Transcript of Conference Call on ‘Google’s Role in Digital Advertising’ with Ari Paparo

On June 26, The Capitol Forum held a conference call on “Google’s Role in Digital Advertising” with Ari Paparo, former Google executive and founder of Marketecture.tv. The full transcript, which has been modified slightly for accuracy, can be found below.

TEDDY DOWNEY: Good morning. And welcome to our Conference Call on Google’s Role in the Digital Advertising Industry. I’m Teddy Downey, Executive Editor at The Capitol Forum.

And joining us today is Ari Paparo, a veteran of the digital advertising world and the CEO of Marketecture Media. Ari is also the author of a forthcoming book, Yield: How Google Bought, Built, and Bullied Its Way to Advertising Dominance, which will be published by Amplify on August 5th. Thank you so much for doing this, Ari.

ARI PAPARO: Yeah, Teddy, thanks for having me. I’m really excited to have this conversation.

TEDDY DOWNEY: And before we get started, just a quick note. If you want to submit questions, we’ll try to get to them towards the end of the call. Put them in the questions pane of the control panel and we’ll collect questions and we’ll try to get to them later.

And Ari, in many ways, you’re one of the only people that could write this book. You’ve had some really interesting roles in the ad world. Can you tell us a little bit about your experience so we can get a sense of your insider perspective?

ARI PAPARO: Yeah, sure. So, I’ve been in ad tech for about 20 years. I’m a recognized expert in the category. I was actually a member of the executive team at DoubleClick when it was acquired by Google, which sort of kicks off the whole story of Google’s dominance in the area. And it’s how I kick off my book as well.

And I was sort of in the meetings where a lot of things happened. It’s funny, people are often asking me, why wasn’t I called to testify at the trial? Why wasn’t I in evidence? Why could I write objectively? And the answer is like I didn’t do anything wrong. I was in the room, but it wasn’t my thing. I was much more on the advertiser side. So, the agency and advertiser side.

So, I spent a couple of years at Google on the team working for Neil Mohan, who’s now the CEO of YouTube, and then had a career at various other ad tech companies. I was the head of product management for AppNexus, which many of you probably heard of is now a division of Microsoft. I started my own company called Beeswax, which we sold to Comcast in 2021. And now I am an independent journalist and analyst on the industry.

TEDDY DOWNEY: And what caused you to write the book when you did? In some respects, timing couldn’t be better with all this antitrust scrutiny. But I’m curious what compelled you to write?

ARI PAPARO: I’m going to hold up the book like I’m on Oprah. So, yeah, it’s interesting.

So, obviously, I’ve been covering this and following it very closely. And I spent three weeks in the Virginia courthouse reporting on it, writing a daily newsletter on the happenings in the trial in front of Judge Brinkema. And I just found it really interesting. I thought these stories were really compelling.

And it wasn’t just this very technical question about how Google manipulated auctions. It was people’s stories about people losing their jobs, about people not being able to do their jobs, feeling like they were being gaslit or otherwise not being treated fairly by Google. And it made a really good story.

There’s a clear start, middle, and end. Start is the invention of the ad exchange and the financialization of advertising. The middle is Google’s just march to dominance where they just spent ten years controlling everything. And the end is the comeuppance in legal circles and eventually to the possible breakup of the company.

TEDDY DOWNEY: And the book has a really useful timeline. It goes through that history, as you mentioned. Can you pull out some of the moments where you thought, all right, these were pivotal moments for Google in their dominance and how they came to become dominant?

ARI PAPARO: Sure, yeah. The story starts, and the most important moment is the acquisition of DoubleClick. So, DoubleClick was a private company at the time owned by a private equity firm. And it was a very dull company. And I worked there. I was a VP there. It was effectively selling software to publishers to serve ads and selling software to agencies to buy ads. It was maybe a $300 million revenue software business, maybe a little more, maybe $400.

And it had been sold previously to the private equity firm, and no one had bid on it. Google didn’t bid. Yahoo didn’t bid. It was seen as a stalwart, not a very exciting company.

And Google and others realized that as ad exchanges were being invented and the entire market, the trillion-dollar advertising market, effectively was going to become a liquid market like Wall Street for the first time really since the invention of advertising, that technology was going to win. And DoubleClick was the leading technology. It had a near monopoly position at the time.

So, that’s the most important part of the story is that they acquired it. And I guess the deal was approved by the FTC and completed in 2008, I believe.

And then moving forward, there were a lot of little incremental changes over the years where Google was changing the way the auctions worked and putting more emphasis on the way they sold to advertisers.

For example, before the DoubleClick acquisition, Google didn’t allow third-party cookies, which meant that they really couldn’t compete with other advertising sources like Yahoo, AOL, and MSN, where third-party cookies were just sort of the lingua franca.

And about 18 months after the DoubleClick acquisition, they made a giant policy change that they allowed third-party cookies. And it was just like taking the reins off of a horse. I don’t know if that’s a good analogy. But basically, it just let them loose and they started selling ads like crazy. They had all this inventory, et cetera.

The interesting story that’s in the book — and also was covered by the press at the time — was later in this whole period, I don’t know, something like 2016, 2017, Google was sort of at the height of its powers. And they decided to try to fix a lot of the problems in the online exchange world. Unlike — I know a lot of people on this call have a Wall Street background. Wall Street is simple and very well regulated compared to advertising. Obviously, Wall Street is not simple. But the advertising industry had no rules and everyone was running auctions in different ways. And as a result, there were lots of inefficiencies.

