Welcome And Why Rand Matters

Chris Norton 0:00

Today's guest needs no introduction, but I'm going to give him one anyway. Rand Fishkin built Moz into the biggest name in SEO. He raised $30 million in venture capital and then watched his decision nearly break it. He stepped down as a CEO during a battle with depression, walked away and started again the very next day. Now he runs SparkToro, a company a twentieth the size of Moz, and he's happier, healthier, and making much more money. In this episode, Rand tells us why raising venture capital was his biggest mistake, why traffic has always been a vanity metric, and why your boss is wrong to give Google the credit for your branded search traffic. This one is brutally honest, and I think it's one of our best episodes ever. You need to listen to this. Let's get into it. Enjoy. Rand Fishkin, welcome to the show.

Rand Fishkin 0:59

Thanks for having me, Chris. Great to be here.

Chris Norton 1:01

I mean, thanks. Yeah, thanks for coming on the show. Delighted to have you. Um, I know you from Moz Fame from many, many years ago. So Will and I are that old that we've o we've before we ran this PR agency, we used to work, we used to work in another PR agency together, and I was massively into SEO as a PR. There wasn't a thing called digital PR back then. Um, and so I was sort of like a PR doing uh social media and uh SEO, and I used to get a lot of my information from Moz. And um obviously we had the you had the little Moz um Moz rank um scoring system uh and everything, and that was all developed by you. I mean, well, not all of it, but it was your idea initially. So, do you want to just give us a bit of um a rundown for how you came about to set up Moz? Because I think that story is fascinating for us and people will like to hear it.

Rand Fishkin 1:58

Yeah, yeah, old story. Um so this is yeah, I you know, I wrote this book about it, Lost and Founder, but the you know, the introduction is basically me dropping out of college, uh, two classes away from graduating at the University of Washington here in Seattle, and then working with my mom, Gillian, who'd been running a you know solo one-person marketing consultancy for 20 years at that point. This is 2001, and basically being a bad web designer who couldn't get work and couldn't get jobs and couldn't find clients who would pay, and got into SEO because we had SEO subcontractors, right, who are doing the SEO for the websites that we built for our clients, and we couldn't afford to pay them. But we needed, you know, we needed the money from our clients and we had promised the work to them. So I had to learn SEO myself and do it. And it it turned out that I was pretty good at it. Um, it just became a you know a passion for for a long time, right? I um basically was at Moz, the company that became Moz for 17 years. Um, I became the CEO in 07 after we had sort of moved from being a consulting business to starting a software company.

Raising Venture Capital Breaks Focus

Rand Fishkin 3:14

I had all these software things that I wanted us to build, and we eventually did. And then uh raised lots of venture capital, which which in retrospect was a mistake, big mistake. Um, I'm not sure it's a marketing mistake, but oh my god, what a it can be founder mistakes are good as well.

Chris Norton 3:32

I did say that.

Will Ockenden 3:33

We're allowed founder mistakes. Yeah, but dive into that because obviously you you raise capital in order to scale the business, but it it created an enormous pressure, didn't it, for you?

Rand Fishkin 3:43

Yeah, and it's not just the pressure, you you shift the company from being product customer and team focused to being low chance of a billion-dollar return focused, right? So rather than at lots of board meetings, right? If if the three of us, let's say the three of us built software, right? We're um SparkToro, right? SparkToro is one of the is my one of my current companies, does audience research software, right? And when when Casey and I get together and we talk about what do we want to build? Who are we building it for? Why are we building it? What are our plans? You know, I got a giant whiteboard in front of me on the other side of the screen you can't see, that's got, you know, all these all these different data sources, things that our customers have told us we wanted. We run surveys. Board of directors at Moz doesn't really care about that at all, right? The the the venture investors care about is this a giant total addressable market? And how are we gonna become the monopoly in a giant total addressable market space? Building things for at the time, right? Building things for SEO professionals and and and PRs who have to do SEO, you know, that's that's little stuff. Like, don't worry about that. Tell me how we're gonna compete with HubSpot. How are we gonna get this to a public company size? Focusing on Google and SEO, that's that's too small a market. We need to go big. How are you gonna go big? And that that's that's when things go crazy, right? And that is that that is the mistake that I made, especially, you know, the first round of venture was so small. It was what one point 1.1 million um in 2007. It really didn't change the company all that much. And it did, you know, it's well to your point. It gave us the money to build what we needed to build, which was, you know, we wanted to build that link index of the web, right? Chris, to your point around MozRank and domain authority and those things. It let us build that. That was incredible. But man, you know, once once we raised $30 million, that's different story, right? Um, so I I raised that money foolishly, uh, stepped down as CEO during about with depression in 2014, and then sort of negotiated a very painful departure from the company uh that that got me out at the start of 2018, and and I started SparkToro the very next day.

Will Ockenden 6:12

I was gonna say, I mean, you you've you I mean we'll we'll dive into some of your current business concerns, but you're you're oh you know, you had now three businesses. How have things kind of changed? I mean, presumably it's now on your own terms, um, and there's a much healthier kind of balance. You're in control of your own destiny. Talk talk to us about that.

Rand Fishkin 6:30

Yeah, I mean, I think this is one of the one of the most powerful things about making big, awful, terrible, life-changing mistakes, is that you get the opportunity to rectify them. You can change. This is one of the most beautiful things about human beings is that we can change our minds, right? We can go from people who stand on stages and glorify the money that we've raised and how powerful and important our company is because of our investor backing and how we're going to take over the world, and how you know we're making 40 or 50 million dollars a year at our business. You know what's you know what's true and deeply embarrassing is I I am more financially successful, I am more emotionally healthy, I am more day-to-day happy, I am a better partner and friend and human being and family member today than I than I ever was when I was very successful, right? Making making 50 million dollars a year, you know. Uh um SparkToro is still uh you know under five million dollars a year, not even not even a tenth, not even a twentieth the size of Moz. And so so much better. And I'm and I'm making more money. It's really weird. It doesn't make any sense, you know.

