Audience Is Everything: Rand Fishkin on AI Search, SEO, and the Future of Marketing

Rand Fishkin is the Co‑founder and CEO of SparkToro, a software company he launched in 2018 to make audience and market research more accessible to marketers and creators. He previously co‑founded Moz (originally SEOmoz) in 2003 and helped scale it into a leading SEO software company before stepping down as CEO in 2014. Rand authored Lost and Founder: A Painfully Honest Field Guide to the Startup World, co‑authored The Art of SEO, and is widely recognized for his Whiteboard Friday video series and keynote speaking at over 100 international events.

Here’s a glimpse of what you’ll learn:

[2:16] Rand Fishkin critiques traditional attribution models and explains why organic marketing is hard to measure
[6:49] Lift-based measurement as a smarter alternative to performance attribution
[14:54] How SparkToro uses behavioral and demographic data to map audience attention
[18:24] Using audience insights to escape the trap of incremental paid media optimization
[21:48] Why post-purchase surveys often mislead marketers about actual referral sources
[24:47] Tracking AI adoption by customer segment and platform relevance for better targeting
[30:06] Why link building is outdated in modern SEO and AI-driven search environments
[34:45] Rand’s philosophy on writing emails as if they’ll be leaked — for ethics and accountability.
[39:16] How childhood experiences shaped Rand’s values and leadership style
[41:13] Why business growth doesn’t matter if it’s not tied to personal values and fulfillment
[42:53] The psychological impact of sudden wealth and Rand’s approach to staying grounded
[45:25] How Rand and his wife give back by funding education and supporting others’ dreams

In this episode…

Modern marketers face a harsh reality: The tools they once relied on to track performance — like attribution models — are now deeply flawed. As data becomes increasingly fragmented across devices, platforms, and privacy barriers, understanding what truly drives customer behavior has grown more complicated than ever. How can brands measure success when attribution data can no longer be trusted?

Rand Fishkin, a digital marketing and audience research specialist, offers a compelling case for abandoning rigid attribution models in favor of more human-centric, lift-based measurement. He recommends marketers return to fundamentals and observe customer behavior across multiple touchpoints, test messaging across regions or channels, and evaluate net impact over time. Rand encourages brands to shift focus from granular ROI tracking to identifying where their audience spends time and investing in channels, like podcasts and blogs, where competitors aren't yet visible. He also cautions against over-optimizing for paid media and emphasizes the power of brand presence and share of mind in an era dominated by AI-driven discovery.

In this episode of the Up Arrow Podcast, William Harris interviews Rand Fishkin, Co-founder and CEO of SparkToro, about how marketers can adapt to the changing digital landscape. Rand explores why traditional attribution is breaking down, how audience research unlocks new opportunities, and the flaws in post-purchase surveys. He also dives into AI’s role in SEO, the influence of brand mentions over backlinks, and the importance of leading with transparency and kindness.

Resources mentioned in this episode

Quotable Moments

“If you don't waste some money on marketing, you will always underinvest.”
“We get so much lift over our competition by being willing to invest in experimental and serendipitous channels.”
“Links are just not where it’s at; AI tools don’t really give two craps about them either.”
“One of the best, most loving, rebellious acts you can take is to be on the other side.”
“Write every email as if it’s going to be leaked; it promotes good behavior and ethics.”

Action Steps

Embrace lift-based measurement for campaigns: It helps you understand real business impact without relying on unreliable attribution data.
Invest in brand presence where competitors aren’t: Being early in underutilized channels gives you mindshare and reduces direct competition.
Use audience behavior data to guide strategy: Knowing where your customers spend time ensures smarter, more effective marketing investments.
Reframe marketing experimentation as necessary waste: Accepting some untracked spend allows for innovation and long-term brand growth.
Be present in media your audience trusts: Visibility on podcasts, blogs, and YouTube builds familiarity and trust beyond ad platforms.

Sponsor for this episode

This episode is brought to you by Elumynt. Elumynt is a performance-driven e-commerce marketing agency focused on finding the best opportunities for you to grow and scale your business.

Our paid search, social, and programmatic services have proven to increase traffic and ROAS, allowing you to make more money efficiently.

To learn more, visit www.elumynt.com.

Episode Transcript

Intro: 00:03

Welcome to the Up Arrow Podcast with William Harris, featuring top business leaders sharing strategies and resources to get to the next level. Now let's get started with the show.

William Harris: 00:15

Hey everyone, I'm William Harris, I'm the founder and CEO of Elumynt and the host of the Up Arrow Podcast, where I feature the best minds in e-commerce to help you scale from 10 million to 100 million and beyond as you grow your business and your personal life. Today, I'm joined by one of the most influential voices in modern marketing. If you've ever googled anything about SEO, there's a good chance that today's guest invented the entire playbook that inspired it. At least, that's what I'd like to say. Rand Fishkin is the co-founder of Moz, where he turned a tiny blog into a massive SaaS juggernaut and helped educate an entire generation of marketers along the way, including myself.

He's also the founder of SparkToro, a radically transparent company building powerful tools to help brands understand who their audience actually is. But Rand isn't just a tech founder or a marketing brain. He's a walking masterclass in how to build with empathy. Speak with honesty and scale without selling your soul. He's called out the myths of attribution, championed the rise of owned media, and more recently, sparked some critical conversations about how AI is changing the search landscape forever.

Rand Fishkin It is truly an honor to have you on the Up Arrow Podcast.

Rand Fishkin: 01:18

Good to be here, William. Thank you.

William Harris: 01:20

Yeah, I want to give a quick shout out to the brilliant Briteny Muller, formerly the head of marketing, over at Hugging Face. She's now an AI consultant, killing it there. She was previously on the podcast and she said, you've got to talk to Rand. And I was like, I would love to talk to Rand. So thank you, Briteny, for putting us in touch.

Rand Fishkin: 01:36

She is a great friend and a wonderful person.

William Harris: 01:39

Yeah. Last interruption that we're going to dig into the good stuff. This episode is brought to you by Elumynt. Elumynt is an award winning advertising agency optimizing ecommerce campaigns around profit. In fact, we've helped 13 of our customers get acquired, with the largest one selling for nearly 800,000,001 that ipo'd.

You can learn more on our website at Elumynt, which is spelled e l u m y n t. All right, let's dig in. In another podcast earlier this year, you critiqued modern attribution models and emphasized behavioral observation and empathy. If traditional attribution is dying, how should brands actually measure what's working?

