Awin is the all-in-one platform for smarter partner marketing. With award-winning technology, expert support and a global network of 1 million partners, Awin helps brands discover, manage and scale high-performing partnerships that drive measurable growth. Learn more.

Marcus Johnson: [00:00:00] Today's episode is sponsored by Awin, the world's leading brands grow through partnership marketing with Awin's enterprise technology, AI-powered insights, and expert US regional support. With 30,000 advertisers, one million partners worldwide, and more than $20 billion in partner-driven revenue generated last year, Awin helps brands turn partnerships into growth.

Learn more at awin.com/us or us. Same thing.

Nate Elliott: Join us for EMARKETER's Future of Digital Summit on September 15th in New York City for a full day of research, expert analysis, and executive conversations on things like AI visibility, commerce media, and the future of marketing. Click on the registration link in the show notes for tickets.

Marcus Johnson: Hey, gang. It's Monday, August 3rd. Nate, Marissa, analysts, welcome to Marketing Podcast made possible by Awin. I'm Marcus. Joining me for today's conversation we have two normally New [00:01:00] York-based people. One of them is on the move. Uh, he's where he sometimes lives, principal analyst for AI, Nate Elliott.

Nate Elliott: Good to be with you.

Marcus Johnson: Hello there. Joining us from North Carolina. And, uh, we do have someone from New York, uh, still there, I believe, unless she's on the move, too, uh, newsletter- I'm

Marisa Jones: actually on the move. I am actually on the move right now, actually.

Marcus Johnson: Are you really? Where where, where

Marisa Jones: are you now? I'm in Ohio right ... I'm in Ohio right now.

Marcus Johnson: In Ohio, okay. Uh- At the moment, yeah ... two people running, apparently running from the law, newsletter analyst Marissa Jones.

Marisa Jones: Hi, thank you for having me.

Marcus Johnson: Welcome, welcome. Uh, today's fact.

Why f- why people are fleeing New York. Maybe that should be today's fact. Uh, but it's actually about the mammals, uh, who sleep the most and the least. Would you guys care to venture a guess at what mammal might sleep the most? Me. That was a little- It's me, for sure it's me Nate, you are on the list, but you are third.

Nate Elliott: Okay. [00:02:00]

Marcus Johnson: It's not- Uh- I don't know if this is one. It's one that you might associate with someone... A, a mammal that sleeps weirdly, in a weird position.

Nate Elliott: Hmm. Are, are sloths mammals?

Marisa Jones: That's what... Yeah, right?

Marcus Johnson: Oh.

Marisa Jones: Oh, are they?

Marcus Johnson: I would venture yes, and they sleep a lot, but they don't sleep the most. Are they on this list of top 10?

No. Okay. But they do sleep a lot. It is- I don't

Nate Elliott: know ...

Marcus Johnson: the brown bat. Ugh, okay. Yeah. The brown bat sleeps at least 20 hours a day. Amazing. How are bats

Nate Elliott: mammals? That

Marisa Jones: sounds wonderful.

Nate Elliott: Really, how are bats mammals?

Marisa Jones: Yeah, I di- I would not have guessed that. What's going on

Nate Elliott: with this system?

Marcus Johnson: Should be the rest of the real question.

Although technically, it's kind of strange, 'cause a lot of it is torpor, which is like a form of hibernation, where your bo- body temperature, heart rate, like slows down- Mm ... metabolic rate to conserve energy. So i- technically, it's not really sleeping, but it's still kind of sleeping. [00:03:00] Um, and apparently flying requires massive amount of metabolic energy, so sleeping for 85% of the day lets bats conserve energy between their intense short hunting windows at night.

Are they flying further than other things that they need to sleep this... That does not- I don't know ... sounds like a lot of sleep ... but I, I, I

Nate Elliott: understand the, the idea that flying requires massive amounts of energy.

Marcus Johnson: All right, fine. 'Cause the next thing is a, um, a opossum and an armadillo. They're in the close second.