So, Google decided to make a clean break and to fix it all at once. And that meant they were going to move from second price auctions to first price auctions, which was a really big improvement because the second price auction, where the bids are reduced, is rife with ability to cheat. And people were cheating, not just Google. Everyone was cheating in different ways in the advertising world.

So, that was really important. They were going to do that. But then they also did something that was really self-interested. They controlled the ad server, which is the software that publishers use to determine what ads show on their site. And the ad server is really complicated software. It has all these tools for allowing publishers to get more revenue or yield, which is the name of my book.

And Google decided to just remove the pricing features in the ad server. They put out a new feature called Universal Pricing Rules or Unified Pricing Rules, UPR. And it was blatantly obvious to the customers that this was self-interested, that it was a way for Google to effectively get higher prices when they sold the ads on behalf of publishers.

And it was not in publishers’ interest, even though the publishers or the customers are paying for the software from Google. The software is not free. They would be paying millions of dollars. And they announced this as if it’s a good feature for publishers. And that was kind of the straw that broke the camel’s back.

First of all, it was a big deal when it happened. It was in all the trade press. People were really angry. They felt like they had just been betrayed by a vendor that had too much power. And it was pretty much, you could draw kind of a dotted line between that event and the legal scrutiny.

Some of the investigations were already underway. But when the folks at the state attorneys generals and the international groups and the DOJ just saw this, it was like, you took a software product and made it worse over the objection of your customers to your benefit. That doesn’t sit well with people.

TEDDY DOWNEY: Yeah. So, was it obvious to you at the time when they bought DoubleClick that that was an anti-competitive merger that would be problematic? Or should have been problematic from a legal standpoint? Were you surprised that it was allowed?

ARI PAPARO: No, I don’t think I was. And if you go back and read the FTC’s approval of the merger, they were doing their best to guess and figure out what was going to happen.

The reality is, first of all, Google at the time and DoubleClick at the time were not competitive at all. So, there was no overlap whatsoever between those businesses. They were compliments, which is important to realize.

The second thing is that the thesis for the deal, the reason Google bought DoubleClick was because Google had all this demand, which was the advertisers. So, you think about supply and demand. The demand is the advertisers, the Google ads, logins, advertisers, with huge budgets going to search. And DoubleClick had all the supply because they had the publisher ad server.

And so, the idea was you bring the two together. The fact that both of those products are dominant in their industry — Google had the most demand of anyone and DoubleClick had the most supply of anyone — did come up. There were Congressional hearings at the time and some experts brought it up.

But the FTC said, more or less, this is so nascent that we don’t know how it’s going to develop. There’s lots of competitors. These companies aren’t competitive. What can we do?

And unfortunately, they said in their document — which is still available online — we will be watching it very carefully. And if there are any abuses, we’ll take another bite of the apple, which they did not do.

So, I have sympathy for them at the time. And certainly, as an executive, I didn’t see any issue with it. I was benefiting from the merger. So, maybe I was self-interested at the time.

TEDDY DOWNEY: Yeah, it’s interesting. FTC did absolutely nothing about it. And DOJ was the one that ultimately scrutinized it and sued.

You were at the trial, the Brinkema trial, the AdTech trial. I hate to just jump into remedies, but I think a lot of people are interested in what is going to happen with these remedies. I think you have some interesting thoughts.

What’s your take on DOJ’s ask for remedies? And would also love to get what you think would be necessary to actually create a functioning open market that isn’t for publishers?

ARI PAPARO: Sure. First of all, the Brinkema decision didn’t surprise me or anyone else. I think the evidence was pretty overwhelming in the case.

And Google’s defense, they largely didn’t even defend their actions. Their defense was entirely legalistic about the Trinco and other cases that they felt were relevant.

So, the decision wasn’t surprising. I think that the DOJ’s proposed remedies were a little surprising because I think, in my opinion, they’re a little too complicated. They’re trying to be a little too on the nose about how they’re trying to create competition.

So, just in a very rough sense, what they’re proposing is that first, the ad exchange gets spun out. And as a result, all of Google’s demand, meaning AdWords and DV360, their buy-side tools, have to effectively compete on the same basis as all the other demand from the internet.

I think that’s great. I think that’s really 90 percent of the battle right there, is that publishers can get the Google demand right aside, right next to other demand from any other source. That’s a big deal.

Then the part where it gets a little complicated is instead of potentially asking Google to immediately spin out its ad server, DFP or GAM, it has different names, which has 90 percent market share, is absolutely dominant. They are trying to create conditions that will incubate competitive ad servers.

So, they’re asking for the algorithms inside the ad server to be open‑sourced. So, potentially a competitive ad server can use those algorithms. And the buy algorithms, we’re talking about how the ads get served, in what order, and stuff like that. It’s pretty complicated.

And then they’re asking for APIs to allow customers to switch to a different ad server. And then they say they want the existing ad server, GAM, to stay at Google for at least three years. And they call it the rump ad server, which is a little insulting. And then after three years, reconsider and maybe force a spin out.

I don’t like this solution at all because it is a 90 percent market share product. So, publishers are on it. They’re stuck on it for a while. In the best case scenario, we’re talking about a year or two before a publisher could potentially viably switch to a different solution.