Rand Fishkin: 02:16

Well it depends. There's nuance here. William. I'm I'm not saying like attribution is dead for everyone. If you're in e-commerce and all you're trying to attribute is your paid performance spending, that's pretty good.

Attribution modeling works pretty darn well. But let's say, William, you would like to know what the contribution of PR, organic PR is, or organic content, or organic social media, or SEO, or being on podcasts and YouTube channels. Sure. Right. Influencer marketing creator marketing attribution is a terrible, terrible way to measure or justify or invest because it that data that used to be semi-present, say, ten years ago, like 2014 or 15 is no longer available to any of us.

It's illegal to collect in most jurisdictions, and even where it's legal to do so, like the United States is, is pretty flexible on most of these things, except for California privacy law, of course. Even in the United States, where some of this is possible, there are huge problems with trying to attribute anything organic. You know, the networks not passing referral data to you, you not knowing what keywords people search for in Google anymore. If they're clicking on organic results zero click everything taking things away, multi-device journeys, death of third party cookies, privacy and ad blockers in browsers which are on more than a third of American browsers. Like all this stuff is going to make attribution look possible, but in actuality be false.

And it's going to lead you to basically overspend on Google and Facebook.

William Harris: 04:10

Yeah. So how do we get around that? Like, what are the better ways to start looking at this then?

Rand Fishkin: 04:15

Okay, so William, I don't know how long you've been in the in the marketing universe, but let's let's imagine that you and I were helping Coca-Cola in 1965, right? We're like, we're on Madison Avenue. We got our fancy offices, and, you know, we wear suits and ties to work.

William Harris: 04:33

And what color is my tie?

Rand Fishkin: 04:36

Let's say the least, sexually harassing color. Whatever. Whatever that is.

William Harris: 04:40

Okay. Got it.

Rand Fishkin: 04:41

Yeah, I really that's the one part of that culture that I don't want anything to do with. But, you know, Mad Men era, what we would do is we would say, hey, Coca-Cola, we believe that this messaging is going to really resonate with new customers that you haven't yet captured in the Midwest. And so what we're going to do is we're going to run an outdoor campaign. We're going to buy billboard advertising in Cleveland, Detroit, Cincinnati, maybe Minneapolis, right. And then we're going to run different messages in each of those.

The billboards are going to be different. The messaging is going to be different. And we're going to see where we get the most lift and same store sales year over year, accounting for historic growth patterns and seasonality. And look, the billboard in Detroit outperformed the one in Chicago and Cincinnati and Cleveland and Minneapolis. So we're going to take the Detroit one, and we're going to run it in Chicago next quarter.

And it worked there too. Now we're going to roll it out throughout the Midwest. We outperformed West Mountain states East coast. That's going to be our national campaign. We have no attribution.

There's no.

William Harris: 05:51

Right.

Rand Fishkin: 05:51

Actually, technically there were. I don't know if you ever heard of these guys. There were license plate spotters.

William Harris: 05:56

No.

Rand Fishkin: 05:57

Yeah, yeah. So the attribution, the old school attribution used to be that there were license plate spotters who worked for the ad agency or for Coca-Cola, and they would track the license plates of people driving down the freeway, and then they would have, like, a guy further down the freeway who was at the convenience store, and they'd be like, hey, did he buy any Coke?

William Harris: 06:18

That's crazy. I had never heard of that.

Rand Fishkin: 06:20

Yeah, yeah. So there were very few of these. It didn't work very well, as you can imagine. Like, it's not a it's not a great they didn't have cell phones. Right.

So we're talking about like like you know anyway like people writing them down and then the other guy writing them down and then they compare the two notebooks at the end of the day. So, but. Right. Market marketers have always been trying to do attribution, but lift based measurement, which is essentially what this Mad Men era style of marketing did. You can do that today.

Rand Fishkin: 06:48

You can do that right now. You can say, hey, you know what we are going to invest in, let's say, a content distribution strategy that is focused on getting our content onto other people's websites and blogs and social media channels that that we think our audience pays attention to. Right. And you could use there's all sorts of audience research sources you could use. SparkToro is one of them, but there's plenty of others. And and then you could say, okay, this podcast, this YouTube channel, this blog, this news source in our industry, like we think those are paid attention to by our customers.

We're going to go be present in those places, and then we're going to measure lift of the products that we promoted on those in those places over the next three, six, nine months, depending on what your sales cycle is. And then we're going to compare that against the, the model of like, you know, what our CFO or our accounting team put together for what we thought sales were going to be. And gosh, that delta that is almost certainly happening from that investment that we made. I think unfortunately, like CFOs and financial teams get together and they're like, well, you can't prove that any one person purchased the product because they hurt us on this podcast. To which my response is, just shut up, Dave.

Like, no offense, but if if the numbers are going in the right direction and you're getting lift over what you expected and you invested in these channels, the absolute worst case scenario is that your business is doing better than you expected, and this investment is actually not contributing to it, which is fine. Go waste some money on marketing, because if you don't waste some money on marketing, you will always under invest. You will always be behind people who are willing to make those investments. And this is why you know the famous like quote about I waste half my advertising, but I wish I knew which half.

William Harris: 08:46

Yes. Henry Ford, I think. Right. Yeah.

Rand Fishkin: 08:49

Well, so I don't think I think it was much later than Henry Ford.

William Harris: 08:52

Yeah, you might be right. Okay.

Rand Fishkin: 08:53

But regardless. Right. Like that quote which has dubious sourcing over the years. But that quote is, is very incomplete because the the rest of the quote should read, we're happy to waste half our advertising dollars because we get so much lift over our competition by being willing to invest in experimental and serendipitous and fortuitous channels that our competitors are unwilling to invest in because they're obsessed with attribution, not growth.

William Harris: 09:26

See, this is good. I wrote about this back in, I want to say 2019 2020 whenever ITP 2.3 came out called the Roas Death Spiral. And that's literally what we were doing at that time, is exactly going back to this saying, okay, you might have some, you know, attribution issues that you're having here. But what's not an attribution issue? You know exactly how much money you spent.

You know exactly how much money you made. Like, those two things are absolute facts. If we do this as a holdout test, a holdout test, to your point, you will know what the actual lift is for those things, and then you can use that to kind of back into that data.