Marisa Jones: I would have not guessed any of those, so.

Marcus Johnson: No. And then the human being- I mean, really

Nate Elliott: the question could be, guess three strange things that are actually mammals.

Marcus Johnson: That's true. Tiger makes the list, 16 hours. Um, that sleeps the least, uh, the giraffe. About- Hmm ... two hours a day. That's not enough. That can't

Marisa Jones: be good

Marcus Johnson: for them.

That doesn't sound fun.

Nate Elliott: But then- I mean, it's gotta, it's gotta be a big risk. It can't. If a giraffe lays down and goes to sleep, the chances of something running over its neck are [00:04:00]

Marcus Johnson: unacceptably high. So, well, this is why they don't... Yeah, that's one of the reasons they don't sleep much, is because of predators.

Um, and it's hard for them to get down and back up again. Um, they... The other reason is because they're such large herbivores, uh, they have to spend about 75% of the day, uh, chewing leaves and grass just to get enough calories. Um, so they don't have as much time, uh, as they might like to to sleep. But they, they don't really sleep, they, they power nap.

It's apparently three to five minute bursts scattered throughout the day and night. Oh.

Nate Elliott: Wait, so they're not laying down. Are they propping themselves against trees? What, what's

Marcus Johnson: happening? They can rest standing up. Um, but they, if they're gonna, uh, really like deep sleep, then they do... They kind of... It's like a yo- like a seated yoga pose, and then they fall asleep.

Marisa Jones: That's kind of cute.

Marcus Johnson: That is quite cute. Giraffes do yoga. Might have made that up. Giraffes do yoga to, to sleep at night. Uh, anyway, today's real topic: the three big questions surrounding Google[00:05:00]

All right, let's set the table. How have Google been doing? Alphabet, I should say. Q2 2026, Alphabet's total revenue was $120 billion. That's up 24% year over year, stronger than the 14% last Q2. 68% of that money comes from ads, which made 82 billion, up 14% year over year, an improvement on the 10% growth from last Q2.

And YouTube made 11 billion. They grew 13% in Q2 year over year, same as last Q2. So that's how they've been doing, but we're here to talk about the big questions surrounding the company at the moment. Nate, what is one of the big ones for you?

Nate Elliott: Well, the big question that investors seem to be asking after they issued those results at Alphabet is, is the company spending too much on AI?

Mm-hmm. Um, the stock got punished pretty aggressively for CapEx spending.

Marcus Johnson: Mm-hmm.

Nate Elliott: Most of it associated with [00:06:00] spending money on various AI projects, whether it's chips or services or consumer devices and services. And i- uh, frankly, it confuses me because This is a company that's doing remarkably well in AI.

Uh, they started, uh, Gemini more than a year after ChatGPT hit the market. But on that earnings call, they announced that Gemini now has 950 million global monthly users, putting it behind only AI mode, another Google product, and ChatGPT, both of which have announced about one billion monthly global users in terms of the largest AI platforms in the world.

And when Gemini hits a billion users, presumably within a few months, Google will have three separate one billion user AI consumer products, Gemini, AI mode, and AIO reviews. At the same time, they're selling their TPUs. [00:07:00] They're making a ton of money on Google Cloud services because of AI. They've got hundreds of billions in backlog waiting to be serviced all around AI, and, uh, yet their decision to spend a little bit of their cash reserves, uh, seems to have been punished by the market who don't want them spending that much on CapEx, and it just completely confuses me.

Marcus Johnson: Mm-hmm. So same. But what are your thoughts on this idea that, um, this came from a, a few people, but Deborah, Mary, Sophia, and Kendrick Kai of Reuters saying, "During the question and answer portion of the earnings call, multiple analysts pressed CEO Sundar Pichai for clarity around whether Google could keep up, uh, with its rivals at the frontier of model development."