And you have Google being forced to continue caring for this unwanted child for years. They’re going to have the worst engineers on it. They’re not going to be able to do anything innovative without the court asking. They even call it a rump product. And it’s all in this pretty speculative belief that a competitor will emerge based on all of this.

I just think it’s too complicated by a couple of measures. I would prefer them to ask for an immediate spin out of GAM. I think as an independent company, it could start innovating and you could put restrictions on its ability to reform the monopoly with a different exchange or things like that. And I think that would be better for Google, better for publishers, better for GAM, and it would be just a lot cleaner.

TEDDY DOWNEY: Would you run into problems if it’s just a monopoly in someone else’s hands?

ARI PAPARO: Yeah, yeah. So, that’s the real problem. And this is the problem also — we’re not really here to talk about the Google search case with Judge Mehta, but it’s the same problem as the Chrome, which is any of the potential buyers would be worse.

So, if you sold GAM to, say, Amazon, they could integrate it with their demand and their supply and create the monopoly. And if you gave it to, say, a private equity fund, likely they would want to extract more profits from it. Because GAM is probably not very profitable right now. All the profits come from the exchange.

So, you might be in a situation where the costs go up quite a bit. So, publishers get stuck with a much more expensive ad server, even if it’s not a monopoly anymore. So, there are some really tricky bits here.

TEDDY DOWNEY: And you mentioned the search case. I am interested in how you think those two things go hand-in-hand. Or the importance of the remedy and search ultimately matching up with what we’re seeing.

ARI PAPARO: Yeah. So, it was pretty interesting that right around the time that the remedies were happening in the search case, the remedy trial or hearings, Google announced that they were not going to get rid of third-party cookies. And that was the end of a five-year odyssey where the Chrome team had effectively dominated the product development roadmaps of every ad tech company for four or five years, and ultimately total waste of time. And the reason I bring that up is because it was illustrative of the power of Chrome on the advertising business.

And the fact that Google and Apple, to some extent, acts as unregulated regulators on this huge swath of the economy. And I think that incident — the fact that the sandbox debacle was a waste of time and that it showed how much power Google has, and Google can just turn off cookies the day after the trials are over if they want to — is just more illustration that Chrome is really important and a potential spin out of Chrome needs to be thought of in terms of its effect on advertising.

The other point I’ll make, which is kind of a little more subtle, is the fact that Google is fighting so hard in the ad tech case is very puzzling. The ad tech business, from a financial perspective, is not very important to Google. We can look at their financial savings. The network business is not growing at all. It is the least profitable. The fact is that any of its growth is likely happening on the mobile side, not from GAM and not from these products that we’re discussing.

So, why doesn’t Google just turn it off? Why don’t they just spin it out? It doesn’t make any sense that they’re fighting this hard. It would actually be better for the company — I think everyone agrees — if they spun it out.

And so, not to put on a tinfoil hat or anything, but it appears that one of the reasons they’re fighting so hard is because they really want the publisher tag on the page. That ad serving tag that goes on the page, meaning JavaScript, HTML, whatever it is, gives them so much insight that feeds into the search business and the other businesses. That’s the asset they’re fighting so hard to keep, even though it’s not part of the financials of the network business.

TEDDY DOWNEY: So, it’s a data play. They would lose data if they lose all this control over the ad ecosystem.

ARI PAPARO: Yeah, I think that’s an opinion. There’s no smoking gun on this. But one thing to note, in terms of evidence, is through the emails, especially in the later portions of the timeline, the Google product managers feel really strongly that they want to have their tag on the page. That’s their North Star about the publisher business. We always need to have tag on the page. We don’t want someone else’s tag ahead of us.

And that data does feed into a lot of things. It also feeds into PMAX. PMAX is Google’s AI play for advertisers, which is a huge moneymaker. And PMAX is the sort of set it and forget it advertising product that takes the advertiser’s money and distributes it to YouTube, to search, to the network, et cetera.

I think a lot of the PMAX data is coming in through the network business, even if the advertising dollars don’t necessarily go out to that. This is a very long answer to your question, which is what is the search business?

And so, the search trial and the ad tech trial are separate, but there’s this data layer that goes between them that’s really important to the overall monetization of the company.

TEDDY DOWNEY: We talked a little bit about this previously, but I’m curious to get your thoughts on whether or not holistically all of these remedies would work together to actually give the publishers an open ad market and a functioning way to get ads on the internet. What’s your take on the workability of this in terms of providing a real solution for the publishers?

ARI PAPARO: Yeah, I think that it’s a little late because most of the ad serving companies that were trying to compete with Google have given up because of the monopoly. So, in the trial, AppNexus, where I used to work, and Kevil, which is a small company based in North Carolina, and OpenX, which is in California — these are all private companies at the time — they all testified that they couldn’t compete with DFB. So, they gave up, more or less, and they exited the market.

So, if they had been successful, or if a new publisher ad server could be incubated because of these remedies, then publishers would have a choice. It would lower costs for publishers. So, they would keep more of every dollar that was being spent. And there could be an opportunity for innovation.

I think another point that’s very important is that the demand that goes into this web publishing, open publishing, world is somewhat stifled because people like Facebook, Meta, do not send any of their demand to the web. If you buy an ad in the Meta console, it will go to the Meta properties and some app stores install ads, but they don’t spend any money on the open web.

And there were documents in the trial that said specifically, we, Meta, don’t want to participate in the open web because no matter what we do, Google’s in between us and the publisher. We can’t get around them. And that’s why they ended up signing that very controversial Jedi Blue deal.

