So I think behaviors have changed. That's gonna impact it. But I also think too that the way people consume content is changing. We're going to like see yeah, so I may sp I might not have 20 tabs open, but you know what? I'm in LLMs all the time and I I listen to like crap loads of podcasts. I'm listening to 10. Some weeks I probably listen to 15 hours, right? That you know, all my commuting time, walking a dog, running in the morning.

I never listened to podcasts before, but now there's so much amazing content. I'm constantly listening to it. I'm reading great. There's a bunch of like Substacks. I of people I just a couple economists I love who are just so smart. They publish incredible content. So I think that I'm consuming more content than ever. And I think that the monetization is going to follow the eyeball. So, you know, maybe audio is going to be take a while because reporting's weird. It's hard to understand like any ROI, but like certainly people.

Yeah, we have sponsors for this podcast. You know, they're happy to pay for it. They they they even though it's harder to quantify the exposure, maybe, they still know they're getting it. So I just think that the way we consume content's changing, it's gonna change. And I think the open web will look very different. I think it's gonna be a bunch of cloth like authenticated kind of hedged gardens, right? And then you know we talked to Scott McKinley from Truth Said, he was explaining to us like authenticated traffic is worth quite literally 10X on authenticated anonymous traffic. So like if you're a publisher, yeah, you're gonna lose traffic.

If you if you authenticate everyone'll be but guess it'll like for every authenticated person it'll be able to be wore ten X. He's like, I don't know why they don't just all do this. That was his point, which actually was kind of interesting.

Kurt Donnell (01:10:10.435)

Yeah. Sure.

Kurt Donnell (01:10:16.079)

It's my newsletter thing. Like email is the the one like cheat code of like create value that people will sign up for emails, then they'll be authenticated and then you can work with you don't have to build it yourself, but work with the third parties, work with the SSPs, do more curation around those things and you know, you can drive value even with a shrunken audience. I completely agree.

Rio Longacre (01:10:37.274)

What's old yeah, what's old is new. It's funny. Like I when I started out in digital media, like my first job as a media buyer, we used to do these, you know, run buy these or run a network banner it, banner it stuff or and then a lot of times there was so this is back in like the you know late 90s, 2000, 2001, there was so little actual like inventory that was legit. We would rent email lists, we'd do anything for brands, right? And so so and and it and Ozid performed r pretty well, what's crazy, right?

Kurt Donnell (01:10:41.484)

Yeah.

Rio Longacre (01:11:02.37)

So like you know, what's old is new or back to the future I guess we're gonna we're emails eas emails become highly valuable again as part as part of a and let's say an omni channel like engagement approach with customers. So agreed.

Kurt Donnell (01:11:02.543)

It didn't.

Kurt Donnell (01:11:13.305)

I I think that happens even with AI models a little bit. I think the pendulum does swing back to if you're one of a few authorit authoritative sources of something, you can really still build a sustainable web traffic because people believe you and it doesn't have the slant or it's your fact check on something. So I I guess I I have to say that, right? My business is based on the web, but I do believe that as a person and you see technology go so far and you're like, hold on, let's hit the brakes here a little bit and

I think real journalism still matters, personally.

Rio Longacre (01:11:47.738)

Let's say you're working with a publisher, like and like what are the first things when you when you go in to meet with them that you what are the first questions you typically ask them, Kurt, as they're like you know, 'cause I I think they're all experiencing similar issues, right, with that that we described. But like what are the first how do you begin these engagements?

Kurt Donnell (01:12:03.563)

say is it's changed over the years. I think a couple of years ago, it's like, what are you doing right now? that we can probably raise your revenue 20, 30% just if we do this or the other thing. What we're really focused on, and it's particularly with our focus on the enterprise publishers, some of the largest in the world, is what's kind of total cost of ownership of what you're doing right now. There are so many ancillary little things that people don't think about, whether it is we're paying for all these different services, that's a piece of the whole thing that's fairly easy to quantify. But

How much time is it taking you each month to reconcile billing and payments from 30 different demand sources and deal with the invalid traffic chargebacks? How much time are you dealing with user complaints because you've got malware issues because you're not using a post-auction ad quality vendor that comes baked in with what we do? Then you've got to go run through your procurement team and your legal team and all of that stuff. Like what is the total cost of ownership of doing this yourself?

