Google Just Lost $5.9B… Or Did It?

Verbatim. Paragraph breaks added for readability; nothing else changed.

Yeah, first episode of the podcast. So, what we're going to be doing is we're going to be talking about some certain news stories. I've collected or gathered all of this. Julian has no clue what's in the next slide, so it'll be interesting. Could be AI station or anything, mate. Yeah. AI girlfriends.

So the first story is 5.9 billion loss. What's your thoughts on this one, Julian? So Google's cash flow has gone. Is that because they're investing so much in AI right now or what's going on there?

Yeah. So this I saw this and I was like right maybe Google maybe SEO is dying but this is actually a little bit of a flawed headline. Because it is their cash flow that is actually negative. So as you can see here like their free cash flow in Q3 of 2025 was $24,000. Q1 2026 it was $10,000 and then Q2 was minus $5,855. So what many people don't know is you can still be a positive company because you've got the assets but you can have a loss for the quarter.

That makes sense. So like they've got assets but they've just lost on this quarter. But I've seen their for example their like Alphabet's shares like the company that owns Google is is absolutely flying right last time.

That's going to be our next slide. So profit actually jumped 298%. So they from what I was able to understand Google have invested into anthropic stocks and also SpaceX. And yeah, they they like their their their profit is at a skyhigh, but their actual money in the bank or the money that they brought in this quarter was minus $5,000, which is pretty crazy. And also I saw particularly with I think it's I think it's Anthropy but it might have been SpaceX but I saw that they actually 100xed their investment which is absolutely I think it'll be interesting to see Q3 figures from Google cuz that's when we can see right okay they they done this or they've done that from the investments that they done this this quarter. But I mean like the the operating profit is still 40.8 billion, which is up by 30%.

It's crazy, isn't it? It's I mean, when you look at the S&P 500, it's like it's it's mostly Google, Apple, and Nvidia out of the top 500 US companies, right?

Yeah. Like I mean, I would love to be Nvidia right now that Nvidia just like literally just shipping out graphics cards and RAM to to all these AI companies.

So here here's like a proper breakdown. Even though SEO is supposedly dying according to no one we SEO or search advertising still up 17% they're overall revenue is up 24% Google cloud's up 82% which is an interesting one Google cloud why do you think Google clouds went up by 82% Julian?

I'm assuming because of the use of AI because like so many people are using cloud models using Google cloud workspace probably even not even the AI But as well as that, you've got people using Google Cloud Workspace as the API inside their AI agents, right?

That that's where my first mind went to cuz like when when when I'm doing like custom tools and stuff like that, there's always going to be something that I'm going to need from the Google API library, whether it's like the Google Sheets API, the Google Search Console API, anything like that. And I reckon they're going to just make a boatload of money. like we we we already saw ahrefs they made the DR metric in their API completely free to use. I think that that's probably why Google clouds also gone up cuz AHREFS has probably seen okay if we don't do this now people are going to build more custom tools and they probably or AHREFS probably wants to be the overall outlier for the the domain rating metric. So I think that kind of goes hand in hand.

Yeah, that makes sense. Like even for example, people people are using like data for SEO and then just plugging that into their own custom tool, right? Instead of using I don't know, but I think we were talking about it the other week, like I literally don't use ahrefs apart from checking backlinks now.

I think the one best tool that ahrefs has is the history. I don't know if you've seen that, but you can track the history of your competitors. So for example, a month ago, you might be in position four and then you've got three competitors above you and now it's like slowly tracking it. So you can see across a 12 month or a six-month period. So yeah, I do agree with you though. Like I use ahrefs less. I could probably count on one hand the amount of times I've used the ahrefs in July. Which is pretty crazy to say.

And then down at the bottom we've got cloud work already sold. So this is for like these are like big contract stuff. So like governments for example using Google cloud and stuff. They still brought in 514 billion. So, it's not like Google is dying by any means. I'm sure they'll be able to survive on 514 billion.

Yeah. The amount of money is just ridiculous, isn't it?

Yeah. The next story, this is an interesting one, and I wanted to pull this up because we've spoken about this before, and I don't know if the audience knows our thought process about this. So, this was a tweet by Dmytro Krasun on Twitter. He owns Screenshot One. Basically Julian, for if you don't know what it does, it's basically like a website. A lot of developers use it. It takes screenshots of websites pretty much, right?

He had a page on his website which was bringing him a lot of customers, a lot of revenue. However, he didn't have it like he didn't update it for two years. As you can see here, this is like the the rank tracker. Overall 10,200 unique visits, load of good sessions, very good time on page, conversion rate was pretty high. It obviously ended up tanking, because they didn't keep it up to date and stuff like that.