Rand Fishkin: 09:59

I think one of the most interesting things from the last 25 years that I have learned for sure is that all of the performance attribution style marketing investments, like the companies that obsess over those things and focused on them universally with, with almost no exceptions, lost out to the folks who threw money at brand and product. If you look at the winners I used to when I worked at Moz, I used to make fun of Super Bowl ads, right? I'd say like, oh my God, you know, they spent $3 million for 30s of Super Bowl advertising. It's the dumbest thing in the world for $3 million. You know how many keywords you could rank number one for in Google?

William Harris: 10:47

Sure. A lot. Who was dumb?

Rand Fishkin: 10:51

I was the dumb one. I was the idiot.

Rand Fishkin: 10:53

The guys on Madison Avenue who are buying the Super Bowl ads, they knew exactly what they were doing, and they were incredibly smart about it. And if you look at the brands that succeeded, they were the ones that invested in hard or impossible to measure channels and didn't give a crap about it because they could see lift.

William Harris: 11:14

What about things even beyond that? Then? I think you've talked about this a little bit, you know, like like dark social or whatever, but just this idea of like word of mouth, you know, where people are actually hanging out. How do you start to understand how. How do you put a value on those things.

So that way you can be more intentional about them?

Rand Fishkin: 11:33

Yeah well, I think some of it is philosophical and some of it is spreadsheet math. Right. So. So the philosophical portion is we want to be present in the places our audience pays attention. And we are willing to devote, let's say, 20, 25, 30% of our budget to being present in those places, even if we cannot measure even lift.

Rand Fishkin: 12:03

Right. But our presence there precludes the presence of competitors. It means that we are always top of mind. It means that our brand will be known and trusted. That's more important to us than measuring any specific return on ad spend. That's the philosophical portion.

And the bad news for spreadsheet marketers is that works extremely well. Most of the most successful brands that you see out there have CMOs and executive teams who believe in that, and that is how they make investments that, you know, all the savviness that you see at whatever P&G or Coca-Cola or brands like that. It comes from that philosophical approach before spreadsheet approach, and that's hard for a lot of digital marketers to wrap their head around. But I, I have to admit that that they beat, they beat us, right? Like they win, we lose.

And then there's the spreadsheet portion where if you want to try and get very, very sophisticated, you can do increasingly interesting things in various categories. Right. So you can, you know, some of the most obvious ones from the past, of course, are like, hey, we have some sort of special offer for people on this podcast. And so we can see some of the direct, you know.

William Harris: 13:27

Yep, yep.

Rand Fishkin: 13:29

You know, measurement. There's billboards. There's billboards now that have location Beam sensors, right? That that sort of track phones that go by and then ping like IP addresses and try and say like, hey, this, you know, this device drove past our billboard. Then we later saw this device in a store then, you know, and because of that, we're going to connect that up to with bank data that you can buy.

Right. We can connect that phone browser behavior to a credit card. And then we can connect that to an address and to a human being. And then we can see that they checked out. And here's how much they purchased.

Those kinds of companies and processes exist, at least in parts of the United States. They exist I think South America, you can do some of that stuff. Europe, you cannot do any of that stuff. If you do like, you will get shut down. Yeah, it's harder in other jurisdictions, but but possible.

William Harris: 14:29

I feel like this is where SparkToro comes in a little bit, at least in my understanding. And this is where I wanted you to kind of walk me through this a little bit. But you talked about it in a LinkedIn post there about the ability to analyze audience behavior tied to search terms, media consumption, demographics. So that way basically like how can brands, how can they identify and target their ideal customers?

Rand Fishkin: 14:51

Yeah, I mean, so SparkToro is shockingly simple software like really, you know, it sounds fancy. You make it sound really fancy.

William Harris: 14:59

Like, I'm trying to do my best here.

Rand Fishkin: 15:03

If we if we used your phrasing, I bet we could sell it for, like, ten times as much. But, you know, instead, I just sort of describe it as like, find where your audience is, right? So you go to SparkToro and you can search in a few different ways, but you could search, for example, for people who visit a particular website. So like, you know, people who visit Architectural Digest, maybe you're you're selling stuff to architects. And so you're like, hey, I want to see give me the demographics and behavioral data about people who visit.

Architectural digest. The magazine turned website. And then you can see like, oh, okay, here's sort of the geographic breakdown of where their visitors are coming from. And here's an analysis of what other websites they visit. Here's the podcast that that audience listens to.

Here's the YouTube channels they subscribe to. Here's the subreddits that they follow. Here's the press and media sources that are influential to them. These are the search and AI tools that they're using and in what proportion. Like, yeah, of course everybody's using Google.

But like, oh, ChatGPT is less popular with this audience than it is with the average American. And perplexity is slightly more popular, you know, whatever it is. And so that that type of data to build a either a persona, right, a marketing persona, or to make a pitch to your team or your boss, or if you're an agency, a client is, is, or to export and then upload to your favorite LLM and be like, hey, brainstorm a bunch of whatever content marketing ideas for that. Reach this audience that, you know, target the search terms that this this group is searching for that we don't have on our website already. And you can use the new thing, you know, learning models.

Great. Yeah. People use SparkToro to do all those kinds of things, but at the core it's just a combination of of three sources of data. It's clickstream data. So you're probably familiar with clickstream panels where they basically like have, you know, millions of devices and they look at every, you know, every URL that the browser visits on, on a phone or a desktop.

And then they anonymize and aggregate those and sell them to people like us. And then we combine that with LinkedIn profile data. So we have a few hundred million LinkedIn profiles that we sort of, you know, I want to say I want to say munged together like Casey does fancy math and statistics and stuff to get get those profiles to match up to, like the behavioral stuff. And then we layer on top Google search data. So like from hundreds of millions of search results, you know, what are the words and phrases people search for and what are the websites that come up for those and questions people ask, blah blah, blah, blah, blah.

Those three sources essentially mash together with a few other little things in there. And that's how we can say these are the behaviors and demographics of nearly any describable audience.

William Harris: 17:52

When I think about brands that get stuck, e-commerce brands especially, they get stuck at that 10 million, 20 million, $30 million range. They want to get to $100 million. One of the things that I talked to them about is, you know, finding out what are the tests that are going to move the needle instead of saying, hey, how do we add $5,000 a month more? How can we add $500,000 a month more to whatever we're doing, right? And one of the things that I feel like it's easy to get stuck in is hyper optimizing.

