Google is Keeping Cookies, Maybe Selling Chrome, and Scaling AI Overviews | Behind the Numbers

Rokt helps marketers reach high-intent customers in the Transaction Moment™—when they’re actively completing a purchase online. Powered by AI and first-party data, Rokt Ads connects your brand with over 400 million global shoppers and delivers outcomes you can count on. Learn more at rokt.com/emarketer

to get started today.

Marcus Johnson (00:00):

In today's economy, every single ad dollar counts. That's why our performance marketers are pivoting to Rokt Ads to reach over 400 million active shoppers in the transaction moment when they're completing a purchase online. You only pay when customers engage. It's a pretty good deal if you ask me. Learn more at Rokt.com/emarketer. That's R-O-K-T.com/eMarketer.

(00:28):

Hey gang, it's Monday, May 5th. Yory, Jeremy, and listeners, welcome to Behind the Numbers, an EMARKETER video podcast made possible by Rokt. I'm Marcus, and today we'll be discussing the impact of Google keeping cookies, how AI overviews is getting on, and what a Chrome sale would look like. Join me for that conversation. We have two people. Let's meet. Then we start with our principal analyst and desk lead for the advertising, media, and tech teams. Based in New Jersey, it's Yory Wurmser.

Yory Wurmser (00:55):

Hey Marcus, how are you?

Marcus Johnson (00:56):

Hey fella. Very good, sir. How are you?

Yory Wurmser (00:58):

Very good.

Marcus Johnson (00:59):

Terrific. Whoa.

Yory Wurmser (01:04):

No, I mean, and yeah for Monday.

Marcus Johnson (01:05):

Not as good is Jeremy apparently who loves Mondays. He's our senior director of briefings living in New York. Jeremy Goldman.

Jeremy Goldman (01:11):

Hello there. Or should I say Hola and happy Cinco de Mayo? I would be happy even if it was not one of my favorite holidays.

Marcus Johnson (01:19):

Jeremy likes Mondays and me and Yory and everyone else on the call we're saying dial it back.

Jeremy Goldman (01:26):

Sorry.

Marcus Johnson (01:27):

It's too much.

Jeremy Goldman (01:27):

Not happening.

Marcus Johnson (01:29):

Anyway, today's fact, the most commonly eaten fish. I don't know if this is a good one. I'm going to be honest. It's either really good or God awful. You decide. The most commonly eaten fish by volume or weight in the world is what?

Jeremy Goldman (01:52):

Sardines?

Marcus Johnson (01:56):

Sardines. No.

Jeremy Goldman (02:00):

Cod?

Marcus Johnson (02:01):

Cod's a good guess. Sardines is a terrible. Sardines? Do people eat those?

Jeremy Goldman (02:06):

Yeah, I mean in some places. Portugal loves them.

Marcus Johnson (02:12):

Tuna.

Jeremy Goldman (02:14):

I thought it was like a trick question. That would've been the first thing I would've said. Aw, man.

Marcus Johnson (02:18):

By value, salmon is the largest single fish commodity in the world, but by volume or weight, it's tuna. In the US salmon's one, tunas two buy per capita consumption. Where are all these fish being caught? Good question.

Jeremy Goldman (02:35):

The water.

Marcus Johnson (02:36):

It is in the ocean. Nearly 60% of all oceanic wild caught fish are captured in the Pacific, 60%. 25% from the Atlantic, 15% from the Indian Ocean and a few from the Southern and the Arctic. That said, Pallavi Rao of the Visual Capitalist was noting, that we now farm more fish than we catch, but wild court fish still make up about 49% of the world's fish production. Is that a good one? Probably not.

Jeremy Goldman (03:11):

Did you fact check that because it seemed a little fishy?

Marcus Johnson (03:17):

Just when you thought it couldn't get worse.

Jeremy Goldman (03:19):

Sorry.

Marcus Johnson (03:21):

I can save the day with another fact. I brought a backup because I thought that would be terrible and I was right. The Smithsonian Ocean notes that there are an estimated 3 million shipwrecks scattered across the ocean floor according to the United Nations Educational, Scientific, and Cultural Organization. You're welcome fish. We're the worst. Anyway, here's real topic. Google cookies, Chrome, and AI overviews. So lots to talk about with regards to Google. We start by talking about cookies. "Google recently said it will keep cookies in their current form on its Chrome browser and will walk back a previous commitment to build a simple opt-out tool for users who don't wish to be tracked," writes Kendra Barnett and Lucinda Southern of Adweek. Jeremy, I'll start with you. What's your biggest takeaway from Google saying actually after five years, however long, decades, we're keeping cookies in their current form?