So, he basically done these six steps, studied his search console, added internal links, republished on a bigger platform, which is an interesting one. I'll talk to you about that in just a second, rewrote the page by hand, asked Google to reindex it, and also made a YouTube video.

Republishing on a bigger platform. What's your thoughts on that?

That's like parasite SEO. Yeah, quite often that can backfire, right? It's like you you republish it on a bigger platform. I remember this with LinkedIn. You remember a couple of years ago, LinkedIn was just so easy to rank on Pulse and then some people would copy and paste the article. I remember there was one from a site that actually got penalized. Someone copy and pasted the article, posted it on LinkedIn, and it started outranking them within like 24 hours.

So this is the issue that I've got with this one. I mean obviously it's he's recovered it so it has worked for him but I would be careful with this. It can backfire but it can also sometimes go in your favor because you can end up getting like double bubble as I like to call it. So your your website's in an ideal world your website's ranking number one your parasites ranking number two and then number three is one of your competitors. So that's a that's an ideal scenario. Usually what ends up happening is the parasite ranks number one, your website ranks number two, or in some cases you might even end up going lower because it's it's republished content.

What about YouTube videos you've been doing? I mean, we've both been doing a lot of YouTube videos. Have you seen a direct correlation between YouTube videos being embedded on a page and it ranking higher?

No, like I've never tested out, but like one thing I'd say is if you've got a YouTube video and you use that content to create like a blog post, I think that's really good because you're adding information gain into the content and it's using your own words and your own data inside that video to create something better that AI can't generate itself. So, I think for that it's really good for actually embedding it into the blog post. I've seen mixed things. I've seen some people say don't embed YouTube videos in. I've seen some people say embed them because it increases the watch time on the page. What's your take?

I am more of a embed on a page because I think it's a twofold. It increases watch time on your website or your web page and it also increases the dwell time on the video as well. And that that's like a like any healthy channel like for example if you look at Matt Diggity or Nathan Gotch and stuff like that they do a lot of embeddings. And usually they they it also helps rank their video. So, it kind of works hand in hand. I think it's more of an ecosystem.

Yeah, that makes sense. Here's what was really interesting. What he was talking about is like the page is already back to position one, but AI overviews are taking up the space above it. A randomized study puts that click loss is at 39.8%. Have you started seeing much loss of clicks on some of your websites because of AI overviews?

Yeah, I've seen it reported. I've seen it on like across some client sites as well. It's like the clicks are down. Sometimes the leads actually go up because the traffic that does go through converts better, but overall yeah, for sure clicks are down and like AI overviews are just squeezing more room on the page.

Yeah, I've noticed that like in in our case, I've noticed that the conversion rate of the traffic is a lot higher. Like with SEO like few years ago when there wasn't AI overviews you might get 100 clicks and out of the 100 you might get 15 or 20 people that end up converting. Now it might be a little bit less clicks. So it might be let's say 70 or 65 clicks but it might be 20 or 30 people converting. So the conversion rate's a little bit higher. That's also working out a little bit more in our favor.

One thing that was really interesting about this tweet, and I know that me and you have spoke about this, is the attributing each channel. Obviously a click from YouTube isn't as high converting as a click from Google. It's not as high converting as a click from Instagram. What's your thoughts on channel attribution?

It depends how much traffic and how many sales you're getting. But I think if you plan to grow your business to seven, eight figures, then you at some point you can't grow without this, I think, cuz you don't know what's working and it's just kind of a big mess.

Yeah. Like I wanted to bring this up because I know that me, myself, and you have spoken about this and it's I think it's probably one of the most dangerous or vicious circles that you can get in when you're starting out your business because some people start out on let's say doing SEO and then they see, okay, Julian's doing YouTube, I'm going to do YouTube. And then they might see Gary Vee doing Instagram. I'm going to do Instagram. And then they might see Alex Hormozi doing TikTok. I'm going to do TikTok. and now it's just a overall spaghetti mess and you don't know where your leads are coming from. You don't know which channel's converting the highest. You don't know what where you're wasting time on.

I do agree with you in the sense that if you're going to be trying to grow to like seven, eight, nine, 10 figures. then you probably or you will need channel attribution. When you're starting out, I think you probably just home in on one one channel and just go go at it cuz I think one channel could probably get you to six or seven figures easy.

Yeah, that makes total sense to me. I mean, like I think by the point we got up to like maybe 30 or 40k a month on our agency, we were literally just relying on Upwork. Like that was it.

Yeah. Do you reckon you could have got to 100 just by doing Upwork?

Yeah, but I think I was like at that stage like you were saying where you start trying to do everything, you know, and you're like, I need to do someone told me to do YouTube. I'm going to do that as well. And then you start doing everything. And yeah, you're right. I think like I mean definitely to six figures. Definitely to six figures. Like one channel is enough.