Chris Norton 7:58

That's why you feel happier. You've got you're making more money and you're not as under as much pressure, but it's also your baby as well. You is it is it I always what when I see people that have obviously me and Will haven't sold out, so we we own our business. Um is it um the fact that you're relinquishing control and that's that when somebody else then has control of your business and they're you're then you know they're in control of you rather than you're in control of your own destiny then? Is that where the stress comes from?

Rand Fishkin 10:06

Okay, here this is even more embarrassing to admit. I think when that happens, and Chris, I've talked to plenty of founders. I I talked to a founder friend of mine, gosh, two weeks ago, who from the outside looks incredibly successful. Like his business is doing what looks amazing, but it but it turns out behind the scenes there's a private equity deal, and you know, he's just in these chains and he's paying, you know, huge amounts of interest on debt that he had to do in order to get a deal done. It's just it's just awful. Um, so yes, that can cause a tremendous amount of stress, but when I was CEO of Moz, all of the pressure was self-created. My documents with my investors said that they couldn't replace me. I if I had not chosen to step down as CEO, I owned two of the four board seats, I had uh veto rights on nearly everything. I like we really negotiated extremely well in all these deals. Uh I would not it was the cultural environment. It was the cultural environment that caused that pressure. Um and I I felt an obligation, a deep obligation to you know, live up to this promise of how am I gonna get you 10 times your money back in the next seven years or whatever it is, right? So you know, they put in 30, I need to get them 300 every year. It's like, all right, we got to make another 25 million this year. We only made 10. You know, like uh oh, right. Um yeah, yeah, it's

When Pressure Is Self-Created

Rand Fishkin 11:48

it's really weird. I I should have just stood at those boarding board meetings and said, hey, you can tell me whatever you want. This is what we're gonna do. We're gonna run a highly successful, profitable business that serves this particular niche. I bet Moz would have been, you know, it would have been SEM rush or or AH refs today, right? With with hundreds of millions of dollars of revenue and happy customers and you know, still a market leader. But instead we chase this, you know, this other outcome, and that is completely my fault. Completely my fault. You know, there's no there's no, you know, we can talk about the circumstances that created that environment, but there's no one to blame but me.

Chris Norton 12:30

I mean, wow, this is like I mean, we had so we've had some things on here, but that is the most honest thing I think I've ever heard. I mean, the pressure. It do you know what it sounded like? It sounded like an episode of succession when you were reading it out. The pressure that they were each one of them trying to succeed their dad, and they they're under all that pressure. And I was just thinking of you in that role. Also, you're wearing a suit, and I could I could see you as the CEO.

Rand Fishkin 12:54

Well, I'm I'm no suit. This is a this is a waistcoat, you know, a little knitted waistcoat and a knitted tie. I'm trying to bring the formality level down here.

Chris Norton 13:02

Right. You're bringing you you're bringing you're bringing the snap to the to the show, anyway. Um you've talked a lot about zero click search. Um so I mean, for marketers out there listening and that aren't geeky, like a bit like we are. What how could you explain zero-click search today in 2026?

Rand Fishkin 13:25

You know what the you know what the easiest way to explain it to you is this is 10 minutes before I came on your show. Geraldine, my wife, uh, and I are getting ready for our day, right? She's she's going out somewhere and I'm getting ready for this. And she is looking at her phone and she says, I just Googled something. Where are the results? Right? Where where are the websites? I can't click on anything. It's just it's all answers and AI and suggestions and stuff that's inside Google. That is that is the zero-click world that we're living in. It's inside Google, it's inside all your AI tools, it's inside all of your social media, Facebook and YouTube, and LinkedIn, and Reddit, and TikTok and uh Instagram, right? All of these platforms and environments want to keep you within their ecosystem. They don't want anybody clicking out. That that doesn't do anything for them unless you pay. Right. And so inside this zero-click world, what's happening is people are completing their entire customer journey from very top of funnel and problem discovery all the way down to brand preference and brand choice and product choice. But they're doing it in an environment that's not your website. They're doing it inside of Instagram, they're doing it inside of Reddit, they're doing it inside

Zero-Click Search In Plain English

Rand Fishkin 14:49

of Google AI mode, they're doing it inside of Google's instant answers, they're doing it in ChatGPT and Claude. Because of that, you need to embrace this idea. We all need to embrace this idea that is completely counter to what we've done for the last quarter century of zero-click marketing. Doing marketing even when it draws no click, but creates influence in the platform and the place where your audience is paying attention. And so this is, you know, this is the book that Amanda and I are writing that's coming out this fall. And we're um we're we're hoping we can help a lot of folks who you know have this historic mindset that that really worked until like 22, 23. It's not it's not all that new yet, right? Right, right, but it's also just a massive sea change. I mean, you guys think about, you know, I you did PR, right? PR, PR is one of those um marketing channels that has always struggled to get attribution. And now every channel struggles to get attribution because there's no click. You can't look in your analytics and see it. You and I both know, right, that uh um if my friend Scott Heimendinger gets his ultrasonic chef's knife, you know, uh a fantastic review in The Guardian, it's gonna drive a ton of sales. And it did, right? This this happened like a few weeks ago, I think. Ton of sales, right? You can log into his sales dashboard, you see this big spike, but you can't prove it. You can't prove that it came from the Guardian. Does this mean Scott should stop doing PR? No, no, that's absolutely mental. But the the model that so many business owners, so many CEOs and CFOs and boards of directors and CMOs have is that if it's not in their analytics dashboard, if they can't see it in the numbers, it doesn't matter. It's not real.

