← Social Media Marketing Podcast
Social Media Marketing Podcast
Zero-Click Challenge: Why Marketers Must Go Back to Fundamentals
00:50:16
Loading summary
Transcript97 lines
- [00:00]
A
After 13 years, we are making a bold move. We've been hosting thousands of marketers in San Diego, California since 2013 for social media Marketing World, and it has been epic. Here's the news. We're making a bold move to anaheim, California in 2026. The decision came down to three very compelling factors. First of all, the location advantages are incredible. Orange County's central location means easier access for our growing international audience. There are five different airports that service the area, including the Los Angeles airport, which means it's a lot more economical for you to get here. Second, the weather. April in Anaheim is absolutely perfect. 75 degrees, sunny skies, less chance of the random rainstorms that sometimes surprise us in San Diego. And third, this is really the big one. And Disneyland, it's only a 20 minute walk away from our venue. Imagine being able to experience Disneyland with your new friends that you make at the conference, literally after the event. I did this with one of my brand new employees. It was an incredible experience. Talk about developing lifelong relationships. This is the way to do it. This is the year to finally come and experience the magic that is Social Media Marketing World. Grab your tickets right now because we have a really big sale going on. Visit social media marketing world.info I can't wait to see you there.
- [01:32]
B
Welcome to the Social Media Marketing Podcast, helping you navigate the social media jungle. And now, here is your host, Michael Stelzner.
- [01:42]
A
Hello. Hello, Hello. Thank you so much for joining me for the Social Media Marketing podcast brought to you by Social Media Examiner. I'm your host, Michael Stelzner. This is the podcast for marketers and business owners who want more exposure, more leads, and more sales. And today we're going to talk about a really fascinating topic. You may have felt as if the whole world has been changing over the last couple of years where it's getting harder to drive traffic to your website. You're not maybe showing up in search. Maybe you're not showing up in those little AI results that are showing up at the top of Google. Or maybe ChatGPT is not even knowing who you are and you're kind of wondering, do I need to rethink how I market? Is there some fundamental shifts that I need to do? Well, today's guest is going to help us unravel it. Today I'll be joined by Rand Fishkin and he's going to help us unravel these very challenges. We're going to explore a concept called zero click everything. And we're also going to go deep into AI and how Is it going to impact, is it impacting jobs? How is AI and search impacting and changing things and how we think? How ought we behave differently? This is going to be one of those must listens because if you don't know who Rand is, he's a really big deal. And I think you're going to find today's interview extremely fascinating. Let's transition over to today's interview with.
- [03:01]
B
Rand Fishkin, helping you to simplify your social safari. Here is this week's expert guide.
- [03:10]
A
Today, I'm very excited to be joined by Rand Fishkin. If you don't know who Rand is, he is CEO and co founder of SparkToro, a software company that helps marketers understand the behavior and demographics of their audiences. Rand is also the co founder of Moz, a leading SEO company and also the author of Lost and founder Ran. Welcome to the show for the very first time. How you doing today?
- [03:31]
B
Great to be here, Michael.
- [03:33]
A
Yeah, it's awesome to have you, man. I should have done this a long time ago. Today, Rand and I are going to explore one of the biggest threats literally happening today with markers, the zero click challenge. We're also going to explore Rand's views on AI and search marketing and a whole lot more. So random. This is a concept that I believe you coined, this concept of zero click everything. Why don't we start by like, first of all unraveling what in the world this means and then we can just kind of go from there. What is zero click everything?
- [04:03]
B
So let's go back in time 10 years and think about what every digital marketer, social media marketer, search marketer was tasked with doing. Go drive traffic from all the platforms back to our website and the content that we created so that we can turn those people into subscribers and hopefully customers.
- [04:24]
A
Right?
- [04:24]
B
That worked for 20 years, right? The, the first 20 years of the Internet, that was what everyone did and it worked quite well. And then Google, followed by Facebook and Instagram and LinkedIn and Reddit and YouTube and TikTok and every other platform you can name, realized that it was in their interest to keep you on their site, not send traffic to other places. And so Google started answering search queries right in the results and LinkedIn started demoting the value or the reach and engagement that posts with links would get. And Twitter built in their algorithm a system to show fewer posts to people if they contained a link. And all the social media platforms and content platforms and search engines followed suit. And then AI hit and the AI tools, as you know, send out almost no traffic at all. And almost never reference the sources where they got their information, which of course means even less traffic opportunity. So even as these platforms have grown 10x100x in terms of their reach and engagement and the people you can reach through them, the amount of traffic that they can directly drive has dropped precipitously. It's not entirely gone. You can still do a little bit. I think this actually fools a lot of marketers into thinking it's their fault, that they're the ones who are supposed to be responsible for driving traffic. And, oh, why aren't we doing as well as we were last month, last year, when in fact what's happened is zero click everything.