Maybe your meta campaigns and you're like, if we do this, you know, then we're going to get this 0.1% more incremental growth, and then.

Rand Fishkin: 18:24

That's maxima versus global maxima.

William Harris: 18:26

Yeah yeah yeah yeah. And that's fine right. Like that's okay to still continue to optimize there. But it might be more a matter of where else are your customers at. And like you said, it's like if you can understand it a little bit better, you might say, that's fine.

We can incrementally grow this 0.1%, or we can go over here like you just called out. Maybe they're spending more time on these podcasts. We're not doing anything on podcast advertising. We need to do some podcast advertising. You know, Athletic Greens has done a fantastic job with that.

Right. So how are some ways we can say, where are our customers hanging out outside of these dominant ad platforms and spend more time there? SparkToro can help unlock some of those data insights for me.

Rand Fishkin: 19:04

Yeah. And obviously we're not the we're not the only way to go about this, I think. Sure. One of the things I've seen ecommerce brands do, speaking of, we talked in the in the pre show about stories. So one of the early SparkToro customers was a a consumer brand in the not in beverage space but in food and beverage world and they their marketing methodology you know for for this kind of stuff was post consumer search post consumer purchase surveys.

Right. So I'm sure you've seen this William. Right. Like you check out online and then like the website says, hey, how did you hear about us? And then you, you know, check a list and you're like, oh, it was from YouTube.

And then they'll say, which YouTube influencer or channel or whatever. And then you can select or you can enter your own or whatever. So there's two, two big problems with this that I, that I don't love. One one is if you market to your point, if you market in places where people have already heard of you, you cannot grow. Right?

So if you're like, oh great, this YouTube channel brought us these visits, let's go do more of that. Well, I'm not saying don't do it, but recognize that that's already sending you people?

William Harris: 20:22

Sure.

Rand Fishkin: 20:23

You're always going to miss the ones that are sending traffic and, you know, visits and purchases to your competitors, right? And the second problem, of course, with this is that data is pretty terrible. Like, no offense to human beings, but, you know, if we were like an alien race, we'd say, oh, well, humans are terrible at at recalling, you know, events that led them to take actions. And sure enough, you can prove this. So anyway, this, this food and beverage company that I was working with in the early days of SparkToro, I convinced them, William, I was like, hey, would you guys be willing to make some changes to your survey, like your post-purchase survey?

They're like, well, maybe tell us what you want to do. I was like, okay, here's my theory. My theory is that people like really don't know and understand this. And in order to prove this to sort of your, you know, your boss and your team, will you do? You know, they have a in the drop down.

One of them is was like influencer or influencer creator or whatever. And so I said, hey, would you be willing to in the influencer creator box? Because that was one of the big ones that people checked put in. I want to say it was like Martha Stewart and Guy Fieri and like, you know, a couple of like famous Food Network food TV type of people. And of course, the response was, but those people have never talked about our product.

And I said, exactly.

William Harris: 21:48

Ooh.

Rand Fishkin: 21:49

Guess what happened.

William Harris: 21:51

You got a whole bunch of responses from people like that.

Rand Fishkin: 21:53

Almost everyone who clicked the influencer box chose one of the. I think we put in three famous food influencers who had never mentioned them. And so then they took that, of course, to their boss and to the team and were like, okay, post-purchase data, like survey data. We can't trust it, right? It's not it's just not accurate.

Like people are not they can't remember where they heard about you first or last, or what sent them to you. I think one of the other big frustrations, this, I'm sure, is how a ton of CEOs dine out, myself included, for years. Which is that what happens when, let's say you and I talk about a product I don't know, like like, have you heard of CS6 shoes?

William Harris: 22:41

Yes. Those guys. Absolutely.

Rand Fishkin: 22:43

Yeah. So I like them a bunch. They're they've got a there's a pair that goes great with this cardigan I'm wearing. Nice. And so like I, I love that they're slip in and out.

I got some for my grandfather like my my wife is into them. Whatever. So someone hears us talking about CS6. They almost certainly will go to Google or Bing or something. And they'll search for kayak and then they'll go to CS6 website.

And if CS6 marketing team is focused on attribution tracking and analytics. They will say, wow, Google sent us yet another purchase. Google didn't send you shit, right? Google is just an intermediary. They're a traffic cop, right?

Like they said, go down the street.

William Harris: 23:32

Yeah.

Rand Fishkin: 23:32

And yet they're the ones who get all the investment when you when you look at how Google like, absolutely dominates the advertising landscape. This is one of the big reasons is because everything that happens on every other channel turns into a Google brand search, which are some of the highest, best. Converting my conspiracy theory. Dude, I think Google took away keyword data because when they sent you keyword data, it showed that your brand was the primary driver of like a, you know, of an overwhelming amount of your conversions. And Google was like, wait a minute, wait a minute.

If we don't tell them that they're going to spend money with us.

William Harris: 24:16

That's wild. That's an interesting conspiracy theory that maybe feels intuitive. They're getting a little bit of their own medicine, though, because they're having a little bit of their lunch eaten here by ChatGPT and perplexity and things like that. You called this out a little bit ago, just, you know, a minute ago to one of the things that you actually tweeted about this, I think, back in July, is not to not to reframe the question, not saying are people switching from Google to ChatGPT, but are my customers switching?

Rand Fishkin: 24:47

Yes, I think I mean, look, this I think this is the question every marketer, every brand, every founder should be asking about their audience generally. Right? There's there's sort of the interesting like, hey, is AI on the rise broadly with the population? That can be interesting. Same thing with social networks, right?

Like TikTok. Okay. TikTok is huge, but is it is it huge with your customers, right. And your More audience. What about Instagram?

What about Reddit? What about Facebook? What about LinkedIn? What about YouTube? What about.

Yeah, perplexity and Bard and ChatGPT. What about DuckDuckGo? Like, do you have I know people who are in this cybersecurity space and privacy-focused browsers, privacy-focused search engines, privacy-focused networks like Mastodon are big with their audience, and so it's less about where's everybody? And it's more about where are my people? And I think the other confusing thing, the other thing that gets frustrating is if they're on that platform or network or website or whatever, are they doing things there that are relevant to my world?

One of the things that I've seen time and time again is that, you know, someone will say, hey, our audience is whatever. Like we serve a lot of people between 21 and 35. Like we're very youth culture focused. So TikTok is our is like a network where we're going to pay a bunch of attention, and then they get no traction, no attention, because their messaging, their marketing, their brand, their product isn't a match for TikTok's algorithm.