Chris Norton 16:45

It's so the thing that Will and I have were just talking about before you came on, we've been talking about it today. We've talked about we were in Paris last week at the WorldCom conference, and we were talking about something that I've I'm fascinated with being a bit of an analytics and SEO geek. Traffic. So we we about nine months ago, when ironically, it was April, I think it was not nine months ago, it was over a year ago now. Uh April 2035, just after the AI overview, I put two and two together here, traffic had dropped on our website. I think it was by 48%. Um, but inquiries, yeah, 48%. And then we I we did some anal I looked, I jumped into the analytics and went through with Will and we were looking at what the problem was. Then I looked at Daily Mail, their traffic was down 62%, I think. I've done a whole there's a whole YouTube webinar we did on it, and we had like 800 people come on it. We'd probably do a new version of that. Anyway, that's fine. Delved into the and we could but I saw that the the leads on our website, which we come through HubSpot and various places, were going were fine, they stayed the same. Traffic down 50%, leads not down at all. However, the last eight weeks, leads are now down, traffic's going down still, and I'm wondering, and because we we did our business, we get a lot of referrals, you know, we're we're with a we're a big, we're fairly we're a medium-sized PR agency, we're the biggest in Yorkshire, which is the biggest state in in the UK. Um, so we get a lot of referrals, and we get um, but we get quite a lot from search marketing, PPC, and stuff like that. But in quite inquiries, leads has started to drop off quite a lot. And um, not from referrals, because obviously we know those people. And I will and I were discussing it.

Traffic Falls While Leads Hold

Chris Norton 18:27

Is are we starting to see the disruption that people were talking about from is it artificial intelligence and is this tied to your zero-click search? Is this AI crash 2.0 that we saw in the 2000s? Is that what's happening now? And is it going to be worse? Or and this is a big question. So I'm I'm open, I just thought it'd be fascinating to get your insight. Yeah. Uh okay, so lots of things are going on here. One is for tons of businesses that we've observed and sort of um people who've given us access to their metrics, we can see exactly what you're describing, right? This this phenomenon. I wrote a popular blog post about this, right? That traffic's down, but revenues up. It doesn't seem to make any sense, right? And and the answer is traffic was always a vanity metric. All of this, you know, hey, we're going to create great content and then we're going to drive people up to our website and from our website, some of them, some small percent are going to become leads, some small percent are going to sign up for our email list. That whole ecosystem is sort of falling apart as all of the big tech platforms try to create ownership and control inside their ecosystems. Now, this being said, if you're if you're the Daily Mail, right? If you're a publishing, and I, you know, I don't think anyone here is going to shed a tear for them, but the if if you are, you know, a publisher, well, what model do you have except traffic? Traffic is how you monetize, traffic is revenue. I think for those folks, this world is uh apocalyptic, right? And they the the the only the only solution is to find another way to monetize your business, find another another business path. Great, you know, two two good examples that I always point to. One is Bloomberg, right? Bloomberg is a giant publishing engine in the financial field, but they don't need traffic. They're happy to keep their you know millions of of uh people who are exposed to their content on YouTube or inside of Google or you know, license their data to third parties because the real value in their publishing is from people using the terminal and trading stocks and bonds, right? And being part of the market. That's that's where they benefit. So they don't they don't need 10 million visitors on their website if they have 10 million viewers to their YouTube. And that's exactly what they're they're finding. The other good example is the New York Times, which the New York Times has become a games company. Almost two-thirds of New York Times visitors, subscribers visit nothing except the games and cooking section. Nothing. Wow. They're not subscribing to the New York Times, they're subscribing to Wordle. Right? This and and it's been hugely successful for them, right? They're essentially a a games platform with this ancient publishing appendage. Um, what are that, like a like a prehensile tale, you know? And this is this is kind of the model going forward. It's really, you know, it's quite tragic if you're someone who uh believes in the power of the free press and and you want to see sort of great journalism. Um I think there's gonna have to be a huge upheaval around that before before we see something settle. But I would point to the second thing you talked about, which is like, hey, now we're seeing our leads drop. One of the things that I would, you know, that I would encourage you to do is is sort of to build a new form of analytics dashboard to be able to reverse engineer what's happening. Right. So at the at the top of your kind of um marketing funnel, my suspicion is there are people who find you via, you know, three or four primary sources. One is sort of press media and blogs and and email newsletters and sort of you know things in our field that send you referrals. They might not even send direct traffic, but your name comes up in those places. Podcasts, absolutely podcasts, great. Yeah, podcasts, YouTube, webinars, right? All that kind of stuff, right? And you can see some rough stuff, right? Like if you use, I don't know, you probably use something uh higher tier or more expensive than Alertmouse, but like, you know, we I I use Alertmouse for this because one of my companies, and see, like, hey, here's all the mentions of SparkToro, here's how many we've gotten, here's how high impact those are. You know, and if that number, if that graph starts dropping, I'm like, aha, our press and media presence is dropping, right? Our sort of third party. So the second, the second one is Google, right? So like it's not clicks, but it's impressions. How many times did our brand come up in Google's results? They're they're now surfacing how many times it came up in um AI answers, right? They they just launched that yesterday, I think, the test. So you're gonna be able to see that soon. You can do this too with third party, you know. Um, I use Gumshoe, which is a company here in Seattle, but like Gumshoe will tell me, hey, you know, you're here's

Rand Fishkin 23:33

whatever, 5,000. Different prompts that people might enter around all the topics of SparkToro and audience research.