Uh, and then you had Megan Broski and Tina Lee of Wall Street Journal saying, "In recent months, Google has been trying to catch up in the AI race. It released Gemini 3 in November that was competitive with top offerings from OpenAI and Anthropic, but since fallen behind them. New York Times reporting in July, Meta released a model that crept ahead of Gemini's [00:08:00] capabilities according to several leaderboards."

Is there any-- how much weight do you attribute to that concern?

Nate Elliott: Well, on the consumer side, the, the quality of the model has almost nothing to do with how many people use the model.

Marcus Johnson: Right.

Nate Elliott: Right. The reason Gemini and Google's other AI services are doing so well is because, yeah, the models are, are good enough.

In some cases, the, you know, on various benchmark tests, they've been the best. Uh, but that situation ebbs and flows, and they're not always going to be the best. The question is, is it good enough that it's giving people the kind of responses and information that they're looking for, and is it accessible?

And Google does really well in both of those regards. Mm-hmm. It's good enough both because the models are good enough, but also because Google has access to the data, information, and behaviors of the billions of people who use their search engine every month, who use their, uh, browser, their mobile operating system, their email service, and lots of other Google products.

And that amount of data allows them [00:09:00] to personalize their responses in a way that takes an already good model and makes the responses even better. But also, those billions of users on all those other services, it's, it's really easy for Google to remind them, "Hey, we've got an AI product over here that can help you with what you're doing," and nudge them in that direction.

Marcus Johnson: Yeah.

Nate Elliott: On the enterprise side, i- it matters more, I think, to have the best model. But I also don't think that most enterprises need frontier models for most of what they do with AI. I mean-

Marcus Johnson: Mm ...

Nate Elliott: certainly there are some instances where you want the best, most powerful model working on your task, but the vast majority of what most companies are doing with AI doesn't require state-of-the-art models.

Yeah. It requires something useful- Especially

Marcus Johnson: the free ones.

Nate Elliott: Yeah. It requires something useful and integrated with everything else you do. And, and Marissa, I'm sure you're, you have some thoughts and you're seeing some stuff in this space as well.

Marisa Jones: Yeah, absolutely. I think the big thing for me is there are these investor concerns about AI spending, but this is, correct me [00:10:00] if I'm wrong, Google's second quarter of over 100 billion in revenues.

So I think their revenues and their growth is strong enough that it kind of, to me, disproves those concerns a bit- Mm ... or at least makes it that I don't think they've really materialized yet, um, in terms of how AI spending and their massive CapEx is kind of affecting their business, or these claims that it's going to meaningfully affect Google.

Yeah.

Marcus Johnson: Yeah.

Nate Elliott: It's their 12th straight quarter of double-digit revenue growth. Mm-hmm. You know, unlike a lot of their competitors, the CapEx spending on AI is working. Yeah, I agree. When, when Meta got punished last year because they were spending too much on AI and hadn't even launched a usable consumer-facing AI product, that made sense to me.

They were spending almost as much as Google, and there was no product to show for it. Google's spending a lot of money, yes, but there's so much to show for it in terms of- Yeah ... coming from behind to being neck and neck with ChatGPT on the consumer [00:11:00] side, in terms of building an incredibly strong enterprise business, and also selling the chips that help power all of this.

Marcus Johnson: Mm-hmm. Yeah, th- I mean, Nate, to what you were saying before, um, most people I don't think are gonna know the difference in how powerful the models are, and if you get 80% of the leading models, 50% of the leading models, if you have 10X the, 20X, whatever is the scale, and you can put that in front of people, that's what is really gonna matter.

Um, $15 billion increase in CapEx spending plans for 2026. Google now in the 195 to 205 billion range for CapEx this year. Um, Brian Mulberry, who manages a portfolio at Sax Investment Management, uh, that includes Google stock, was saying the $200 billion was the, apparently the do not cross line, um, which led, or apparently led to its stock being hit, and, uh, uh, uh, investors expressing frustration about, um, what they said Google's lack of explanation for how [00:12:00] its AI spending plans would translate into more revenue for the company.