William Harris: 26:28

Yeah. That happens. You're right.

Rand Fishkin: 26:31

Yeah. And then and then they're like, gosh, we're when we do a little bit on YouTube, we do really well. Why can't we get that to work on TikTok? And the answer is like, you got to match up both these things. Is your audience there?

And are you uniquely good at reaching people through that medium because they actually care about it on that platform? Like people use Reddit for a different reason than TikTok, than different reason than YouTube, different reason than LinkedIn. So make sure you're, you know, stacking up all the checking, all those boxes.

William Harris: 27:02

Yeah. Well, relevance is so important in every area of life, right? Like where you are when a conversation comes up, you're like, I might actually be interested in XYZ, but at this moment I'm not. I'm at work, right? And a good example could even be it's like if your kid, you know, you're you're working from home or something, your kid's like, hey, do you want mac and cheese?

And you're like, I mean, sometimes, yes, but it's 3:00 in the afternoon and I'm jumping into a podcast. Not right now, but like, context matters.

Rand Fishkin: 27:29

Yeah. Yeah. And I mean, I think that the, the really interesting thing about audience research today, this is this is a flaw of SparkToro. I'll actually say, right, is that if you look at the data and you see, for example, gosh, our audience is using ChatGPT as a good example, using ChatGPT and AI tools more than average, right? More than the average American web browser by let's say 60% or something, I would question whether your immediate response should be, hey, let's go and do a bunch of sort of PR and brand work to get associated with words on the internet that people might be using in ChatGPT before you do that.

I might be interrogating the hey, what are people doing with AI in our sector? And is that actually relevant? Is that something that is going to be useful and productive for us to be in those conversations? Like, are they actually searching for, you know, whatever it is, beverages they're going to buy at the convenience store on ChatGPT? I'd question that if if it's something like, hey, are they looking for.

You know, a new like vitamin supplement to, to add to their regime around like health stuff maybe like I does get used for more of those kinds of things. But again, I do I do some real interrogation of the data behind that before I'd make that investment. And SparkToro doesn't show you that. It just shows you literally what's happening.

William Harris: 29:07

Okay. I may be digging up some old stuff here, because I'm going to go back into the SEO world for a little bit, but I'm going to call it SEO, the AI optimization. And there's a whole lot of, you know, debate about what we're going to call that. I don't want it to be called GE. Oh, by the way, because I always talk about geo holdouts already.

I don't want it to be Geo. But anyway, so.

Rand Fishkin: 29:27

Geo stands for goose egg omelet I don't know.

William Harris: 29:32

I haven't heard that one, but I like that one too. But okay. So again, going back to the Maz days a little bit, I don't know if you remember this. I had a an article that I wrote called the SEO Food Pyramid and you commented on it. This was the first post that I put on GrowthHacker.com you commented on it. I was like, yes, like Rand knows that I exist. This is a really good thing. And so I have to still start a little bit with the SEO stuff. What do you think are the most common SEO and let's say AI optimization mistakes that e-commerce brands make right now.

Rand Fishkin: 30:06

Oh, there's definitely there's still a weird belief in link building. I look, I mean, I don't want to rain on people's Parade. I know a lot of people still make a living trying to build links, but it it really looks like from the DOJ investigation and like the Google leak from last year, you know, that that folks looked into and I got I got to break the story. But I wasn't like in depth looking at it. The reality seems to be that Google cares less about links than they ever have, and AI tools don't really give two craps about them either.

Yeah, right. We are. We are in a world where the content is what's being analyzed and the user usage and behavior is being analyzed. And then there's a lot of model tuning and vector space. Models and that kind of stuff.

But links are just not where it's at. And so look, if you are trying to rank higher in Google, if you're trying to appear in AI tools, when people, you know bring up certain words and phrases, links are not the path, my friend.

William Harris: 31:25

Sure. Mentions still worthwhile though, right? PR mentions you called that out. So PR mentions fine links. Not as important anymore.

Rand Fishkin: 31:34

Yeah, yeah. I mean, I know you know how this works, right? But like the, the vector space model that all AI tools are based on is essentially like a Amanda and I like to call it spicy autocomplete. Right. Because it's essentially that.

Right. It's like words that frequently come after other words in documents on the internet like that, and books and other stuff that has been scanned and used by whichever, you know, Claude or OpenAI or whoever. And what that model does is essentially say, like, you know, if you search for soda, well, let's say you say, what are popular soda brands in the United States the way that that ChatGPT or Gemini or, you know, Claude, behind the scenes, the way that they're determining what what answer to give you are they look for documents that contain soda and beverages and other synonyms and, you know, are relevant to this topic and, and on this. And then they sort of like crunch. They have they have that, that all those words like crunch together in, in multiple dimensional vector space, which you don't have to worry about how that works.

But essentially it's like, okay, soda beverage, Coca Cola often appear together. So there's a high probability that Coca-Cola will be one of, if not the first answer After almost every time you do this, because the probabilistic system is going to say that's the that's the thing there. And how do you influence that? How do you change that if you're, you know, whatever doctor Pepper or your. What are some of the new beverages like pop pop or you know.

William Harris: 33:15

Yeah. Poppy. Yeah yeah yeah yeah.

Rand Fishkin: 33:17

Like if.

William Harris: 33:17

You're. Yeah.

Rand Fishkin: 33:18

If you're one of those guys and you want to appear high on that list, the way to change that is to get press, media, blogs, podcasts, YouTube channels, social media accounts, all the stuff that's getting indexed by by these models. They really care about recency and frequency. So if in the past 30 to 90 days, you know, Poppy is getting way more mentions alongside soda and beverages and most popular beverages and popular US beverages and best-selling beverages. United States, like in all those places, all those documents, it is very likely that now they will be the one who is mentioned at the top, even higher than Coca-Cola, because the historic mentions might say Coke is the one that's that's there the most. But the recent mentions are putting this new brand up there.

That's how you influence it.

William Harris: 34:15

Yeah, yeah, I think it's wild to see that. I think you're absolutely right to call that out. We did talk about that with Brittany Muller a little bit in the vector. The way that the vectors sound is.

Rand Fishkin: 34:23

Yeah. She understands this stuff like way deeper than I do. Yeah.