Jeremy Goldman (04:28):

Yeah, I mean this is obviously a really big deal because this is something that has been, I would say, brewing for how many years at this point, right?

Marcus Johnson (04:36):

Yeah.

Jeremy Goldman (04:36):

It's interesting that I think on one hand you can make an argument that Google's tried to be responsive to advertisers who felt a little bit concerned about the new model and Privacy Sandbox in general, but also at the same time, is this necessarily good if you've had years to figure this out and develop a roadmap and get people rowing in the same direction and on the plan that you thought was better for you as Google and Chrome and the internet in general and the advertising ecosystem, and I think that this is kind of a clear admission that note, there were still a lot of things to be sorted out and not everybody was really comfortable with Privacy Sandbox moving forward as the de facto way of advertising operating on the internet.

Marcus Johnson (05:29):

Yeah.

Yory Wurmser (05:31):

No, I mean I think it's good news. I agree with Jeremy. I think it's good news for advertisers because having the signal of cookies, it's just the best. It's better signal than you would get without it for advertising and the targeting. On the flip side, the majority of internet visits, paid internet advertising is still not going to have access to cookies, so all the movement towards alternative identity solutions is still going to be necessary even with cookies still remaining.

Marcus Johnson (06:07):

Yeah, that's the big part of this, isn't it? There was a quote from, exactly what you're saying, Yory. Eric Seifert, I'm saying that right. A leading media strategist was saying, "The privacy environment can only get more restrictive since the industry is moving away from deterministic IDs. Even if third party cookies aren't going anywhere in Chrome, the marketing ecosystem is moving or evolving away from deterministic identity." There's still quite a lot of people using them though. We had this chart that said one in five Americans always opted into cookies when given a choice according to a study we did last year, but if you add up the people who said always, often, or sometimes it's 80% of people, so there's still definitely a factor. I thought this quote from that Adweek piece was quite fitting, kind of sums up how a lot of people probably are feeling given this news. "Digital ad ecosystem, which has invested significant resources and manpower to develop cookie less ways to track and target audiences and measure media performance responded with a mix of shock, exasperation, and relief."

Jeremy Goldman (07:08):

Yeah, there was actually some data that we cited by Supermetrics in a recent briefing to clients that I thought it was really interesting. Marketers foresaw that with the decline of third-party cookies, 57% believed that they would see a decrease in effectiveness of targeted advertising. And to Yory's point before, I mean that's a significant number when you think about how many dollars are thrown into this, and you can just imagine if you're Google, do you really want to risk having less people throw money at a channel that you largely dominate because they don't think that the targeted ads are going to necessarily have the same ROI?

Marcus Johnson (07:51):

Yeah. What's the real impact here? Because this piece, it wasn't Adweek. Yeah, they split out kind of the winners and the losers from this. They were saying some of the winners, a lot of winners, The Trade Desk, [inaudible 00:08:02], BoxMedia, et cetera, et cetera. They had a lot of folks, a lot of different groups. The biggest losers from this, they say Google, reputational damage, privacy advocates, third-party cookies still exist and consumers losing privacy protections that Google had promised. But is this just going to be a situation where Google said, actually we changed our mind. We're not doing this. All parties kind of walk away from the table and there are no kind of ramifications. I guess Google didn't have to do this. Maybe it's better to say actually we've tried it for however long and it's not something we want to move ahead with just because we started it. Does Google suffer much as a result of this at all?

Yory Wurmser (08:38):

I think Google, it's a relatively smart move for Google to do right now for a couple of reasons. One, they're facing antitrust issues. Part of what happened in Europe is that Apple has been fined for its app tracking transparency, its version of taking away identity and providing an alternate identity. By taking away cookies, their alternative was the Privacy Sandbox and that is something that would likely have given them an advantage. So from an antitrust perspective, it makes a lot of sense.

Jeremy Goldman (09:17):

But at the same time, and Yory and I, we were kind of debating this a little bit recently. I can see it both ways in the sense that it also helps shine a light on look at the relative market power of Google, that they can do this move and everybody says, wow, this has a major implication for the advertising ecosystem. Does it show maybe that they're a good actor in some ways? I guess, but if I was them and I saw the writing on the wall, it would've been more fortuitous to try to do a move like this maybe six months ago. That's all.