So, in an alternative world, you can imagine Meta wanting to have as big a presence in the open web as they do in the app world or on their own apps. And that could be billions of dollars of new media dollars that could go to publishers.

Now, I think there’s a lot of things that would have to happen for Meta to decide to rejoin the open web, but it’s important to realize that they’ve been prevented from doing so by this monopoly.

TEDDY DOWNEY: I’ve got so many questions here. You mentioned the online ad market. Do you look at all at the gaming ad market? Are you familiar with that at all?

ARI PAPARO: Yeah, sure. There’s two segments. There’s the end‑game market, which is like console games and desktop, which is pretty small and mostly sponsorship still. And then there’s the in‑app market, which the majority of in-app spending was game related.

TEDDY DOWNEY: Yeah. We’ve done some reporting on AppLovin and they’re sort of taking the –it’s got a very similar business model to Google in many respects.

ARI PAPARO: It does.

TEDDY DOWNEY: Sort of emulated what they’ve done. Do you think there’s any hints of market power issue there in that market or any scrutiny deserving of AppLovin for how they run their business in the mobile ad market?

ARI PAPARO: I don’t think market power is the real issue there. The app market remains fragmented. So, I don’t have exact numbers. But between AppLovin, Meta, Google, Unity, Iron Source, and a bunch of smaller competitors, it remains fragmented. I don’t think anyone has more than 50 percent share by any means. I don’t know what AppLovin share is, but I can’t imagine it’s close to 50.

So, it’s still pretty fragmented. I think that the mobile market has some of the problems that the display market had, that I mentioned earlier, which is effectively cheating. Basically, not great ads shown in not great places using data that may be not allowed by app stores.

So, there’s a lot of shady stuff that happens in the app market that’s hard to detect and may not be illegal, but may not be exactly what advertisers are expecting their dollars to be doing. So, that’s really the enforcement edge in app.

TEDDY DOWNEY: And we’ll go back to Google in a second, but isn’t AppLovin vertically integrated?

ARI PAPARO: Yeah.

TEDDY DOWNEY: And aren’t some of those ads sort of more dominant in the ad tech stack for mobile ads?

ARI PAPARO: Yeah. So, they are vertically integrated. They have apps they own, which they’ve discussed spinning out. So, they’ve owned and operated properties. They have the SDK and the Ad Exchange, and then they have the demand side.

So, that’s a play that other people have made too. Being on the buy and sell side together is not necessarily an area for abuse. It actually can produce better results. To some extent, Facebook’s in that area. They have an SDK and they have a demand. So, they connect.

And like I said, I don’t think they’re anywhere near dominant. I mean, if you’re an app provider, you do not have to work with AppLovin. And if you’re an advertiser, you don’t have to work with AppLovin. You can buy ads through Meta and Google and other places like that.

TEDDY DOWNEY: And getting back to the AdTech case, one thing that I think has been interesting is Google. obviously, the judge has said they’re a monopolist in ad tech. There are big class action lawsuits by publishers against Google. But those are really focused on the take rate that Google gets in each ad.

And I’m wondering if you can discuss the difference that premium publishers — also, we’ve done some calls on this as well — would have a claim if they directly sue Google around the manipulation that Google was doing in the ad market on individual ads. Can you tell us a little bit about the difference in how those two rip-offs work and how big potentially the size of that manipulation rip-off would be?

ARI PAPARO: Sure, sure. So, two caveats. First is it’s very important to say I’m not a lawyer. I’m a businessperson. So, do not take legal advice from me. Secondly, I’m not sure rip-off is the right word.

TEDDY DOWNEY: What’s a good word? Illegal conduct.

ARI PAPARO: Potential violations of their contracts.

TEDDY DOWNEY: But it’s already been — the court’s already said it.

ARI PAPARO: No, the court didn’t say that. The court didn’t say that. In the DOJ case, the DOJ decided not to bring forth evidence about auction manipulation that was largely outside of the scope of the case, probably to simplify it because it’s a complex case. In the Texas case, which is going to start August 11th, they are very focused on contract manipulation and bad dealings.

But in terms of overcharge, let’s talk about overcharge first. So, the evidence shows that for open exchange auctions, which is the most common form of auctions, Google charged 20 percent fee and they never discounted it. Their effective fee was something crazy like 19.9 percent or something like that. That shows how much power they had. Even their biggest customers, people like weather.com, News Corp, they would never get a discount even if they asked for it.

And so, that 20 percent fee, there was evidence shown that it was higher than the competition and that, while the competition moved up and down quite a bit, they never moved. It was just a flat line. And the expert witness claimed that the proper rate was 16 percent. And there was a lot of discussion in the court around that. Was that the right number or not? And I think this will happen in the remedies side. There’ll be a lot of discussion about whether 16 is the right number.

So, in terms of volume, you could do some real back of the envelope calculation here, which is that the ad exchange at Google transacted about $10 billion in media per year in the latest year in which it was disclosed, which was 2020.

So, 4 percent overcharge times $10 billion, $400 million a year in overcharge across all publishers globally. So, that’s back of the envelope. That does not account for other parts of Google that may have been taking additional rents. The Google ad system, for example, or the buy side, DV360, and a bunch of other stuff. But $400 million in overcharge to publishers per year is a good back of the envelope number.

So, then the second question is, I think, the manipulation of the auctions and whether that was a breach of contract or not to their customers’ benefits.