William Harris: 34:26

It's so crazy. I want to talk to you a little bit about. There was something else that you said that I really liked, which was if every email should be written as if it was going to be leaked. And again, you're transparent person. We talked about at the beginning, why should you write every email as if it's going to be leaked?

Rand Fishkin: 34:45

I think there's I think there's two reasons behind it. One is that it nullifies risk, like it removes a lot of risk around your organization. If you're worried about ever being hacked, or you're worried about someone leaving your organization in anger and, you know, sharing an email, or you're worried about your own, your own behavior, right? Like, man, am I pissed at this person or this event? And I'm tired and I'm hungry.

And you know what? All the swear words and terrible things that you know are going on in my brain, I just want to put them into an email and send it thinking about, wait a minute, how will the internet, the broader population, like react and respond to this? It's just a great way to prevent that risk from existing. And the second reason is because it promotes good behavior. Like if you if you are considering, how would this be perceived?

You will also be more likely to make moral and ethical decisions just like full, you know, full stop. That is a great reason to do it even if there was no risk of leakage. That's a great reason. And this is you know, this is. I recognize it.

We are living in a weird time where it seems like nothing matters. And no, no amount of, you know, whatever flying children to islands for, you know, horrible acts by politicians seems to make any, any difference. But I would I would argue that that is going to change at some point. Like it seesaws back and forth. Right.

The people's perception of morality and ethics. And you can see that even though, you know, you might say like, hey, whatever. Like politicians and billionaires are not being held to this standard. People who buy from companies do hold standards. Boards of directors do hold standards.

Employees hold standards. Your coworkers hold standards. Those people do care, and you should care about it as well. So that's why I. I believe in that.

William Harris: 36:59

I like that you called out that it builds the right habits to your point too. Like, I really like that too. I, I'm still somebody who says please and thank you or whatever even to like I and one of my reasons is because I was like, I'd never want to become so accustomed to not being polite that to even to something that almost feels sentient, that like now that becomes the habit. And the way that I talk to my fellow human beings, I want to continue to make sure that I build the right habits.

Rand Fishkin: 37:24

Yeah. I mean, politeness is that's how we show respect to other people, right? It's just a and I think it's a wonderful habit to get into. And there's nothing I don't know. To my mind, there's there's nothing better than being in a world surrounded by good people who care about other people.

And you know what? In a world where that seems to be going away, one of the best, most loving, rebellious acts you can take is to be on the other side of it.

William Harris: 37:51

I like that. I've heard before that every generation rebels like that's almost like a thing that we just assume, right? Like, I don't know if it's been proven, but it feels fairly intuitive when you look at it. The kids like to rebel against their parents. And so to your point, whenever there is something that maybe frustrates me about what's going on, I look at it and I say, there's a good chance that the group that's growing up is also frustrated, and they're likely enough of them that we're going to rebel against that.

And that's a good thing.

Rand Fishkin: 38:16

I surely hope so. I mean, I'll definitely say like one of my one of my saddest days was post US 2024 election, looking at the demographic breakdown of voting patterns and seeing that for the first time in history recorded statistic history, American youth swung to the right, which that historically, you know, it's always been like the younger you are, the more left you swing. And this was like the rare, rare exception to that rule. And so I sort of temporarily lost all hope in my country. But we'll see.

We'll see if that changes in the future.

William Harris: 38:56

One of the things that I appreciate about you is, is how, like you said, transparent you are about the way that you are thinking about things. And one of the things that I've read about you is that you had a tough childhood growing up. How do you think that has helped shape you to be the leader that you are today?

Rand Fishkin: 39:16

Okay, first off, I want to clarify my. Oh, sure.

William Harris: 39:21

Sure, sure.

Rand Fishkin: 39:22

Yeah. Like my childhood, so many, so many people had such worst childhoods. Absolutely. Like, you know, my parents were pretty middle-class. Like, my dad was very, very obsessed with saving money.

So extremely, extremely frugal in some, some wacky ways. There's there's stories, but. Yeah. I mean, my parents definitely loved me there. There were two of them, right?

Like they were together? Yeah. I had a brother and sister who, you know, loved me and cared for me a lot. We had lots of our problems and challenges, and my dad was not physically abusive, but. But quite emotionally, at least problematic.

If not. If not, you know, seriously abusive and not great to my mom in my opinion. Anyway. Blah blah blah. Yeah.

But yeah, I think part of my speaking of rebellion, right. Part of my rebellion was like being a little kid and, you know, going to my room a lot after something had happened with my parents and being like, man, I that that wasn't right. I don't want I don't want to ever be like that. And yeah, I think that that shaped a lot of who I am and how I am. I'm weird.

William. I don't want to, like.

William Harris: 40:36

I like weird.

Rand Fishkin: 40:37

I don't want to rain on your parade or, like, the position. But you know, when you said to me at the start of the podcast like, hey, you know, we help businesses grow from, from tens of millions to hundreds of millions or, you know, 10 to 100 million in revenue. I was like, is that what matters?

William Harris: 40:55

It doesn't.

Rand Fishkin: 40:56

Is that what.

William Harris: 40:57

It doesn't matter.

Rand Fishkin: 40:58

Yeah.

William Harris: 40:59

Let me let me let me call that out. It doesn't matter. But that is the role that I'm playing because there are some people who that is what they are looking for. That is what they want. And I will help them educate that ranking high in Google also doesn't matter.

It doesn't.

Rand Fishkin: 41:13

Doesn't.

William Harris: 41:13

Matter. But like there are reasons. And so I would say we've actually done this on a couple of podcasts where we've talked about it's like, what if that's not the goal? What if your goal is to do almost the opposite, right. And so I would say that is like from a important in life thing growing from 10 million to 200 million, going from 1 million to 10 million.

Doesn't matter if it does matter to you for one reason or another. We want to be the ones to help you do it.

Rand Fishkin: 41:35

Right. Yeah. But like, if you're if you're on that journey, you can help. This podcast can help. I think that's a great thing. I would, I would caution. Here's one thing I'd say, which is that I don't know, I don't know about, you know, your network.

I suspect you have people in your network who are like this, right? But, you know, Geraldine and I, my wife and I have because we're in, like, Seattle and in tech, right? Like, we have friends who've gone from net worth of next to nothing to millions. Most of them stayed cool. We have people who've gone from, you know, into networth of like 10 million plus.