Marcus Johnson (09:53):

Yeah.

Yory Wurmser (09:54):

I mean, they've gone back and forth so long that it's been handled incredibly poorly. No doubt. So in that way, I agree with you Jeremy.

Marcus Johnson (10:03):

Yeah, there's been a lot of antitrust talk. We covered the ad tech monopoly, so two big ones. One is the ad tech monopoly antitrust trial. We covered that in an episode that came out last Monday, April 28th. The other, the search monopoly antitrust trial is in remedies phase. Google said it will appeal once the remedies portion is finished because they lost that case. They have been called an illegal monopoly by Judge Amit Mehta from last year. One suggestion, one of the remedies suggested from the DOJ is that Google divest from Chrome, it's web browser. Raman Jones of the BBC noting that ChatGPT maker OpenAI would be interested in buying Chrome if Google was forced to sell it. So is AI company Perplexity, they're interested and internet staple Yahoo as well. Yory, what would the search world look like if Google was forced to sell Chrome?

Yory Wurmser (10:58):

All of these revenues are going at really two things. One is Google's built-in advantage in data that just makes its algorithms work better and it also, the fact that they've created these, Google is a default search engine across the internet. By taking away Chrome, both those things are seriously reduced. Chrome is just a really strong source of signal for Google, not just in advertising but in search functionality. It along with the other remedies, could just totally alter search by giving other search engines the chance to be the default engine, to work better, and just to have just a lot more visibility and functionality than they do now.

Jeremy Goldman (11:44):

I do. Yeah, I mean I fully agree with all that and because of that, I am not going to just echo what Yory just said. I really think that if you look at the overall evaluation of Chrome, and if you look at the fact that there are two cases going on, one of them tied to, for lack of a better term, like search and then the other one more so on ad tech, I think it makes sense to see a bit of a compromise on the ad tech side that leaves some of Google's consumer search business more whole in part because how many people out there can reasonably credibly purchase Chrome? It's a very high risk move for anybody who could actually buy it, but there are very few people. Well, simply because of the price that it's going to command. Right. You can have somebody who drives up the price simply because they're not going to buy it, but they know that you'll pay a bit more, right?

(12:38):

It is more difficult to get a strong ROI when you're paying out $50 billion with a B, which is what some people think that Chrome could command on the open market. And also that part of the business is what Google would like to hang on to. The ad tech business is lucrative, but it's still a fraction of Google's overall revenues, so they would probably prefer to give that up as a bit of a compromise in order to keep the consumer search business intact.

Marcus Johnson (13:07):

Yeah.

Yory Wurmser (13:09):

It's not just about search though, it's also about AI and Google is moving in the same direction with Gemini, it's AI chatbot as it had done with search and trying to lock it in all over the internet. And without the data from Chrome, without that default engine, it puts it behind, it hinders its efforts to become the leader in search that it had become. And I mean, the leader in AI that it become in search.

Jeremy Goldman (13:38):

Absolutely. It's more like if you took the AI out of the picture, they would really want to hold onto Chrome, but when you layer what you just said, they really want to hold onto Chrome even more.

Marcus Johnson (13:48):

It also seems like OpenAI, ChatGPT maker putting their hand up and saying, Hey, we'll buy it. Dan Primack of Axios was noting wouldn't this just be solving one antitrust problem to begin building another? Because they have, I mean, there's some data here from Similarweb. Global monthly web visits to AI chatbots. ChatGPT close to 300 million, second place DeepSeek and Gemini about six times fewer, and DeepSeek is going in the wrong direction. ChatGPT is going up and up and up. So a lot of the problems with these antitrust cases as we talked about in the episode, is you have good intentions, but you end up kind of breaking things off and that just kind of forms, whether you sell it or whether you just break it off, that ends up forming its own monopoly elsewhere later, in like 20 years or so or longer.

Yory Wurmser (14:43):

I think-

Marcus Johnson (14:44):

[inaudible 00:14:44]. Or yeah.

Yory Wurmser (14:44):

Sorry.

Marcus Johnson (14:46):

Please. No, no, no. Please. Please. No, no, no, no. Hop in.