And that’s really interesting. That evidence has not been shown. I mean, a lot of the evidence is out there in the public because the emails are out there, but it has not been adjudicated yet.

So, some of the things that Google did were they would have a contract with a publisher saying you’re going to get this 20 percent take rate. And then in real time, they would change the take rate so that they matched better. So, if the demand from Google was not going to win when reduced by 20 percent, on that specific auction, they would only reduce by 15 percent and they would win the auction.

So, it benefited Google overall because they won the demand and the supply from Google matched better than it would have normally. And the competitors were worse off, but the publishers were not necessarily worse off. The publishers actually got a lower rate than their contracted rate. That was called Dynamic Revenue Share, DRS.

But then they’d move onto the next phase and they would say, well, on some auctions, it’s not competitive at all and the buy side is willing to pay a lot more. Let’s even it out. So, we only charged 15 percent on this last auction. Let’s charge 25 percent on this next auction and it will average to 20.

And there was a bunch of that. I could probably talk about it for an hour, which you don’t want. But the bottom line was they were really skirting the edge of what their contract said they were allowed to do versus what they did.

And most of it was to the benefit of publishers. So, the evidence shows that it hurt competitors, but it helped publishers, in most cases, that they made more money.

So, it’s dicey to talk about these manipulations of the auction as necessarily immediately turning into a big windfall for publishers.

TEDDY DOWNEY: That’s interesting. We’ve had some experts say that it was probably billions of dollars in lost revenue, but that’s interesting to get your take that it might not be that big.

ARI PAPARO: We don’t talk about advertisers. So, the Google advertisers on Google ads were systematically overcharged, I think. That’s what the evidence shows is the demand side was subsidizing the sell side. The sell side of Google AdX was not a great business and was very competitive with people like Magnite, Pubmatic, et cetera, had to do all these manipulations to win more options.

But the buy side customers — small mom and pops who had no ability to see how much they were being charged. All they knew was like they put a credit card in — those people were paying outsized amounts.

TEDDY DOWNEY: Interesting, interesting, yeah. So, in terms of some of the remedies that are on the table, we’ve got a question here on the possibility of Google divesting its ad exchange into an open source platform. Do you think the judge will ultimately agree that, hey, these divestitures are necessary?

You’re even saying the consensus is that this would be good for Google, just in terms of if you take away the data part. So, it seems like maybe the feasibility or the costs or the whatever is probably overcomeable. But what do you see in terms of what the judge will do and in terms of like is it achievable and how easy or difficult that would be?

ARI PAPARO: Yeah, so the main interesting thing that’s happened recently is that the judge has ordered Google to turn over its internal documents about the feasibility of this. And there’s documents under the phrase Project Sunday and Project Monday that Google has done analyses, non-legal analyses, of how difficult it would be to spin these things out. And Google obviously did not want to turn that over.

So, I’m anxiously awaiting seeing those documents. Because it’s not going to be easy. These products are all very intertwined. Just thinking about cookies. The cookie ad at doubleclick.net, that probably everyone has seen at some point in their career surfing around the web, is used by a lot of products inside Google. You can’t just say, oh, well, the publisher side is going to switch cookies when it gets spun out. It’s really complicated, that very little thing. And that’s only one part. There’s so many ties between these things.

But ultimately, the government and the judge shouldn’t care. If they want to break this thing up and it’s costly and difficult, that’s just the cost of doing business. There’s no reason it can’t be done. It’s just work, engineering work, financial work.

I think my biggest concern, and a lot of people’s biggest concern, is how viable these companies are if they’re spun out. The AdX, the ad exchange, on its own would be just another competitor without any real competitive advantage versus other exchanges. It has 50 percent market share, but the market share is highly buoyed by the fact that Google ads will only bid into the ad exchange.

So, you have the captive Google ads demand that 90 percent of that demand goes to AdX. As soon as AdX is spun out, that 90 percent becomes a much — I don’t know what number — becomes a much lower number. So, a huge amount of the demand gets cut off from AdX as soon as Google ads starts bidding fairly.

And the other thing Google AdX has is it’s directly tied into the publisher ad server, GAM. And that gets broken by — that’s the whole point is to break that.

So, now you have an ad exchange that has no real competitive advantage and has probably very high costs because it’s built by Google engineers on the Google stack. And it’s competing with standalone companies like Magnite and Pubmatic, both of whom are public companies who are highly optimized for what they do. So, this spun out thing might just be a failure. And that would be a shame, but it could happen.

But with GAM itself, the ad server, that failing is not an option because it has 90 percent market share. Basically, every ad supported business on the entire internet is dependent on that software. And if it were to falter or start going down or prices went up or whatever, that would have an immediate impact.

So, I think that’s the much more delicate one. And that’s probably why the DOJ has recommended doing something a little more staged.

TEDDY DOWNEY: But when they spin this off in a sale, ostensibly the price will reflect that risk.

ARI PAPARO: Yes.

TEDDY DOWNEY: It’s like, all right, 50 percent share, but not as much — you’re just going to end up being a commodity or whatever. You’re going to end up being a normal competitor against these other companies.

ARI PAPARO: Right. So, the court doesn’t care.

TEDDY DOWNEY: Cost disadvantages, yeah.

ARI PAPARO: Yeah, the court doesn’t care about preserving the equity value of Google. But what the court does care about is if it affected ‑‑ I think the court cares — about the stream of revenue going to publishers because the publishers are the complainants and you don’t really want this to interrupt the revenue flow and throw out the baby with the bathwater.