Do you think that that's like one of the most commonly mistaken things by new businesses where they just try a little bit of everything and not all of one thing?

Yeah, for sure. I mean like I mean you do YouTube, I do YouTube is like just YouTube alone is is an absolutely brutal platform to compete for attention. It's like you you have to just get really really good at that and learn and learn and relentlessly keep improving every single day. And so like to do 10 or you know to do two different channels and probably at that stage as well of business you don't have a big team that's handling everything for you. Like for me we can post on Twitter cuz I literally I don't even log in. like it's all just there in the background with my team. But if it's you doing everything, I can't even imagine how stressful that would be.

Yeah. You're playing whack-a-mole, aren't you?

Yeah. I think if you're confident enough, do YouTube. I think the days of like old school like blogging is kind of just like being taken by AI overviews. I don't know what the the ROI of something like that would be personally. Do you do you still do much like blogs on your actual website or do you just do like video format?

Yeah, we do like maybe three or four new pages a day on each of our websites. And I know like the aiprofitboardroom.com particularly like I see a lot of sales come through on Skool from that. So I know it works pretty well. But at the same time like I know that most of our sales come from YouTube. YouTube is the place to be.

Yeah. Yeah. Yeah, definitely. Here's an interesting one. I don't know if you know about this. Yelp just licensed its reviews to ChatGPT. So, that's 330 million reviews and millions of business listings.

This is an opportunity. If I was if I'm selling reputation management, this is amazing.

Yeah. So I done some tests on this. So I don't know if many people know about this but whenever ChatGPT like you ask ChatGPT like hey who's the best SEO agencies in Canada it will actually do sub queries or query fanouts underneath that. So it will break down your search into maybe 10 different searches. What's interesting is it looks like ChatGPT has certain trusted websites. Yelp is probably obviously now one of them. Another one was Clutch. like it specifically looked at Clutch and SEO agencies that were listed on Clutch and the conclusion was if you weren't listed on Clutch, you were not going to get listed in the final answer on ChatGPT for best SEO agencies in Canada.

So now it looks like there's going to be certain trusted websites. I don't know every single one of them. I'm going to assume Yelp's going to be one. Clutch is probably going to be another. It might there might be like Yellow Pages as well. And if you're not listed on any of those directory sites, you're just not going to get shown up in in the actual final result, which is a bit crazy really.

Yeah. I mean, like I was speaking to the founder of Radar Kit the other day and he was saying like how with ChatGPT like sometimes it won't search the web or searches even. And so like you can only rank if you were indexing the last training data. And if you're not and it's a query where it doesn't do any web searches, you kind of have to do the work now to be in the next update. And I was like, that's crazy, isn't it? When you think about it, you're just invisible for 6 months.

Yeah. So, what one thing that I've noticed, ChatGPT has obviously like the three modes. It's got instant, it's got medium, and high. I believe that's there's three settings on all of its models, right? You you you might correct me if I'm wrong there. If you're on instant, I believe it uses the internal ChatGPT memory. I've never seen it do a fan out on instant, but again, I could be wrong. Medium and high, there are a lot more fan out, so it does search the web on those, but by default, ChatGPT is set on instant. So, if if you just like sign up to ChatGPT today, I don't know why you would be signing up to ChatGPT today, unless you've been living under a rock, but let's say you did. By default, it's set to instant. So, a lot of people will not probably change it from instant to medium, and they are literally just using ChatGPT's internal cache because I'm going to assume they've set it up that way because it's the cheapest to answer people's questions.

Yeah, that makes sense. And then the other thing I heard was from an SEO perspective, check it, run it, run the search for the keyword that you want to rank for, and if it does a web search more often than not, that's an opportunity for you to rank now. But if it doesn't do a web search, maybe like only 30% of the time it does a web search, then actually, yeah, you might have to wait six months before you're even in the training data and you have to start doing the work now.

Yeah, that's insane to me. And luckily, as we're recording this, it's the 30th of July, 2026. They're like Google's traffic is still like so much more than ChatGPT. So don't don't start crying if you're not in ChatGPT just yet.

Here's one for you. These are some quickfire ones. Did you know that Claude uses Brave Search engine to actually do any of the searches?

That rings a bell now you say, but I wouldn't have known that straight off the bat.

Yeah. Yeah. So, this was I initially saw this on a tweet by a guy called Rolo. You might have actually met him. I think he lives in Thailand or Vietnam. So he done a very interesting breakdown where he was he analyzed the searches and he said basically Google's displaying this Claude's using these sources something's not right and then he ended up breaking it down and it 86.7% of it overlaps with what Brave shows up. So, if you are struggling when it comes to ranking in Claude or being cited in Claude, go over to Brave and see if your pages are even indexed in Brave because a lot of the times they're not.