Yory Wurmser (14:47):

I mean, you're making Google's argument because part of what Google has been saying in the remedies trial is that their lead in search is significantly less than it was a year ago or as the trial was actually been argued, that ChatGPT's increased traffic and increased presence. Within the chatbot space it's really merging and overlapping with search and that's kind of Google's argument right now, why we don't think these remedies make sense.

Marcus Johnson (15:23):

Yeah.

Jeremy Goldman (15:23):

I mean it is a tricky thing to say it would've made more sense to break us up a few years ago, which I know is not exactly their argument, but just because again, I'm trying to make a bit of the counterpoint to yours, just because it made more sense to try to break up or have some remedies against Google a few years ago doesn't mean that it doesn't make sense now. It is very true that somebody benefits and then a few years from now you could very well see a situation where you have an antitrust case against somebody who benefited in the same way that Google benefited from the Microsoft case from the late-90s. Right. But it doesn't mean that it's not necessarily worth doing from the government's perspective, even though it does to your point, create an issue that has to be dealt with later.

Marcus Johnson (16:14):

Yeah. So Google's AI overviews now reach over 1.5 billion people every month, Google just said. Despite that, Mark Stenberg and Paul Hiebert of Adweek are noting that one year in Google's AI overviews are not a chronic condition for publishers yet. Yory, should Google be happy with the impact AI overviews have had on search so far?

Yory Wurmser (16:37):

I think that they probably are pretty thrilled by it so far. I mean, with their end earnings, we did not see a meaningful dip in search ad revenues. To the contrary to did pretty well. It means they're monetizing AI overviews well. They're able to monetize it really equivalently to this point, they were with regular search and that was the real danger, that AI overviews and generative search would cannibalize its existing business model and so far they've been able to navigate that pretty well and have avoided that happening.

Marcus Johnson (17:13):

Yeah. Yeah. 10% growth in Q1 for the search piece of Google's ad business. Jeremy, what's your take here? Because there was some numbers. Actually, Marissa Jones who's on your team was saying that they, actually AI overviews appear to be affecting click-through rates as she was writing that a new, is it RFs study, I think it's how you say it, found that Google AI overviews and search results significantly decreased click-through rates by 34% for top ranking pages compared to similar informational keywords without AI overviews.

Jeremy Goldman (17:47):

Yeah, so one thing that I think is really important for us to, like we talk about publishers like they're a monolith and the truth of the matter is there are different types of publishers and we've looked into this and not just the RF study but in few other places, and it turns out it depends on what kind of thing you're publishing, for lack of a better term, like a world encyclopedia for instance, that somebody has the fact that has been surfaced through an AI overview. You need no additional information and then you're not clicking through there. But then there are other types of publishers that somebody very well might click through at a more consistent basis. So this is something that Google has to get right, because there have been other instances where Google has, let's say starved a publisher to some degree from revenue and traffic, which then keeps that publisher afloat and obviously Google thrives when it has a vibrant publisher ecosystem to pull data from, right?

(18:47):

They don't actually want to necessarily fund all of that themselves. They want publishers to be creating great things that then they can turn them into AI overviews. So it's a really tricky balance where they have to keep their ad revenue going up higher and higher year after year as a public company, while at the same time making sure that these publishers, I don't want to say they have to necessarily thrive, but they certainly have to stay in business in order to keep feeding the beast that is Google, so to speak.

Marcus Johnson (19:17):

Yeah. Yeah, these numbers didn't help the antitrust case at all. "Google might be threatened with a breakup after losing two antitrust cases, but in the meantime it can console itself with piles of money," joked David Streitfeld of The New York Times. Google hit $67 billion in ad revenue in Q1, up around 8% year-on-year solid, but down from a 13% growth from the last Q1. Jeremy, which part of Google's Q1 performance did you find most interesting and why?

Jeremy Goldman (19:53):

So a warning, this is a bit of a cheat, but bear with me. I think part of it, it's the fact that its business model is starting to look like a very well oiled flywheel. The Cloud business, which we don't talk about as much, is still growing at a pretty rapid rate even though there's been some deceleration there overall year to year. And all of this appetite for AI that Google is driving is actually helping those Cloud revenues go up. But also all of this investment in AI is creating new ad products that people are investing in, not to mention Gemini and all of the impact on workspaces. So I think the very fact that all of these different components are working very nicely together is something that stuck out to me a bit more this quarter.