TEDDY DOWNEY: And let’s talk about, I’m interested to get your take if you’re following Europe and Canada and what you think. Or even, we touched on the state case, in terms of what would need to happen in those cases to further address Google’s dominance. Like what would you like to see? Or what do you think needs to happen in all of those jurisdictions for there to be a real solution here?

ARI PAPARO: So, Europe effectively is asking for a full spin out of AdX and GAM. It was even reported by, I think, Reuters last fall that Google had offered Europe to spin out AdX. So, that seal’s been broken. Google is willing to spin out edX to make the Europeans happy. And European publishers said, no, we want the ad server spun out too.

So, that’s already under the bridge. I’m a little less familiar with how the rulings work. But my understanding is that the European decisions are effectively baked in. They have found them to be a monopoly and they’re sort of waiting for Brinkema before they dive in with their hard-nosed demands. But I think Europe is going to want a more aggressive spin out than what the DOJ is asking for. And we’ll see how that gets negotiated.

Canada, same thing. Canada is sort of a carbon copy of the European, but it’s earlier. It just got filed in December. Strangely, Mexico said Google’s fine, no monopoly. So, I don’t know if that matters.

But the Texas case is the most interesting thing. So, the state case led by Texas and a bunch of other sort of red-leaning states has been actually going on a lot longer than the DOJ case. It’s just the DOJ case had the benefit of being in the rocket docket.

The Texas case has moved around jurisdictions multiple times. It starts August 11th in Plano, Texas. I’m going to be spending my August in the heat of Texas. That sounds great for me. And my understanding — and this is, again, not a lawyer — is that Texas is going to spend a lot more time on these auction manipulations than the DOJ did.

They have a jury trial. So, the argument is likely going to be around fairness, that Google acted unfair. And that Texas also is trying to make a much larger financial impact here because their allegations are on breach of contract and things like that, that all have fines associated with it. And the volume of transactions makes this like almost a trillion dollar settlement kind of thing. Obviously, that’s not realistic.

We also have the data point that Google just gave into Texas and Ken Paxton for $1.5 billion earlier this year for a nonsense suit about Chrome privacy mode. I say nonsense. It was just a trivial suit and they were able to extract $1.5 billion from Google on that.

So, I think Texas sees blood in the water and I wouldn’t be surprised to see just an absolutely eye-popping dollar amount out of that suit. Whether it actually gets paid or how it gets resolved is kind of above my pay grade.

TEDDY DOWNEY: You mean in a judgment by the jury?

ARI PAPARO: Yeah.

TEDDY DOWNEY: Are they going to ask for non‑monetary remedies also?

ARI PAPARO: Yeah, they would probably also ask for that.

TEDDY DOWNEY: And do you think there’s anything in there beyond what we’re seeing in the DOJ case that would be — or is it probably going to end up being along the same lines?

ARI PAPARO: I don’t know offhand. I think that the Texas case has a bunch of elements that were not in the DOJ case. Like they have the Facebook Jedi Blue relationship is part of that case, or may or may not be, depends on what day you ask. I think they’re just going to try to throw the book at Google and see what happens.

TEDDY DOWNEY: One thing that’s sort of looming over these Google decisions is the shift to AI. How do you think about any remedies related to AI as being important or not important to a solution here?

ARI PAPARO: Yeah, AI is not that relevant directly to the ad tech case. It’s incredibly relevant to the search case. And I think Judge Mehta is taking it into consideration there about how, first of all, Google is trying to run the same playbook in AI that they ran in search by becoming the preferred provider. There was some evidence that they had tried to do a deal with Samsung to become the ‑‑ make Gemini the preferred AI provider.

There’s also the question about robots.txt and how Google is effectively forcing publishers to either block both search and AI or neither, which is not really a fair way of looking at things. And once again, having all this data through Chrome, through the tag, it all helps the AI.

From a publisher’s perspective, obviously, AI appears to be really bad for advertising supported publishers because the referral traffic from Google as well as from social networks is declining precipitously. And to the extent that traffic turns into advertising dollars, that’s bad for them. But it’s not directly — it has nothing to do directly with the allegations in the ad tech trial.

TEDDY DOWNEY: So, taking all this into account, you’ve got DOJ. You’ve got Texas. You’ve got Europe. You’ve got Canada. In a world in which Google gets broken up, how do you see things playing out going forward in the market? Like where do you think the opportunity is? Where are you going to be focusing your attention? How do you think this all plays out?

And it seems like you think Google is going to be fine. They’ve got YouTube. They’ve got all these other assets that they can go monetize ostensibly. But curious how you think the market develops in a post-litigation world.

ARI PAPARO: Well, the market, and there’s Google’s involvement. The market’s interesting because we’re seeing this long-term secular decline of what’s the shorthand is the open web. When I say the open web, you probably picture websites with news on them and banner ads. And that sector is in decline. There’s no way around that. Consumers get their news from social, consumers move to mobile and in-app, which is less conducive to reading news. And like I said earlier, referral traffic because of AI and other things is declining.

Meanwhile, the advertising business is thriving because consumers are seeing ads and interacting with ads in all kinds of new places like Netflix and Spotify and e-commerce sites, Walmart, DoorDash.