Nate Elliott: That's absurd to me. I- I mean- Yeah. Tell me ... how dare they spend an extra $15 billion in CapEx to service a half trillion dollar backlog for its, for AI services? Mm-hmm. I mean, the, the ROI on, on that is, is tremendous. They've got something like $500 billion in, in backlog of- Yeah ... cloud services waiting- Shocking

to be addressed, and they need to spend a little more on AI CapEx to get there, then that seems pretty reasonable to me.

Marcus Johnson: Yeah. Um, pr- probably wanna get used to it. Uh, no end in sight, it seems. AI giants expected to spend over 700 billion, only initially. It's probably higher than that. Um, at the start of the year, Morgan Stanley pegging the estimate at spend, uh, for CapEx for the AI giants at over a trillion next year.

Google saying already it's planning to spend more in 2027. Um, final thing [00:13:00] on this, then we'll move on. I also wonder, guys, how much it, uh, their stock was affected by the Chinese startup Moonshot AI. They released an AI technology that was said to be as nearly as powerful as leading American model Claude Fable 5 from Anthropic.

So maybe that gave the markets a bit of a scare as well. Um, Marisa, what you got?

Marisa Jones: Yeah, uh, so my question is, what's the next phase of YouTube growth beyond building their ad revenues? So we saw this quarter, YouTube's ad revenues grew 13%, which is still pretty strong, but we are seeing other parts of their business growing at a much faster rate.

But growth as video platforms fragment and budgets fragment a- across these platforms is likely to kind of moderate. So what's YouTube's next phase of growth? Is it potentially their subscriptions business, um, is something that they could be very heavily leaning on? Um, as part of the earnings release, Google Senior Vice President and Chief Business Officer, Philipp [00:14:00] Schindler, said that beyond advertising, their subscription business is growing at a faster rate than their- Mm-hmm

ad revenues. Um, we saw strong growth across Y- Google's subscription and platforms and devices business, um, which grew 19% year over year, but we don't know what percent of that is coming from YouTube subscription growth to things like YouTube TV and YouTube Premium. Um, so I guess the broader question is, are subscriptions YouTube's next phase of growth?

Marcus Johnson: Mm-hmm. Yes, this is a good one. YouTube, um, 11 billion in ad revenue in the quarter, as I mentioned, 10 in Q1. Uh, for the year, last year, 40 billion. It, it, it's so interesting, Nate, 'cause it, it, 2.6 billion people in the world use this product, and people are expecting the monetization to be higher given the scale.

Any thoughts here on, um, the next phase of YouTube's growth?

Nate Elliott: I mean, I, I've seen the chart that we published recently [00:15:00] about YouTube's monetization per user versus some of the other- Mm-hmm ... large global properties and, and how much more able they are to monetize their users. Um, it, it's a pretty remarkable gap.

Uh, and it's surprising to me in large part because YouTube has the most attractive ad opportunity of any of those platforms. Mm-hmm. It may not have quite as much data in terms of how people use YouTube, but of course, many of those YouTube accounts are associated with Google accounts, with people's use of Gmail and browsers and mobile phones and everything else, and that gives them great targeting information and targeting data.

But more to the point, advertisers have consistently, throughout history, been willing to pay more for TV style ads than for any other kind of advertising, and of course, that's what YouTube has to offer. So- It, it, it does surprise me that they're not better at monetizing those users, and I, I [00:16:00] wonder if we'll see them start to turn the dial slightly on allowing people to opt out of pre-roll ads, or perhaps not allowing as many people to opt out of as many pre-roll ads, uh, in terms of them pushing more aggressively users towards subscriptions if they do want an ad-free experience.

And of course, that's part of what happens right now. But, you know, there's not really a tax on not subscribing to YouTube because most of these pre-roll and interstitial ads you can, you can skip out of pretty quickly. Mm-hmm. Um, if they make that experience a little bit more onerous, they, uh, have the potential to make more money, either from the ads that people don't skip or from the subscriptions they buy to avoid the ads entirely.