I do think that most pubs are leaving money on the table. We've got great scale and relationships on the buy side. So we've of course got preferred rates, proprietary technology to do that. Most publishers have an engineer that kind of pretends to be an ad tech person, but editorial always gets the engineering resources unless it's truly a kind of digital first ad tech company that turned into publishing organization. They usually don't have the resources to do that. So how much are you leaving on the table? 'Cause you're four pre-bid versions behind and all of those things. So

The conversation is a little bit more about like, let's peel back the full onion here and what are the layers to this whole thing than just, you're AdSense only right now? Okay, we can fix that for you. Better demand. Boom, boom, boom. I think it's a more holistic kind of audit. That's been and it's been resonant with folks because it makes them think about their business differently and try to be very consultative in that.

Rio Longacre (01:13:46.916)

When looking at publishers' assets, one of those assets I always have felt was incredibly undervalued in let's say in digital media was the publisher first party reven signals, whether that's like CRM data, whether that's contextual signals from

You know, a lot of it just does not go into bidstream, right? I mean, they have a ton of it, right? And I know some of it gets gets pushed from supply over to demand. And then, you know, DSPs can only take so much signal, so they prioritize based on ID. A lot of the other stuff just gets I think ninety percent plus just gets left out because it's just too much when when you we need to transact in two milliseconds to address your point earlier, or you just you can't consider it and the compute costs would be too much, the latency would be too much. So like what are you seeing in terms of publishers being able to to to leverage that that incredible

like source of of signal that they have in in in different types of monetization programs. What are you seeing?

Kurt Donnell (01:14:45.199)

I think you're right. There's unbelievable treasure trove of first party data that's tough to do. Part of it is there is the technical elements you laid out. Part of it's a trust thing. And for better or worse, that trust has got an a roaded hero there. But are you gonna believe me if I tell you everybody on my website's an auto intender? Like I would love the fifty dollar CPM or whatever that is for that auto intender audience. And, you know, you create you know, certain outcomes and you drive certain behavior from the less you know.

The shadier folks in the industry. So that's a tough one to be able to just trust publishers. And it sucks to have to paint the good ones with the same brush as you paint the bad ones, but there's a trust thing. So we've ended up in this world where you need sort of third party verification. The curation vendors, I think, are a piece of that in the SSPs that are doing that. I think it's feeding in into that and making sure that you've got your audience kind of classified right with the people that do have that third party trust. And that could be at the SSP, DSP level, the gomaras of the world to a different, you know, different extent. But

Some sort of good housekeeping seal of approval on this first party data means that it fits in this group. This group has been blessed by the buy side because they trust whatever that intermediary is has been what's needed to happen. This goes back to the gentic buying thing. Is there a version of the world where you can build a little more lock and key around? No, this really is an auto intender because it's on this site, in this section, whatever that is, because the signals are easier to do, less intermediaries, potentially. I think there's a

A real trust thing you can do. Like the blockchain stuff, it is interesting when you think of like a bill of lading of like this good goes from wherever it's manufactured to the shipper and all of that stuff. Like how efficient you can make that with blockchain and digital. We could kind of do that same thing from an impression verification that doesn't get mutated and shifted around and sent through four different supply paths. And with that, you may be able to validate that first party data better to say that, no, this really is a logged in user. Here's the hem, here's the whatever on this specific URL.

Like I'll believe then that that's not on tender. So an interesting not interesting, I mean an application certainly of where a gente could fill in some gaps that have been a trust issue in the past.

Rio Longacre (01:16:53.561)

Cool, I like that. All right, let's move on to some quick hits. Sound good, Kurt?

Kurt Donnell (01:16:58.383)

All right, let's do it.

Rio Longacre (01:17:01.037)

All right, quick hit. So the I'll I'm gonna rattle off a few and answer them. You don't have to answer in one word, but the goal is to rip through these quickly, give me shorter answers, right? So the the most misleading metric in publisher monetization, what do you think it is?

Kurt Donnell (01:17:08.834)

Okay.

Kurt Donnell (01:17:13.111)

Kurt Donnell (01:17:18.295)

It's probably two, C PM and fill rate, 'cause I can screw with either one of those. I can give you a hundred percent fill rate at a penny, or I can give you a thousand dollar C PM and fill one ad. So I think it's looking at RPM.

Rio Longacre (01:17:31.501)

that. That's that's smart. Yeah. I was gonna say I I think CPM was was gonna be my answer, but you combine them together, it's even better. Yeah. What's one programmatic practice that the it industry should just stick a stick a fork in?