Measuring Influence Without Clicks

Rand Fishkin 23:40

I'm like, hey, great. You know, here's how many times you were mentioned versus your competitors. Terrific. Like that really helps too. That tells me what's going on in AI. And then social media. What's my LinkedIn impression count? What's my Twitter impression count? What's my thread impress threads impression count? What's my Reddit impressions count? What's my YouTube views? All of those. So, you know, you put it sort of put things into these four categories and then you can see if one of them is declining. I don't know if it's the AI apocalypse yet. My suspicion is no, it's not. It's probably something that's going on in one of these four channels where, you know, maybe Google started recommending one of your competitors more than you because of I, you know, this is a a theory, but like a lot of people are doing AI spamming these days. It's really easy, right? If you want, if you want to create a fake PR firm tomorrow, and you know, we can contact a bunch of people I know and be like, hey, make this the top PR firm that's most mentioned for you know public relations in Yorkshire. No problem.

Chris Norton 24:38

Can we speak to your mate? Because that'd be quite handy.

Rand Fishkin 24:42

I mean, the the question is whether you want to do risky things with your with your real company.

Chris Norton 24:49

No, I because I I suppose is the the and I'll let we'll can come in in a minute. I just wanted to ask one more question. So did because that ties to what we talk about a lot on this show, which is about the power of brand. It's not why we started the podcast, by the way, but brand actually has done us quite well with the podcast because the brand the podcast is doing well. Number one in Apple, by the way, in business. No, not in business, in marketing. Three in business in the UK. Um, but um uh yeah, I but is pot is like brand really the it sounds to me, because when I've I when we're discussing it on our side, we felt brand's the way to go because brand will stay strong. If you've got a strong brand, that'll perform, might not right now, but it will hold better than these little tricks as all these black-hatting AI robotic monsters pop up, like they did with SEO and like they did with um social media. That is that where is that stage, aren't we?

Rand Fishkin 25:46

Yeah, 100% agree with you. I look, there's lots of AI tricks. I think it's fun to play with, I think it's fun to you know become the thing that ChatGPT recommends the most or Claude recommends the most. And and then three months later, right, you fall off because you know their their systems get more sophisticated, just like the early days of SEO. But I'd be very cautious about doing that with your primary brand that you're trying to build for the long term. Brand plays on compound interest, right? People know you, like you, trust you, they use you, they refer more people to you. You keep that same brand name, you keep that same focus. This is this is how Moz was built, right? Moz was SEO software. It was track your rankings, it was get your domain authority, it was see the links that are coming to you. It was, you know, all these things, right? Uh crawl your website and find errors and opportunities. And then, and then it tried to be

Brand Compound Interest Beats Hacks

Rand Fishkin 26:37

10 other things, and the brand got diluted by the product, by the marketing, by the by by the CEO, right? Um, and yeah, huge, huge mistake in judgment and and really compromised, you know, sort of um took the legs out from under the brand's ability to benefit from compound interest. So sometimes, you know, um we've done this at SparkToro before. You kind of go hedgehog mode, right? Where you're just like burrow down, keep trying to keep staying focused on your one thing, your little hole in the ground, and you you wait out these challenges and you emerge stronger on the other side with more opportunity while a bunch of people are flailing around and trying to figure this out. It's yeah, it can be it can be tough. And I I don't think this AI apocalypse is quite gonna hit marketing yet. I think where the AI apocalypse is going to hit is on the financial side, right? There's just, you know, this $8 trillion of build-out for data centers that can't possibly monetize. I think what the number has to be like something like five times the revenue of all big tech companies combined within four years. I the numbers don't work, do they? Yeah, no one thinks the numbers work, right? And so you can see all that all these market players are like, well, you know, I kind of just got my finger on the sell button waiting for it to waiting for a big receptor.

Will Ockenden 28:09

Yeah, yeah. So um I've I've got a question, and and you've you've you've sort of touched on this, Rand, but you you talked earlier about zero-click marketing, you know, and the need to be kind of proactive when it comes to zero-click marketing. Now you've talked about things like building brand. You know, and this is probably quite a big question, but where should where should brands be focusing their time and effort? And and the second part to that question is can anyone do it? You know, if we're a mum and dad corner shop or restaurant, can we compete?

Rand Fishkin 28:40

Yeah, yeah. Fantastic question, Will. Okay, so here's one of my pet peeves about marketing world is that the stories in marketing that often get the most attention, go big and viral, everyone points to, they come from big brands. And there is not a lot that a mom and pop shop that a small service business or a retail business or a restaurant can learn from what Nike did right with their rebrand or what Apple did wrong, or yeah, like all those case studies are just really useless if you're not doing a billion dollars a turnover. Uh okay, my recommendation, my recommendation for anyone who is trying to do zero-click marketing or really marketing of any kind. If you're small, stay very focused. You're gonna hear advice from you know people like uh Gary Vaynerchuck, or you know, those types of folks where they're like, you gotta be everywhere. Your customers are in these, you know, 15 different social channels and Google and AI tools. And you might have heard us talking today about a bunch of different channels and you know, these four different columns. You don't have to be in all those places. If you are great at one or two forms of