Of course, the, the threat there is that they could make people angry and send them elsewhere. But, um, I honestly, for the kind of mid and long-form video content that seems to be YouTube's bread and butter these days, I don't know where else they're gonna go.

Marcus Johnson: Mm-hmm.

Marisa Jones: Yeah. [00:17:00]

Marcus Johnson: Barbara Oruttaye of the Associated Press saying World Cup advertising boosted Google's ad revenue, especially on YouTube.

Um, Google saying close to two billion unique global viewers watched videos related to the World Cup during the event, but that grew the same in Q2 this year as they did last year. Uh, so you didn't really see that bump i- in, in the numbers. Um, Nate, what else do you have?

Nate Elliott: Well, I, I'm curious. My, my perhaps biggest question, definitely not my most consequential question for Google, but my biggest question for them is, when are they gonna rationalize the naming convention around their AI products?

I- I've been amazed for a year or two now that somehow Microsoft has a better, more coherent naming system for their AI tools. Uh, everything is Copilot. Literally anything that's AI that has anything to do with anything Microsoft makes is called Copilot this or Copilot that. Uh, Google, in the meantime, has had NotebookLM and Gemini and AI [00:18:00] Modes and AI Overview.

They started with Bard. Um, they've been all over the map, and, uh, as a market researcher who has to ask people survey questions about which Google AI tools they use, it makes my life harder than it might be because I don't think people understand what is AI Mode and what is AI Overview and- Mm ... and what is, you know, a bunch of other Google tools.

Having said that, they've started perhaps the process of rationalizing this. So they announced recently that NotebookLM is going to be called Gemini Notebook. It's the first time they've used the same name on more than one AI product. It's the first time they've taken- Mm ... the Gemini name and applied it to something meaningful outside of the Gemini app experience.

And so I look forward perhaps to a future in which AI Overviews become Gemini Overviews and AI Mode becomes- Mm ... Gemini Mode, and I understand the internal organizational structures of [00:19:00] Google make that more difficult. But it's not like Microsoft isn't a tangled web of a company as well, and they've figured it out.

So I'd love if they just rationalized it, uh, convince people that everything is called Gemini, and were able to further strengthen their brand and perhaps their grip on consumers by doing so.

Marcus Johnson: Mm-hmm. It, it does seem strange they're taking so long to do that, but I do wonder because the vast majority of people don't really, um, know when they're using AI, having AI in the title helps now, and then transitioning to Gemini and people having that being synonymous with AI, um, is what their long-term strategy is here.

But it does seem a bit all over the map at the moment.

Nate Elliott: No, and, and it is a great point because their entire strategy is to convince people that they don't need to go somewhere besides- Yes ... Google to get AI. So I, I, I see the point in using the letters AI for AI [00:20:00] Overviews and for AI Mode. Having said that, there surely was and is some way that they could both stress the fact that this is an AI response or an AI experience, while also using a consistent brand name across the different things that they offer.

Marcus Johnson: Mm-hmm. Yeah, 'cause A- Apple started with Apple Intelligence, and then they've moved to Siri AI, and so I wonder if there is some of that, um, we want people to know that this is an AI product- Mm-hmm ... before transitioning them. But there's, there's, there's a lot going on, so they could do with, uh, consolidating to a certain extent.

Marisa, what else do you have for us? Yeah.

Marisa Jones: Um, my other question is, are AI and Cloud becoming Alphabet's main growth driver, and can that growth scale without putting too much pressure on returns? So search we still see as Google's core business, but Cloud is showing pretty significant growth compared with other segments of Google's business.

It grew over 80% to about 25 billion, [00:21:00] so very significant growth for a single quarter. Um, and that's the kind of acceleration that investors are kind of looking for to understand whether Google's growth mix is shifting toward a more diversified model, a more Cloud heavy, um, growth story. So I think that's a very interesting thing to watch.