Cloudflare blocking AI crawlers. I don't know if Do you use Cloudflare?

I think we do for juliangoldie.com. For the new AI websites, we've not been using it, but that's pretty crazy. So, like you basically can't get indexed inside of Cloudflare or how does that work?

So basically, let me so this was the one about Claude split your traffic by channels. So Cloudflare by default currently you need to select a setting which allows AI crawlers after 15th of September they're flicking it. So by default it's going to allow AI crawlers. And if you don't want AI to use your data as training for the LLMs, you need to specifically tell it like it's just a button. and you just click and it's going to disable it. And that's a bit of an interesting one for me because I think Cloudflare has been paid there.

Ah, so you think what paid by the AI or

Yeah, I could be wrong, but Cloudflare about 12 months ago was like really adamant like we're not going to sell like any of the data by default. We've actually got it disabled and now they've kind of went back on their word which is a bit of an interesting one.

AI browsers. Are you using AI browsers?

Not at all. I mean, Chrome kind of has AI built in, but I literally I don't like them. I've not seen like any good ones. I mean, like I use Claude's AI Chrome extension, like the automated one, but like I just thought it was so slow. Like it takes screenshots of things. I think it was one step in the right direction, but 10 steps backwards.

But an interesting stat, 15 to 20% of the market by December is going to be shifting from Chrome to AI browsers.

Interesting. Yeah, I can imagine that. I mean, it's I guess like people probably use it the most when they're using, for example, Codex and then they'll just stay inside the Codex app.

Yeah, it'll be an interesting one because I think I have actually seen a few new AI like for example right now the browser that I'm using is Arc browser and that has got an AI module somewhere in it. I never use it. I have started seeing a lot more AI browsers outside of like the ChatGPT and the perplexity and stuff like that. But again it's not for me.

How come you're using Arc?

So, I like it, right? Because it can it's got like the tabs on the like it's just so like let's say if if I've got like 20 or 30 tabs down here. It's like a scroll for me as opposed to having it horizontal where I don't see all of my tabs.

Yeah, that makes sense.

I mean, yeah, it's probably just my OCD.

Right, so anthropic claude split your chat. Right. Okay. So the these are some recaps right and for Anthropic using Brave and splitting your traffic by channel try and do this if you are using multiple different channels to drive traffic because I can guarantee the KPIs of SEO is going to be different to the KPIs of let's say YouTube.

And then last one, Reddit potentially might be cutting off Google because they don't want I'm going to assume they've just had so much spam from people trying to manipulate the Google AI overviews. What's your thoughts on this?

Yeah, I don't think they were best pleased. And also there's been the weird situation with like SERP API, isn't there, where like and and Reddit were kind of doing something similar as well with someone. I don't know if it was SERP API directly or someone else. But yeah, I mean I can't see Reddit cutting off Google, but they do once a year they seem to like threaten Google like, "Listen, mate, if you you carry on the way you're doing, we're going to stop here."

So to my understanding, right, how this came about is Google invested money into Reddit I believe and then in return Reddit allowed Google to use their their Reddit forums and Reddit posts and stuff for Google's LLM training data. Now again, this is slightly like this is early days. It's not fully confirmed yet. But to my understanding, from what people are saying that there's been so much money that Reddit has had to spend in fighting spam where it's almost become like not profitable for them. So now what they're trying to do is think of like a middle ground where they can still potentially get the money from Google, but they don't want all the spam and that comes along with it, which is a bit of a difficult one.

SEOs ruin everything. That's basically it, right?

I yeah I mean the title for this should be SEOs ruin Reddit. Let's do that.

So yeah any any closing points?

That's it for me. But yeah no interesting lots of interesting stuff going on in SEO right now.

Yeah it's going to be an interesting like two or three months. Is there any LLMs that you are excited for that are going to potentially be launching?

I think that with Opus 5 becoming at the same level as Fable 5, they've got to release Fable, you know, 5.1 or Fable 6 pretty soon when that comes. I mean, it's already really really good. It's also a lot faster than Opus, but I'm excited to see that for sure.

Yeah, that that will be a definite one to check out. What's the one LLM that you're using the most right now?

Pretty much Claude. And then Yeah, I mean, as far as LLMs go, I use Claude quite a lot. Yeah, but I'm always testing new ones pretty much every single day. How about you?

Claude, I've started actually using Codex a little bit. Just testing it out and I think ChatGPT slightly won me over a little bit. Like I'm still like 91% Claude, 8% ChatGPT. So yeah.

All right, that concludes the first podcast episode.

Awesome. Cheers everyone for watching. Make sure you subscribe.