Yory Wurmser (20:44):

Yeah, I think you're right. And their AI is really, they're succeeding with AI much more so than they were a half a year ago. Gemini is on objective benchmarks is doing well. They're gaining users pretty quickly, still behind ChatGPT, but they're doing well in that way and they're still investing heavily in AI capital expenditures. They haven't really slowed down as some of the other companies like Microsoft has done. So their AI business is really, really doing well. I mean, they still have a lot of work to do to ward off, fend off OpenAI, but at the moment they're doing pretty well.

Marcus Johnson (21:25):

It helps that they are insanely profitable because as soon as your margin starts to go down, investors start questioning why are you putting so much money into all these bells and whistles when the core product's not churning out the kind of profit that we're hoping for. They made $35 billion in profit in Q1, which is up nearly 50% year-on-year. They were saying much of that is from equity investments, not ops, but it's still money, so it's not bad. It did seem that there was a kind of noticeable slowdown though the year-on-year. Ad revenue overall, as I said, went from 13% growth to eight. Search piece, which Yory was talking about earlier, went from 14 last Q1 to 10, this Q1 percent growth and YouTube went from 21 down to 10. I mean, how concerning is that? Is that any indicator of things to come, do you guys think?

Yory Wurmser (22:22):

I think it's actually relatively good news. Everyone has been expecting some slowdown in the beginning of this year just because of business uncertainty.

Marcus Johnson (22:31):

Yeah.

Yory Wurmser (22:31):

This is Q1, so this proceeds to giant disruption caused by those April 2nd Liberation Day tariffs. But even so there was a lot of uncertainty. It's a sign that ad revenues and therefore business was still chugging along at a pretty good rate in first quarter, even if it wasn't quite as strong as it was last year, year-over-year. So I think it's generally good news, but going forward Q2 and everything, it's a whole different story.

Marcus Johnson (23:02):

Yeah. And Jeremy, we had forecasted a slowdown. Our forecasting team had expected four year growth for Google's ad revenues to be 12% last year, 9% this year. So we were baking in some of this.

Jeremy Goldman (23:13):

Yeah, exactly. We were baking in some of this. Also on top of that, Google's really good at figuring out what to guide people to and then how to beat it a little bit, and when they're a bigger company than most, that looks quite good. I think it's also important to note that in Q1 tariffs were a lot of noise, not much bite, but the real test probably does lie ahead for Google. I think they were even asked that on their call. Tariffs didn't really dent their Q1 performance at all, but the real test is going to lie ahead, just like Yory was saying. Some independent data from IAB said that budgets are going to really be impacted the most, ad budgets in Q2, that that's where some of the greatest pain is going to occur. And not just for Google, but a lot of other companies. And we have already seen pullbacks like companies like Temu and Shein pulling back their Google shopping and their meta ads, so.

Marcus Johnson (23:13):

Two of the biggest advertisers.

Jeremy Goldman (24:13):

Yeah. So you're going to expect to see a lot more of that, that we might be guided towards a little rougher path ahead for Google and the rest of the sector.

Marcus Johnson (24:24):

Yeah. Yeah, both of those companies being Chinese and this growing trade war between the US and China and they're not helping things. Final thing for me on the performance, one thing, I mean all positive across the line items for top line Google Ads, for search, Cloud doing well, for YouTube, all that stuff, but Google Network piece of Amazon's business still making about $7 billion a quarter, so not terrible, but they've been falling by a few percent year-on-year for nearly 12 straight quarters now, three years. So something to keep an eye on. All right, that's all we've got time for, for today's episode. Thank you so much to my guest for hanging out with me today. Thank you to Yory.

Yory Wurmser (25:01):

Always great to be here.

Marcus Johnson (25:02):

Yes, sir. Thank you to Jeremy.

Jeremy Goldman (25:04):

Fantastic as always, sir.

Marcus Johnson (25:06):

And thank you to the whole editing crew, Victoria, Lance, Danny, and John who posted a clip of me saying that he's my favorite to the whole team, the scandal. Thank you. Stuart runs the team. Sophie does our social media. Thanks to everyone for listening to Behind the Numbers, an EMARKETER video podcast made possible by Rokt. Make sure you subscribe and follow and leave a rating and review if you could please. They mean the world to us honestly. Sara will be discussing the attack of the private labels on Wednesday, and I will see you hopefully on Friday.