So, the advertising business is evolving and shifting from that first model that was really invented in the late 90s to a new model, which, for better or worse, is more controlled by fewer companies. But from an advertiser perspective, there’s just new ways to reach your consumers.

Google’s involvement, well, I think it’s pretty clear that Google’s top priority in advertising is YouTube. I was just at the Cannes Festival in Southern France and Google has an amazing, huge presence on the beach. It was all YouTube. It was YouTube, YouTube, YouTube.

Neil Mohan spoke. It was about creators. They don’t talk about the network business or the ad server. They don’t care. It’s just not important. YouTube is the future for their advertising business.

Which is why I would advocate pretty strongly — if I was the Google CEO, suddenly that was a job that was made available to me, I would spin the network business. It’s just a big distraction. It’s where almost all the privacy problems happen. It’s not very profitable. And Facebook doesn’t have one. TikTok doesn’t have one and they’re doing fine. It’s a strategy whose time has come and gone for the company and they should accept that.

TEDDY DOWNEY: If there’s no real strategic value, and obviously we can get into the point around the marker where they keep access to all the data, but even that, this seems strategically makes no sense. You’re saying this just doesn’t make any sense. Why are they doing it?

I mean, is it about power? Is it about ego? Is it about just fighting government so that you keep their eye off the ball so they don’t regulate YouTube, so they don’t go after another part of the business? What’s the point here?

ARI PAPARO: We don’t know, but there’s a couple of hypotheses. The first one we talked about earlier is the data. The second one is — particularly we’ve been talking about the sell side — but on the buy side, the idea that you control all these customers, millions of advertisers, and you control the algorithm that says where those dollars are going is pretty important, especially with regard to YouTube.

Basically, if you’re using their DV360 tool, you’re taking those budgets of top 500 advertisers and you’re putting some of it to YouTube and you’re not putting any of it to TikTok or to Meta. So, that you could make a pretty strong strategic argument, still has some relevance.

And then the third theory or hypothesis is that there’s no benefit from a game theory perspective for giving up until you get what you want. And what they want is to maintain their Chrome and search monopoly.

So, why settle this when you have the bigger case, the more important case still pending? Let all the chips come out and then look for a grand unified bargain that gets you the best deal you can.

TEDDY DOWNEY: I see. So, basically make a play to the White House, to Trump or whoever’s in the White House at the end, saying, hey, we’ll give you everything else, we want to keep Chrome.

ARI PAPARO: Maybe. I think they would take that deal today.

If they could get that deal, make the search case go away, stop paying Apple the $20 billion and get rid of the network business, that would be good for Google. They would take that deal.

TEDDY DOWNEY: That’s so interesting. That’s so interesting. I’ve been amazed. D.C. is much more focused on, oh, can they unscramble the eggs? Like, oh, is that something that could really happen?

Whereas, I think on Wall Street, the consensus is this would just not be hard. This is actually totally doable. We spin companies out all the time. There’s going to be a price for this asset.

ARI PAPARO: Yeah.

TEDDY DOWNEY: Google’s going to not really — if they lose some money, who cares? Like, whatever.

ARI PAPARO: Yeah.

TEDDY DOWNEY: Spinning it out. But it’s a funny disconnect hearing what you say versus what the lawyers talk about and worry about. And you’re in the courtroom, reading the decision. Honestly, reading Brinkema’s decision is easier to read than the DOJ complaint.

ARI PAPARO: Yes.

TEDDY DOWNEY: I actually think the understanding of the market is pretty thorough. I wanted to get your perspective as an industry insider that the lawyers, the judges in the cases, ability to understand this super complex market. I mean, you go through in your book how it’s really complex.

ARI PAPARO: Yeah, it is.

TEDDY DOWNEY: I mean, it’s hard to follow some of this stuff. You have amazing diagrams. You lay it out in, I think, a very accessible way, as does Judge Brinkema. But what was your take on how the judges understand the market?

ARI PAPARO: I think she did a great job. Or her and her team did a great job. It’s very readable. And it’s pretty much in plain English that the ad server is a monopoly. I mean, there’s overwhelming evidence of that. The ad exchange is a monopoly and is tied to the ad server. And this is not to the benefit of publishers. That was the conclusion.

And I think the evidence that the DOJ presented was pretty overwhelming on those subjects. I think the DOJ did quite a good job of explaining things. They made a couple of sort of shortcuts or simplifications that some people nitpicked about. But I thought it was very good.

I think the Google defense was not good at all. Like I said earlier on this call, they didn’t defend their actions at all for the most part. They really just went to this absurd market definition argument that made no sense, where they said, everything is advertising. There’s one transaction. There’s no such thing as an ad server market, which just didn’t hold any water. And then the rest was just legalese around the Trinco precedent and a bunch of others.

So, I don’t think Google did itself any favors. I think Google is still not doing itself favors in the way they respond to all the remedies as just being ‑‑ the Google response to the remedies was like a three‑year oversight. And that was crazy. It’s just astounding how really arrogant they’re being about this whole process.

TEDDY DOWNEY: And if you take another step back ‑‑ and your book goes through a lot of this history. There’s market power. There’s acquisition. There’s cheating. There’s sort of scheming. Beyond breaking up companies. Sensibly, you could go back and DOJ or FTC could have blocked the double-click merger.

But a lot of this is ‑‑ and you mentioned this in the gaming ad market ‑‑ there’s a lot of cheating going on. There’s a lot of scheming. We’ve written a lot about this company. There’s a lot of privacy problems, data control problems. The trade desk we’ve written about, and other ad tech companies with sort of – I would call them less than scrupulous ‑‑ data protection policies.