Marcus Johnson: Mm-hmm. Yeah, I, I like this one. Uh, we think about these companies the way that we've always thought about them, and when you look under the hood, they are changing. They are making money i- in, in different ways. Um, if you look at, uh, Google, uh, they're making 68... Now I know the cloud and the adverti- uh, and AI has worked into the advertising, but, um, 68% of, uh, their money coming from Google advertising today, um, that was, uh, about 80% in about 2021, so that has changed.

Similar to Amazon. Amazon, uh, around the pandemic was making half of its money from online store sales, and now it's closer to a [00:22:00] third. Mm-hmm. And so these companies, as we know, Amazon is the r- online retail giant, is quickly becoming, uh, the company that makes money i- in a bunch of different ways. So I think this is a really good one.

Nate, any thoughts here?

Nate Elliott: Yeah. It's... I, I mean, I, I am continually impressed with how much Google's ad business is growing. The fact that they keep growing it by double digits quarter after quarter is incredible. Uh, I wrote the report about our search advertising forecast recently, and, you know, Google's share of the US search ad business is declining, but not because they're making less money on it, just because Amazon, as you just mentioned, continues to grow pretty quickly in terms of the amount of search advertising that they're selling.

But it, it, it's remarkable to me that as large and as mature a business as Google's search ad business is, continues to grow at a really good clip, but, uh, there's no way for a business of that size and maturity to be the fastest growing piece. And as Nerissa [00:23:00] said, um, the more important AI gets, the more important the cloud business looks for Google.

And we talked about that enormous backlog. Um, that all apparently is tied to cloud services, um, especially around AI. Yeah. So this is, you know, this is where we are likely to see a lot of growth and some pretty quick growth from them over the next few years.

Marcus Johnson: Yeah. Um, to your point, Nate, double-digit growth, uh, in 10 of the last 11 quarters, uh, for, for Google advertising.

A- and you don't have to go too far back in the history books to Q4 2022 and Q1 2023 when they were recording negative, uh, two quarters, back-to-back quarters of negative growth. Slight, but still negative. And since then, they've just been, uh, on a rocket ship, um, given how big, uh, the base they're growing from.

Uh- Just a quarter

Nate Elliott: trillion dollars a year.

Marcus Johnson: That's all. Um, all right, folks, let's pick a top three, consensus top three. So of the ones we've talked about, [00:24:00] Nate, what is your top, top question for Google right now?

Nate Elliott: I actually like Marissa's question about how YouTube's going to grow. Uh, I, you know, as an AI analyst, the, the key question for the AI platforms I cover is how much money are they, are they going to make from advertising?

How much are they gonna make from subscriptions, and what does that mix look like? And to, to have a good look at what that looks like for a larger, more established business, I think is really interesting on its own, and also ties in really well with some of the stuff that I'm thinking about on the AI side.

Marcus Johnson: Mm-hmm. Very nice. Marissa, how about you?

Marisa Jones: Um, I think my favorite question is the first one that Nate posed about Google's AI spending, how investors feel about it, um, and whether those concerns are really justified.

Marcus Johnson: Very nice. Very nice. And I will steal, um, Nate's second one, uh, because it needs to get answered at, at, at some point.

Um- Yeah ... will Google finally rationalize the AI product naming convention? Um, very good. Well, there are our top three. Thank you so much for joining me, gang. That's all we've got time for for today's [00:25:00] episode. Thank you, uh, to my fellow list makers. Thanks first to Nate.

Nate Elliott: Thank you very much, Marcus.

Marcus Johnson: Pleasure.

And to Marissa.

Marisa Jones: Thank you for having me.

Marcus Johnson: Thank you, thank you. And thanks to Lance, who's helping us out on the production crew with this one. Thanks to everyone for listening to Mind the Machine Market podcast, made possible by Awin. Suzy will be here on Wednesday talking back to school, and I'll be back on Friday discussing how AI is changing health wearables.