Pick Two Channels And Commit

Rand Fishkin 29:52

marketing, just keep hitting those channels. Almost any of them for a small business, for a medium-sized business, even if you're doing a few million pounds of turnover, totally big enough. That is that is completely big enough to go after um two channels, right? So you might say, hey, you know what? I'm I'm actually really good at answering people's questions on Reddit. I got an account there, I read it all the time. I sort of get the culture and the lingo and the vibe. Keep contributing to Reddit, keep being a helpful person in the, you know, the whatever HVAC subreddit, right? For people who are installing air conditioning in the UK because global warming means that temperatures in the UK are going crazy in the spring. Yeah, great, fantastic. Like, like, do it, get in there, be that person. Maybe you hate Reddit, maybe you think it's just a cesspool, but you're pretty good on video. YouTube Shorts is just waiting for you, right? Or you hate video, you hate Reddit, no way are you doing those things. But turns out, you know, you got a killer voice for radio, you're really good at going on other people's podcasts, you have a bunch of interesting things to say. Guess what? There's gonna be, I don't care what industry you're in. There's a thousand podcasts in your industry, webinars and you know, people with other channels, pitch them. Do interesting research, do interesting things, have interesting stories to tell, and then go pitch those stories and that research and those interesting things. Um, this can work in literally any channel you can think of. Maybe you're terrible at all those things, but you are absolutely fantastic at on-the-street marketing, right? Like fantastic storefront, great window display and design. Uh, maybe you're excellent at um, you know, uh in real life advertising, right? So putting up posters and billboards and you know, uh and ads like that. Fantastic. Lean into it. Like, don't don't let the world of digital tell you that you can't still bring people to your business in the real world. You absolutely can. This is about leaning into three things. One, is my audience actually paying attention to this place? Whatever it is, right? This channel. If you're going after podcasts, but you're in HVAC, I'm not sure. Like it's a little dicey, right? It might might not work so well. So that's the first one. The second one is are you do you have unique value that you can offer there? Something that is different from what the rest of the market is doing and differentiate it. And the third one is do you have passion and interest around this? It's it's not it's not just the value you provide, right? If you I've never heard anyone say, for example, Will, never heard this. Oh, I hate Instagram so much. I wish that app would just fall in the ocean, but I'm so good at it. Nope. Right, doesn't happen. So there's a thing that you hate and you think you're supposed to do it. Maybe go find the thing that you like, the thing you brings you energy, because even if you suck at that thing today, you will be putting more energy and more revolutions into it. And like we said, the most powerful force in marketing is compound interest of brand.

Will Ockenden 32:59

Love it, love it. And then um specifically, if I'm gonna push you specifically, do certain channels have undue influence when it comes to um zero-click marketing, GEO summaries? You know, if if if we if we're serious about being seen in those, I know it varies platform to platform, but you know, is it is it PR, is it earned media, is it directories, where should we be looking? Yeah, yeah, yeah.

Rand Fishkin 33:23

So uh, okay, so if we're talking about specifically zero-click marketing for AI answers or or instant search results, right, at the top of Google, the AI overviews, uh AI mode, Claude, ChatGPT, all that kind of stuff. That that kind of what we call it like zero-click marketing for AI or SEO as opposed to for social or PR or content, those draw from two kinds of sources. And this is it's quite frustrating to understand. We we sort of wrote a whole chapter on this in the book with a bunch of research. But okay, there's two ways that they come up with this. One is called, uh, is is pulled from the training models themselves, right? The models train on a bunch of content that is crawled from the internet and taken from books and taken from YouTube transcripts and and television transcripts and podcast podcast transcripts, right? It's like this historical mass of data that tells them which words frequently come after other words. This is why a lot of people call AI spicy autocomplete, which I think is a good shorthand for it. So if your brand gets mentioned alongside the words and phrases, right? You want to be the most mentioned uh uh PR agency in Yorkshire in ChatGPT, well, all the documents on the web, especially the the recent web, the last six months, nine months, if a high percentage of them who say PR agency or public relations agency or or any other synonym also mention your agency, boom, you you are gonna be in those top few results. The second kind on the training data is is called RAG RAG.

How AI Answers Choose Sources

Rand Fishkin 35:06

It's retrieval augmented generation. And this is when the AI tool, uh, which which can be Google as well, does a search, a classic web search or many web searches, which is called called query fan out, which you don't need to know too much about, but but essentially is performing a bunch of searches just like a human being might to figure out what are the top sources or things that it should pull from or or missed opportunities. Rag is happening a lot now. You know, it used to be maybe 10% of AI prompts would result in a in a rag uh uh return, but now it's 60%, 70%, could be higher depending on your industry. And in those cases, you need to do two things. You've got to be in that training data, and you need to be in those top results. So this gets weird, right? Like suddenly I'm doing classic SEO to rank well or be in the results. It doesn't have to be your website that comes up, it can just be your brand that's mentioned on a page that does well in the results. But you want to be in those top results, not because anyone's ever gonna see you or click you there, but because the AI, right, is gonna perform this search and then show you. So it's yeah. And I do want to say this, right? And we we we talk about this in the book too. The last 90 seconds of advice that I've given you, you ask me again in six months, the answer might totally change. Like AI is so it it's it changes so frequently, right? Like that last year, people were talking about how basically you could you could submit um 10 posts to a bunch of subreddits that had never existed before yesterday with zero upvotes, no one's ever upvoted them, there's no comments, no conversation, and you could like that instantly improve your ChatGPT mentions, right? Like ChatGPT was just just pulling from Reddit like there's no tomorrow, and it didn't care whether there was any. And then, of course, that changed. And then and then there was like a YouTube era that we're we're currently in a bit of a YouTube era with Google's AI answers, where you guys have probably seen this, right? You can put up a bunch of YouTube videos with your brand name, even if they get zero views. You you can get a lot of you know, a lot of pull there. Is that gonna last? No way. Google's not gonna let that stand for long, right? That would that would totally undermine their result quality, but you know, it works today. And I don't know, are the three of us gonna tell people you shouldn't spam AI?