Kurt Donnell (01:17:46.628)

do think it's the duplication winning. Like the more times you send something, you can drive more yield is just stupid. It's like that should not work. It still does right now. We're working on fixing it, but it's just it's just stupid. So I I just the I'm an efficiency guy. Like I hate waste and it just seems so wasteful. So that

Rio Longacre (01:17:57.269)

It doesn't work. It doesn't work.

Rio Longacre (01:18:06.647)

Yeah, no, I never th understood why publishers like, we want to integrate with like twenty SSPs. No, you don't. It's not gonna help, right? I I I I th I g I guess I understand why they think it does it sometimes.

Kurt Donnell (01:18:13.751)

It does sometimes. That's the like

Kurt Donnell (01:18:19.063)

It does. Like that's the dumb part, is that's the problem and what has bred this is it does still drive yield. So unless we've got better rules or people are actually on the buy side saying, No, I'm not listening to all these signals, like we've got a problem. So that's what we need to fix.

Rio Longacre (01:18:36.919)

What's a publisher rever s revenue stream that you feel has been underrated?

Kurt Donnell (01:18:43.993)

We talked about this earlier. I think it really depends on the publisher. I would say maybe it's not a revenue stream, but we I think email is a holy grail here. If you want to build an audience, be useful via email that will keep you top of mind for them and get you going back. And you could monetize that too, but email as a mechanism for your other pieces of your business.

Rio Longacre (01:19:05.272)

Love it. What's one walled garden advantage that the open web and publishers should be copying immediately or should consider adopting?

Kurt Donnell (01:19:17.039)

Sheer ease of understanding ROAS. I put a dollar in the machine and here's what comes out the other side. And it's tough because the 40 line spreadsheet I said earlier, but we've got to emulate that one way or the other if we wanna compete and beat them. It's not better content, it is better ease of access and reporting.

Rio Longacre (01:19:35.993)

Makes sense. And last one here, what's one decision an AI yield system should never make autonomously for a publisher?

Kurt Donnell (01:19:47.118)

Publisher and advertiser. Publisher, I would say it would be maybe block lists on categories of advertisers. And then the flip side is advertisers using block lists on sites and just blindly using something that seems like a swear word, but it's somebody's last name. Like we shouldn't be doing that. And we can maybe get there with better context, but blind blocking doesn't work well on either side.

Rio Longacre (01:19:52.695)

Yes, yes, love it.

Rio Longacre (01:20:04.898)

Yeah.

Rio Longacre (01:20:10.328)

Yeah, we could probably have a whole episode about that. Totally agree. I mean, the you know, the the the blunt object of like keyword and domain blocking is just not I mean it's demonetizing anywhere from I think twenty to forty percent of all of all like news publishers' content. It's crazy and like make is ridiculous. So cool. Well, Kurt, this was this is fun. I mean, I I feel you could have gone on. I know I've I've we both have a hard stop in a few minutes, but this is a cool conversation that you know you know you know your stuff, man. So it's been it's been fun. And for those who want to get a

Kurt Donnell (01:20:24.11)

Terrible.

Kurt Donnell (01:20:30.863)

Great man.

Kurt Donnell (01:20:37.977)

Well.

Rio Longacre (01:20:40.268)

Th those of you who want to get a hold of you, how would how would they do that?

Kurt Donnell (01:20:44.351)

give us a look on freestar dot com. We publish a lot on LinkedIn as well. So take a peek there, a lot of content. And feel free to shoot me an an email if you want to chat. Kurt dot donnell at freestar dot com putting it out there. Let's go.

Rio Longacre (01:20:59.148)

Awesome. So Kurt, this is fun, like I said. So for the listeners who have made it this far, thanks for listening to the SIG this episode of Signal and Noise. Another great discussion about kind of sell side and publishers and supply. Certainly where I think a lot of the interesting innovation is happening in the industry right now. So you can find this podcast and all of your and all of our signal and noise content on anywhere you get your podcast, that would be YouTube, Spotify, Apple Podcasts. We publish full transcripts in our website if you're interested.

We also publish shorts across YouTube shorts, Instagram, even TikTok. Last few months, our TikTok numbers are actually are actually going up pretty pretty well as well as on our LinkedIn page. So thanks everyone for tuning in and looking forward to catching up with you next episode. Thanks again, Kurt.

Kurt Donnell (01:21:45.345)

Awesome man. Thank you so much for having me. This is fun.

Can the Open Web Survive the AI Age? Kurt Donnell on Publisher Economics, Radical Transparency, and pubOS

Recent Posts

Comments