What kind of laws or rules could be in place? You mentioned Wall Street is more heavily regulated. Which, by the way, I don’t get any sense that Wall Street is the most regulated place in the world by any stretch of the imagination. But obviously, some laws are different from none.

What kind of rules or regulations do you think the government should or could explore so that it’s not just totally the wild, wild west and companies like Google regulating the market as they see fit, as opposed to a functioning, competitive market with just some basic rules, like you said, on Wall Street to, I don’t know, prevent conflicts of interest or whatever type of thing you think would address some of the privacy and cheating problems?

ARI PAPARO: Sure. Well, on privacy, I think it’s just astounding that the Congress has not made any effort to create a national privacy bill. The industry wants a national privacy bill.

The industry doesn’t want state-by-state privacy. They want a single bill that is sort of GDPR light, less regulatory overhead than GDPR, but pretty straightforward requirements to consumers around privacy protection, et cetera. The industry has wanted this for going on ten years and it gets nowhere in the Senate and it’s really shocking that it hasn’t happened.

On the cheating and stuff like that, it’s a little trickier. To my knowledge, there hasn’t really been a strong effort in Congress to do anything in that regard. It could be, there definitely could be, it could even be self-regulatory, but that hasn’t happened either, where companies are required to run fair options with defined capabilities and also give their customers full transparency of what happened, so log-level data.

One of the things that came up in the trial that didn’t really make it in my book is that Google would stonewall people who wanted to get log-level transparency about the auctions in their accounts because they said there was privacy issues. They would just use every excuse in the book not to give people raw data. Because if you had the raw data, you would have been able to see that the auctions weren’t working the way you thought they were supposed to be working.

So, that would be something. The only thing that’s been going on in Congress is the America Act, which I think is Amy Klobuchar and Mike Lee, I want to say, and they’ve brought this up a couple years in a row. And effectively, it’s a Google breakup act. It requires, based on the way it’s written, Google to spin out its network business. It’s pretty simple, really.

There’s also an advocacy group called Check My Ads that some of you may have heard of in the D.C. circle. They just published their manifesto for how the industry should be regulated, and they call for log files as well. And they also call for the conflict of interest, that you shouldn’t be able to be both a buyer and a seller, or depending on how you think about it.

So, those are all really interesting ideas, but the reality is, even in Europe, this is just totally unregulated other than privacy. And it’s buyer beware about what you’re buying, how much you’re paying for it, and how it’s being measured.

TEDDY DOWNEY: Are there any companies, innovative companies, that you see that are kind of innovating in a pro-privacy or an honest way, like sort of not cheating. Or are there any honest businesses that you see as innovative, that are like, hey, these are something to watch, these are models to watch?

ARI PAPARO: Well, I think you kind of threw the trade desk under the bus, which I don’t think was deserved.

TEDDY DOWNEY: Yeah, we’ve done reporting on them in the past. I mean, the GDPR violations, things like that.

ARI PAPARO: Yeah, sure. Privacy-Enabling Technologies, or PETS, are still pretty early and nascent. The 180 that Google spent on the sandbox really kind of threw a wrench in a lot of the PET development, because it went from being super urgent to being much less urgent. And so, most of the companies that are doing that have exited for smaller amounts.

So, we saw the WPP, the holding company, bought InfoSum, which was a leading PET company. And then, last year LiveRamp bought Habu, which was another kind of clean room data collection company. And Snowflake offers this data.

So, these sort of like privacy-preserving technologies used in ad tech are becoming pretty mainstream, but it hasn’t produced a winner, hasn’t produced a big public company on its own. And then in terms of like ad tech, well, I think most of the large ad tech companies are not cheating. I use the word in quotes, “cheating”.

But it’s so hard to tell. It’s so hard to tell. Because basically, you run some ads. You sell some products. And then who takes credit for it is a jump ball, right? So, that’s the nature of the business.

TEDDY DOWNEY: And this has been such a fascinating conversation. You are such an incredible expert. The book, “Yield, How Google Bought, Built, and Bullied Its Way to Advertising Dominance”. Can you tell us a little bit more about your podcast and your newsletters so we can all go sign up for them? I’m excited to get more involved in the podcast universe.

ARI PAPARO: Yeah, so I run the company. It’s called Marketecture, like architecture with an M at the beginning. And we publish a bunch of newsletters and podcasts. I personally publish a podcast called the Marketecture Podcast, available anywhere you listen, every Friday. We have an interesting guest and talk about kind of the advertising news.

And then I also write a newsletter that comes out every Monday, which is analysis of the latest and greatest going on in advertising. You could go to marketecture.tv to access all of that stuff.

We also have a policy newsletter, which might be interesting to this listenership, called the Monopoly Report. So, I don’t write every episode of that. I have a colleague, Alan Chappelle, who’s a real expert in privacy and policy stuff. So, you can subscribe to that as well. It’s all free.

TEDDY DOWNEY: All right. Well, I hope you’re getting good ads, honest ads, good revenue on those newsletters.

ARI PAPARO: Yeah.

TEDDY DOWNEY: Thank you so much for doing this. The book is amazing. And thank you to everyone else for joining us today. This concludes the call.

ARI PAPARO: Thanks, Teddy.

TEDDY DOWNEY: Bye, Ari.