Chris Norton 37:44

Let's start spamming AI. AI spamming everybody. I I see so many people rant uh so many people ranting on LinkedIn now. Um, you ranting about AI slop and the fact that and in fact, there was another there was an interview today on one of the news sites in the UK, and it was it was journalists moaning that PRs are now using AI to pitch to them, even really good quality ones are using it. I think, yeah, but then you could flip that on its head. Journalists, we there was uh a couple of examples where a journalist, and it's there's an article, it was published, right? And at the bottom it said, um, the usual bit that it says, would you like me to amend this? And that was in the article. He'd obviously forgot they cut and pasted it, the the article, and left the would you like me to uh the actual prompt from the AR. Also, why do they leave that bit in on the copy and paste? I never know, you know what? Nobody, nobody, if you're listening out there, Mr. Altman, if you're listening out there, get rid of the copy of the end of the conversation. We just want the content copied, don't we? Who wants that?

Rand Fishkin 38:52

Oh my god. It's it's a wild era for AI and content creation. You know, I think in 10 years, 20 years, people are gonna look back and be like, oh my god, what were what were they thinking? What were they doing, right? How did they let this technology just run amok? They were letting students use it for for school. People learned nothing, you know. It's just um yeah, and this is how it goes with new technologies, right? Uh, there's always there's always that pushback. I'm sure you guys remember, uh, because you look like you're about my age, right? The the late 90s, early 2000s, when there was tons of media and journalism about how the web was ruining kids and ruining social relationships and you know, ruining the planet, and and and how we would all give it up, right? The world wide web would would be used by only a handful of businesses in a few years. It's just this new thing, right? And yeah.

Chris Norton 39:48

We're now getting to the I mean when we're in the bloody social media um policy. Gov governments are now talking about control, you know, Australia are controlling uh age groups. We're now getting into that era of where social media is is the tr that each government is trying to find a way to control it and make it well, not as uh corrosive and horrible.

Rand Fishkin 40:08

Um I mean it's you know, it's equally I I think from a health perspective and from a societal perspective, it's equally dangerous as cigarettes or alcohol or or opium or you know, any of these other things, right? And it always takes time, it takes years and decades before human beings and societies and eventually governments take action to regulate and control. I, you know, I've often I've often thought to myself that there will be a time in the future when human societies look back and they say, wait a minute, you let anyone post anything they want to social media, then you build algorithms to optimize not for the health and and truth of any of that, but for engagement. And then you were surprised when fascists came into power, you know, and started wars. How can you be surprised about that, you idiots? You know, right? And so, like, this this feels like the type of thing where, you know, it's like carbon monoxide. You gotta regulate it. It's you can't just be spewing, you know, unlimited quantities into the atmosphere.

Chris Norton 41:11

No. So so we haven't really we've we've touched on a mistake you've made, and you you found have you got a marketing mistake to share on the show then?

Rand Fishkin 41:18

Oh man, so many. Oof. God, have I done marketing mistakes. Um yeah, so let's see. So uh some of I think uh my biggest recent marketing mistakes uh have been with positioning. Positioning is one of those things that I I think because my background was SEO, you know, when you I I like to say, you know, when you uh when you hand in an SEO a problem, they they they think every they they just take their their keyword rankings hammer to it, right? So for you know, for 17 years of my career, all I thought about was how do I move up in position in Google? How do I how do I become uh higher ranked in Google and get more clicks and then turn more of the those clicks into conversions? And what I never thought about was why does this business exist? Why is this business the uh a better alternative to solving this customer problem than anything else they might consider? And then how do I make that really clear over and over again in the places where my audience pays attention? That that was not really a consideration. I think I accidentally was good at it with Moz. But with SparkToro, we got positioning wrong several times. We might not even have it right now, you know. We um we started we started as sort of a uh social uh

Positioning Mistakes And Fixes

Rand Fishkin 42:53

intelligence riff on social listening data. That that was not really what we were, but we thought we should describe ourselves that way because we had this sort of graph of all the social data that we were using to provide, you know, information about like where do you where do your audience go and and what do they pay attention to, what are they listening to, what are they following, what are they watching on YouTube, which podcast are they subscribing to, what conferences they go to. So we had all this great data, but we were describing ourselves in a way that people didn't quickly understand and that they didn't associate with the problem we were solving. Then we pivoted to audience intelligence. We were like, okay, we're gonna create this category of audience intelligence. Category creation is very expensive, very time consuming, super resource heavy, and we were a two-person team with you know a million dollars of funding that we needed to spend on, you know, our our software. Again, didn't didn't really work. We pivoted a third time. We are now in audience research, which is basically like, I want to understand more about my customers and potential customers. Yeah. So we, you know, with that that thing of like, what's the problem that you solve, and how do you describe both that problem and your solution in a compelling way? That has just been an ongoing battle for us, and and I think never never quite hit. So if you guys have great ideas, I'm I'm open to it. But this has been a you know, it's a real thorny pickle.

Will Ockenden 44:27

It's interesting hearing you explain it like that because actually it's so simple, isn't it? You know, it's it's it's you know what what's unique about yourself and how do you communicate that in an engaged and it's amazing how easy it is to forget that and get caught up in a tweet.

Chris Norton 44:42

In a tweet. That's that was the beauty of Twitter, wasn't it? You had to get it in 140 characters, and really, Rand, you should be able to do your elevator pitch in 140 characters, and I should know exactly what you are, who you are, and what you do.