Most of them have stayed cool. And then we have network of like a, a smaller group of people who've gone to 50 or 100 or more. Almost none of them have been cool. Like when I say cool, I mean, like people you'd want to continue to hang out with and the conversations you have with them are kind and friendly and normal, and they can relate to their fellow human beings and their sort of ethics stay good. And I don't know, it just it worries me, I think.

I think there's, you know, there's something to that old like Bible adage about, you know, the camel fitting through the eye of a needle.

William Harris: 42:53

Absolutely.

Rand Fishkin: 42:54

And yeah. And I just I just caution anyone like, you know, a few years ago, my, my old company, Moz sold. I was not on the board of directors anymore. I wasn't part of the company I had left in 2018. It sold in 2021.

But, you know, we I got a phone call and then a few months later, I got a giant lottery winning size check in the mail. Right. And. I was scared. I was scared I was going to become a shitty person.

Yeah. And and part of part of fighting. Fighting back against that. Like rebelling against that. I'm sure this has something to do with my childhood, too.

Is is like, okay, well, how do we help out a bunch of people give a bunch of way, you know, put it put it in. We found a financial advisor who sort of manages the money. I don't want to manage the money. And I don't find that fun and interesting. So, yeah, like doing that kind of stuff and then being able to, like, we just went to our niece's graduation whose college we were able to pay for.

William Harris: 44:02

Wow.

Rand Fishkin: 44:02

You know, because her parents couldn't. And we have like, I think we're going to be paying for like half a dozen other people's college. Like, it's it's awesome, you know, like that. That kind of stuff is, is a great like I feel a lot of wonderful things about that. And I feel very grateful that, you know, that money exists.

You know, we I've started multiple other companies and projects, and it's really nice to go out to dinner and be able to pay for your friends who couldn't afford it, you know, like, it's it's really cool. But yeah, it's just it's just thinking about that, like goals and big long term things. And what are you trying to accomplish? I don't want to take away from the value of money. I recognize, you know, like, Geraldine grew up broke.

I grew up with my parents acting like we were broke. And man it, America has gotten way harder since then, right? Since the 80s and 90s. And so I'm not going to rain on anyone's parade who's trying to, like, make enough money to send their kids to college and, like, try and help out. I want to help entrepreneurs succeed.

I love help like you. I love helping founders, like, get their business to that next level. And also I just I just want to have that, like caution about what happens to you and the the great person that you might be today when a, when a big check comes in.

William Harris: 45:25

Yeah, I'm glad that you called that out. It is something that I'm trying to be as conscious and aware of as well. One of our motto. It's not called a motto, but I can't think of what it's actually called, like our business slogan or whatever is, is to amplify joy through profitable business growth. But amplify joy is the first part, because there are going to be times where you might not grow the business, right?

But it's what you're coming to me to, to grow the business, and we're going to do everything we can. Covid could happen. A lot of things can happen. But ultimately I want to amplify that joy. Meaning there's a really good picture painting by Richard Sergeant where it's four different pieces and he goes, it's called anger transference.

And it's it's this boss yelling at this guy, and then the guy's yelling at his wife, and then the wife's yelling at the kid, and then the kid's yelling at the cat. The anger transferred from one person to the next, the same thing that happens to joy. And if somebody goes out of their way to do something nice for you, you almost can't help but want to do something nice to the person next to you. Literally, you might a stranger and you're just like, I don't want to do something nice for them, and I want to amplify that joy. So the way that we interact with people that we're talking about, that is even more important than taking you from 10 to 100 million.

Rand Fishkin: 46:39

I couldn't have loved this conversation anymore. This is. Yeah. I think I think one of the most beautiful things in this world is to recognize that pattern and to invest in kindness and joy and amplification of those things as a result. And the other beautiful thing to do is to take terrible things that happen to you and turn them around.

William Harris: 47:02

Yeah.

Rand Fishkin: 47:03

Right. Which which is harder. Like it is really harder because of exactly this, you know, this this sergeant piece that you're describing, which I just pulled up and was looking at. But yeah, I mean, like the my I remember being a teenager, being in my early 20s and having those whatever, you know, something didn't go my way. And I wanted to react the way my dad did, which is just to, you know, explode and yell at everyone and blame other people and and take out my anger on the people that I love the most.

Which which, you know, was like a big tendency of his and. Eventually getting to a place where I could recognize that and say something that bothers me a lot. I'm going to take that in and I'm going to, like, turn it into love back out.

William Harris: 47:55

I like that.

Rand Fishkin: 47:56

Or or I'm going to ignore it. Right? Like, you know, there's there's some people who, you know, whatever they but.

William Harris: 48:04

That energy exists, that energy exists and that something needs to be done with it. And it's very hard to just metabolize it. So if you could turn it into something else, that's better.

Rand Fishkin: 48:13

Yeah, and I think this is true in life, but it's also true in the business world, right? So like a bunch of people this is this is a like marketing example of this. But a bunch of people have been seeing lately have said, hey, my traffic is dying. I need to get my like. My traffic is dying to my website.

How do I get my traffic back? And my answer has been my friend. It's gone. You cannot take this thing that used to exist and like and, like, claw it back from these networks whose executives are incentivized not to send you traffic anymore. Sure.

What you can do is turn that into an insight. It's not just my traffic that's going away. It's everybody's traffic that's going away. Instead of getting angry and getting even and yelling at my marketing team and, you know, going and pounding the street for a new consultant and firing my old one, I'm going to say, wait, what? What's the structural change that's going on on the internet?

And how do I take advantage of this next phase of the world, which is essentially traffic is dying, but sales are not dying.

William Harris: 49:21

Right.

Rand Fishkin: 49:21

You just have to influence people in the places they pay attention in the places that get the traffic and are hoarding the traffic and then have them come to you kind of, you know, when they're ready to buy, but not much before. It's just a different it's a different kind of world from the first, you know, 25 years of the internet being able to, as you say, metabolize that and internalize and turn it from something, you know, reactive, reactive and reactionary into something, into learning. That's that's a superpower.

William Harris: 49:56

Another way to deal with some of those things that you and I share. You don't know that yet because I haven't told you about it. But like, you like to cook. And I want to know when when did this start? And like, what's your favorite recipe?

Like, if I was coming over and you're like, dude, I gotta show off. Here's my, here's my show off recipe.

Rand Fishkin: 50:17

Okay. My my show off recipe. I used to have like, a few things that I was really good at. Now I'd say I'm more of a I don't know. Poly chef.