Rand Fishkin 44:54

Yeah, sometimes, you know, so so I I practice this a bunch, right? Where it's like, hey, you what you what you want to do, right, as a marketer, is is try or a founder, is try and talk to a whole bunch of people who are your target customer and look for the lines where their eyes light up and they they can't wait to do the thing, right? They go from yeah, yeah, yeah. Wait, what? Wait, what did you just say? You you do this? Oh my God, I've been dying for that. That that's the thing you want to find. And it is hard. It's hard to find. Um, I think you got to have a lot of conversations with a lot of people. You got to figure out whether the ones that you're Talking to are a big enough market and the right market and the right match, and whether they have the authority and influence to make the purchase decision and um whether they can convince their boss or team or or clients or whatever to you know jump on board with this idea. It's a it's a challenging thing. Um and sometimes I think one of the one of the big challenges with um SparkToro is that we're we're in a new space, right? Like there's not prior to SparkToro, if you came to me and you're like, hey, Rand, I really need to know what podcasts chemical engineers in the UK listen to, I'd be like, well, we're gonna have a $50,000 you know budget for a big old market research survey, and we're gonna do a bunch of interviews, and we're gonna, you know, um try and do an audit of a whole bunch of podcasts and their audience data and pull this all together. You know, and now you can just be like chemical engineers in the UK, type it into SparkToro and then scroll down to the podcast lists, and it's like here's one through 50. Ta-da.

Will Ockenden 46:40

Um, I've I've got a quick question, Rand. I appreciate we're approaching that the hour mark. Um, this is a little bit like asking somebody who their favorite child is, but you've got three very different businesses. Um Snackbar Studio, Alertmouse, SparkToro, or SparkToro. Um, have you got which which one are you most excited about? You know, which which have you got a favorite? You know, um where's where's where's your where's your where does your true passion lie?

Rand Fishkin 47:05

Yeah, uh let's see. So I SparkToro is um wonderful in a million ways. I would say it is uh where I put the vast majority of my effort. Um and it's certainly you can you can hear like when I talk about zero-click marketing and like trying to change marketers' minds and stuff. This is like my my long-term big mission is to make it such that even though big tech is trying to screw you, you're gonna turn around and and have success despite all their nefarious efforts to own and control the web. That you know, that's a big long mission for me. I I um so it's hard, it's hard not not to say that one, but but I will say you guys, running a video game company is so fun, it is so delightful every day. I here look this morning. Our our animator, Adrian, uh he's he's working on, you know, this is so the video game. If you guys if you guys haven't seen it, I don't know if you played around with uh uh Snackbar at all, but it is you know, you play this um uh chef in a magical version of 1960s Italy, and you uh you you have to fight all these um ingredients, right? Magical ingredients that you bring back to your restaurant. And so, for example, like here you can see he animated the the onion, the little cipola that you're gonna have to fight in the wilds and and bring back to your restaurant so that you can you know make your uh risotto milanese or whatever it is. And it's just delightful. You know, here's a here's a a guy who's working in town on an engine, you know, and he's uh and that yeah, the game is just so enjoyable to work on. The team is wonderful, recruiting for it has been a fascinating experience, and it's super different from B2B software, right? So SparkToro and Alertmouse and Moz, right? My previous company, the whole goal, guys, was how do you remove friction, right? How do you how do you take friction away from the user experience, make it easier for them to solve the problem? When you make a video game, which is also software development, you are trying to intentionally add friction. You are trying to create problems for the user that they can enjoyably solve, and then you complexify those problems into just this right, you know, cadence and and mode. It is it's mind-meltingly uh interesting and and so different. So that's been that's been a total joy to work on. Working with creatives, I absolutely love it, right? Artists and um and engineers, and sound designers, and sound effects people, and animators, and illustrators, and narrative designers. What a what a joy. Um, so I that's certainly that one's the most fun.

Chris Norton 50:13

Yeah, so that gives you joy. What's the most annoying thing in marketing in 2026?

Rand Fishkin 50:19

Historical addiction to attribution. The the CFOs and CEOs and CMOs who who cannot get it out of their head that they have to be able to not just measure a channel, but uh perfectly attribute it or attribute it in some way. And so they refuse to invest in sources of influence that they can't see in a dashboard, right? If this, if a whole bunch of people who listen to this podcast are like, wait a minute, wait a minute, you have a better version of Google Alerts? I'm I'm gonna go Google Alertmouse, and then like we see a spike in signups for Alertmouse. Well, we can't prove that it came from this podcast. It could have been anything, right? Like Alertmouse is gonna get mentioned a bunch of times on the air. Someone posted about it on LinkedIn just this morning. And it's like, well, does that mean Rand should stop talking about it on podcasts? You know, and and that addiction to attribution is just drives me insane. Because here's what happens you hear about Alertmouse on a podcast or any product, you hear about it on a podcast, you you see a sign on the street, your friend tells you about it, uh, you get an email about it, it comes up in an AI answer. And then what's the first thing you do? You go to Google and you search for that brand. And then when you search for that brand, in your analytics, Google gets the credit. And so the boss thinks, hey, let's put more budget into Google. And it drives me bananas. Google did not all branded traffic. Anyone who searches

Attribution Addiction And Branded Search

Rand Fishkin 51:55

for your brand in Google, Google should get zero credit for that because they did not create that demand, right? They are merely a navigation source. They're they're just the road that took you, you know, whatever, whatever street you're on. You know, if you're on uh I don't know, um, Barnaby Street, like Barnaby Street shouldn't get credit for driving traffic to your florist. That's not, it just happened to be there, right? And Google is the same way, they're just a navigation path. So that that's the thing that drives me the most bananas. And for some reason, I think, you know, this these 25 years of people being addicted to attribution and learning to do that and just being basically saying, like, hey, Rand, I agree with you. The zero-click world is real, a lot of branded traffic. Google shouldn't get credit for. But what am I gonna do? I gotta prove it. I gotta prove my marketing, or I'm never gonna get buy-in and investment. I'm sorry, you know, like you make a great case, but we're just we just we're just gonna keep putting money toward paid channels where I know I've got attribution. It just makes me pull out my hair.

Will Ockenden 53:03

Have you been interviewing our customers? I just I've had 25 years of that with PR. We know it's important for reputation. We know reputation's the most important thing about being in business because our share price. Um, but what's the what's the what's the ROI of a press release in the in a national newspaper if AVI AVA isn't real? Oh just all these questions we get asked all the time in PR.

Rand Fishkin 53:30

Hey, do you guys I got I got a weird question for you. Have you found the AI era to be helpful as a as a sort of point of leverage for PR? Because you can say you can basically say, hey, it's not just about the whatever the mention. Even if no one reads this story, AI will read this story and start associating your brand, you know, with the problem that you solve. And and so, like, is that helping your pitches now?

Chris Norton 53:53

Yeah, it is, but there's a lot of um, there's a lot of GEO, AEO, SEO bullshit out there that I think a lot all the same people that sprung up when they were industry and suddenly they're now experts in this. And and we've deliberately tried, we've done a couple of webinars, but we've not really launched massive services and things like that.

Rand Fishkin 54:13

Yeah, I I mean I was gonna say, you know, one thing that I have found to be true for sure, Chris, and that I fully believe in long term, right? We talked about like all these short-term hacks where you, you know, oh, hey, I'm gonna post a bunch of stuff to Reddit or post a bunch of stuff to YouTube or send out a bunch of low-quality press releases, right? And then the AI tools will will put my brand at the top. But I 100% believe that in order to get sophisticated in the long term, what's going to happen is the same PR techniques that got people paying attention, real people paying attention, and that click data and and usage data, that will eventually make its way into the models. And so five years from now, ChatGPT or Claude or whatever replaces them, Google as well, they're not gonna care about all these like sort of spam techniques, but they are gonna care if you got into the publications and media and sources that your audience actually pays attention to. That's the only thing they're gonna care about.

Will Ockenden 55:11

A strong brand and a strong reputation is always gonna be a strong reputation. You can fake it and you can fake it till you make it. You've got to make it, you've got to have a strong reputation. And that is the same in 1998, 2000 when Google came out, and it's when social media came out because you want it to be a high profile and reputation on social, and now it's gonna be so. Yeah, the cat, I suppose it's we should look at it as a great opportunity and not an AI apocalypse. What a great place to end it.

Rand Fishkin 55:41

The older I get, the older I get, the more respect I have for people who did great marketing in 1825 and 1925 and 2025, right? Because because the same principles worked, because human psychology has not changed. We have not evolved as a species all that much in that time. And so a lot of those things of like, hey, build a brand that people know, like, trust, and remember that's easy to spell and easy to say and easy to associate with the problem. Number two, do your positioning so that people know what is the unique value that you provide, right? Over why they should choose you over any other, you know, problem-solving choice. Three, make the problem that you solve clear in the pain points, right? And and repeat that message over and over again in the places where people pay attention. People, where people pay attention. You know, I hacked at SEO for forever and did all sorts of stuff that nobody ever saw, but it influenced Google, so I did it. Now a bunch of people are doing a whole bunch of things that nobody sees, but they influence AI and Google, right? And so people do it. And I it's it's not gonna last, right? Just like all those other tricks in SEO, it's not gonna last.

Chris Norton 56:52

Well, um, thanks so much for coming on the show, Ron. That's really great.

Will Ockenden 56:56

People will wanna people will want to order this book that you're writing. You said it comes out in four. Is there any kind of pre-order or can they register interest in it? I mean, and how do they connect with you as well?

Rand Fishkin 57:05

Yeah, um, yeah, yeah. Pre-orders launched just a couple weeks ago. Um, if you're in the UK uh or outside the United States, I recommend waiting a few more months. I think uh September, it'll be available on like you know, Waterstones and Barnes and Noble and Amazon and all that. And the shipping cost goes way down. Shipping cost right now is insane because of weird Trump stuff. Um, so just I would I would wait to pre-order if you're outside the US. If you're in the US, you can pre-order now and the shipping price is normal. I think it's very low, but everywhere else, wait until September.

Will Ockenden 57:42

And the last question we ask every guest before we go, which is you've been on the show now, um, you know how it works. If you were us, who's the next person you would interview on this show and why?

Rand Fishkin 57:54

Ooh. Okay, uh, I have so many I have so many great people for you, but um I there's a fantastic PR uh here in Seattle or just outside Seattle, Britt Klontz. Do you do you know her? No, no, yeah, Britt. She does the digital PR Explained podcast. Um, I was I was on her show. I'm uh also have become friends with her over the last few years. She used to be at Distilled, the search agency in London. Um and yeah, just a top-notch. The the stories she got are amazing, right? Like what she's done for clients, especially as a solo PR, really impressive.

Book Pre-Order And Guest Picks

Rand Fishkin 58:34

Uh, second person I would talk to, if you haven't already, um emotional targeting author, uh Talia Wolf. She runs uh what's the name of her yeah, Talia's fantastic. Uh wait, her company is called oops, uh GetUplift. And she is she's very charismatic, like just great, you know, great speaker, great thinker. Her book is excellent. Um yeah, I love that. And uh third, if you are looking for folks who really get and know this sort of world of of um AI SEO, and like how are these things being influenced right now? And you want someone to go you know deep down the right rabbit hole of like rag and all that. Um I I have been very impressed with some of the research that Cyrus Shepard has been publishing. Um he's uh he used to work with me at Moz and outstanding guy. Um also fun fact he used to be an actor and he was in a bunch of TV shows and movies. So if you ask him about that, he'll be like, Oh yeah, that's wild.