I can I can make almost anything competently. And so what I love to do is ask people about their favorite meals, favorite dishes, things that they love from whatever someplace that they moved from. You know, I have friends from Chicago who were like, hey, we want to Chicago beef sandwich. And I was like, oh, okay. I'll, you know, I'll, I'll research that I homemade the jardineria, you know.

William Harris: 50:56

Yeah.

Rand Fishkin: 50:57

Got the right like got the right cut of beef from my butcher and then, you know, made it the day before and then sliced it and whatever prepared it the next day. That was, that was an intense amount of work. Oh my.

William Harris: 51:08

God.

Rand Fishkin: 51:09

But here's here's why I love cooking. Generally, it is an act of deep care.

Rand Fishkin: 51:17

Yeah. Like it is a way to show kindness with your time and effort and an energy that is, it's not performative. It's extremely human. There's there's something that is both art and science to it. And I love the art and science of everything, and almost every human being can appreciate it. Like most everyone has things that they love about food.

Food was a part of their childhood, or it's a part of their relationship, or it's a part of their history, their geography, their ethnicity and background. I think that's I think that's beautiful. And so I love being able to cook and, you know, impress someone with something that they love. I work around a lot of dietary requirements because I live in Seattle and people in the northwest have a lot of dietary issues. So yeah, today, today after this call, I am making carnitas with Like a a green of mole verde sauce for some friends tonight.

So yeah, it's going to be like a bunch of hours of braising. Good thing it's only a 75 today in Seattle.

William Harris: 52:33

Yeah, no, that sounds fantastic, though. I feel the same way. So it was only I. I don't think I cooked much more than mac and cheese up until about two years ago. Oh, and then my, my wife started coaching volleyball.

And it was it took place right during dinner time. And so it was like, well, I had to figure out how I was going to cook for our kids and everything. I was like, well, if I'm going to cook, I gotta cook my way, right? Like I gotta cook. And so it's like I just started like it turned into a creative outlet, kind of like you said, where it was like, oh, this is fun.

And I really got into it. And I've been cooking all kinds of stuff. I, I cooked like a chicken with Ethiopian lentils. Right. Like some really interesting things.

And to your point, it's like everybody feels really special when you can cook something interesting for them. And so I've I've really grown to appreciate it. Okay, so if I was cooking for you, though, you were coming over. So you said you'd want to research. What?

Everybody else. What would you be like? Oh, this would be special if you made this.

Rand Fishkin: 53:29

Oh, well, gosh, I mean, my. You know, my childhood was not a lot of traditional foods. I would say we were like a, you know, very 80s Americana family. And so a lot of lot of frozen and microwave stuff. But.

But I there were, there were two things that really stood out. One was my grandparents. Well, all four of my grandparents are are Jewish and traditional Jewish dishes. Right. Like latkes or beef brisket or.

William Harris: 54:08

Yeah.

Rand Fishkin: 54:09

You know, I.

William Harris: 54:12

When you say the brisket, I can't help but think of the Big Bang Theory. Do you know the. Do you know the. The show?

Rand Fishkin: 54:19

I haven't seen it, but I'm aware of it.

William Harris: 54:21

Yeah, yeah, yeah, yeah. Well, the one Jewish guy, Howard, he's always like his mom's always got, like, a fresh brisket. Always. And so it's just. That's good.

Rand Fishkin: 54:29

Great bagels. Right?

William Harris: 54:30

Like I love.

Rand Fishkin: 54:31

Bagels, but the. Yeah. The other thing, my mom used to make crepes for us, like on a very special weekend morning.

William Harris: 54:40

Savory or sweet?

Rand Fishkin: 54:43

Sweet. Usually. Usually sweet. And so with, like, you know, lemon and sugar or or, you know, a fresh jam or something like made with cherries. And that, that was definitely like a very special meal.

When I was in Paris, I was like, ooh, crepes. Yeah, hook me up.

William Harris: 55:01

All right, last question. And then I know I got to wrap it up here. What's some of the best advice that Geraldine has given you that you didn't take it first and you're like, yeah, you were right. I should have taken that.

Rand Fishkin: 55:13

She. She told me to lean into things that I love. She was like, just. Even if it seems impractical and it doesn't seem like the right time and you know, you're overwhelmed with other things. Just do these things that you are deeply passionate about.

So that is that is how. Snack bar. Snack bar. Studio. One of the other companies that I run came about, which is snack bar is bigger than SparkToro now.

William Harris: 55:44

Really?

Rand Fishkin: 55:45

Yeah. Raised. Raised almost twice as much money using, you know, this unique funding structure that I like that doesn't sort of get you into VC world. And we have a team of eight people now and probably going to be growing more soon. So yeah, the reason I started this video game studio and basically created this game is because Geraldine said, don't wait.

Don't wait until SparkToro is done. Don't wait until you're, you know, who knows in your 50s or 60s or something. And maybe you won't have the energy. Maybe you won't have the passion. Maybe the world will have moved on.

Maybe people won't be, you know, buying games on steam to the same degree that they are now. And the indie game scene won't be as hot. Do it now. Like do this thing that you love. And I think that, yeah, that is that is deep care and kindness from a partner.

Right. Like being able being passionate about seeing their dreams through. I hope I hope that I do a good job living up to that in the other direction too, and helping her dreams come true.

William Harris: 56:58

That is beautiful, Rand. As I said before, it's been an absolute honor to have you on the show. I love that you've shared your time, your wisdom with us. If people want to work with you or follow you, what's the best way for them to get in touch and follow you.

Rand Fishkin: 57:12

Yeah, I am most active these days on on BlueSky where I'm at Rand Fish or LinkedIn it’s Rand Fishkin. And if you want to check out SparkToro and try that for free, you can just go to Sparktoro.com. And if you're interested in playing a chef in a magical version of 1960s Italy and, you know, fighting magical boars out in the forest and bringing back your pancetta to your kitchen to to make your spaghetti alla carbonara, you can check out Snackbar studio.com and sign up for the email list there.

William Harris: 57:45

I'm so hungry now. Rand, it has been so good talking to you. Thank you for sharing your time today and thank you everyone for listening. Hope you have a great rest of your day.

Rand Fishkin: 57:54

Thanks for having me, William. Take care.

Outro: 57:56

Thanks for listening to the Up Arrow Podcast with William Harris. We'll see you again next time, and be sure to click subscribe to get future episodes.

‍