- [05:56]
A
It's such a fascinating concept because I've seen it coming and you've seen it coming for years. Right. This idea that there was a river between the social platforms and our websites back in the day, that was extremely easy. That's why we all put these social share buttons all over our articles and our websites, because we wanted to get traffic to the website. And then the slow closing of the gates, right, they closed and then they locked and then they permanently locked. And that has resulted kind of in a fundamental shift, Right. A paradigm shift in how we market. Right. So like, let's talk about that paradigm shift, because there are still people operating off the old paradigm.
- [06:32]
B
Absolutely. I mean, if you want to get deep into it, there's a darkness there. Those social buttons that you mentioned that Facebook and Instagram and Reddit and everybody wanted you to put on your website, when you put those on your website, when you put their cookies on there, when you put their pixels on there, that gave them the detail and the data to monetize attention and influence, to know what everyone on the Internet was interested in so that their algorithms could deliver exactly that and keep them on their sites. So we were unwittingly the architects of our own demise.
- [07:05]
A
By the way, just for the record, I knew this years ago and I decided to have a pixel free website. Okay. I would not let Facebook fire their pixels on my website years ago because it, it became very clear to me it was an unfair advantage. I was effectively giving them data on my customers and everyone else for free. Now I, I agreed to allow Google to do it because I needed Google Analytics because there was utility there.
- [07:29]
B
Yeah.
- [07:29]
A
But the reality is, if we're fundamentally honest with each other, we were giving them valuable data that they monetized through advertising. Right. And they never paid us a dime. I mean, that's a bit of a challenge. Isn't it?
- [07:41]
B
I mean, just, just think of the AI tools too, right? They basically crawled all of our content secretly under the radar without our knowledge, often using bot names that couldn't be traced back to them and third party services, and then built these giant indices so that they could train their learning models right off of all of our content anyway, blah, blah, blah.
- [08:00]
A
Oh, but it's an important blah, blah, blah. Because people need to be aware.
- [08:03]
B
Yeah, yeah.
- [08:04]
A
That pretty much as long as we've been in digital marketing, there has always been an unfair advantage. The platforms. And we've loved the platforms when there seemed to be an exchange. Right. When we got traffic from them.
- [08:17]
B
Exactly.
- [08:17]
A
But now this fundamental paradigm shift. Talk to me about this, because it does require us, to many of us to really rethink what it means to be a digital marketer. Right?
- [08:27]
B
Yeah. For this, I like to go back in time. I like to go back to 1965. Right.
- [08:32]
A
The.
- [08:32]
B
The madman era. And if you imagine Don Draper and his cronies or, you know, whatever, Ogilvy and Mather, the big advertising companies, like, they would meet with you. Let's say you're the Coca Cola Corporation, right. They'd meet with you and they'd say, hey, Michael, in order to get the best results out of Coca Cola's ad dollars, we are going to run a series of two dozen different billboards in cities across the United States. We're going to pick cities and regions in those cities that have similar demographics and psychographics, and we're going to put up a series of billboards that we think are going to be most effective. And then we're going to learn by watching same store sales in a 5 mile or 10 mile or whatever, the driving distance, the average driving distance in those cities is watching that average. We're going to see whether Billboard A or Billboard B had better results in the Midwest versus the west coast versus the South. And then we're going to take the winning billboard and we're going to roll those out across those regions and see whether there should be nationwide ones to all that kind of thing. They do the same thing with television advertising. They use panels like the Nielsen, Right. TV families to go figure out, hey, if we run this ad during, I don't know, Mork and Mindy or MASH or Gilligan's island, you know, whatever it is, do we see a lift that is greater than what our model would predict for financial performance? Right. In terms of same store sales year over year. And then they'd see which ads worked and on which networks and what market share those networks had, how many people watched them, whether those people were their demographics, all that kind of stuff. That model of advertising and marketing requires implied conversion and implied attribution rather than direct attribution. When digital marketing came along, Google and Facebook especially said, whoa, whoa, whoa, whoa, whoa. Why are you throwing your money away at ads who you can't perfectly measure every customer who came from them? You should instead spend money with us, right, with the big tech platforms. Because what we can tell you is that this sale visited these three pages on your site and these 10 other websites and saw your ad in these places and watched these videos, and then they made a conversion. And you should attribute 10% to whatever your YouTube pre roll ad and 5% to, you know, your Google AdWords ad and et cetera, et cetera. And they built these attribution models and they convinced, especially CFOs. CFOs absolutely love this attribution data and they got addicted to it. And what they didn't realize was again, you know, Trojan horse, secret style. What Google and Facebook were doing in a lot of cases was taking credit for sales that already would have happened, right? What does Google and Facebook know better than anything thanks to those pixels that we all put on our websites, right? And all the engagement we give them, they know where you're going to go before you buy. And as a result, they can make sure that whether you saw the ad or not, whether you consciously processed it, whether it was off on the side, while, you know, your browser is too small, it doesn't matter. They can say, hey, when Rand was browsing on the web before he bought his tuxedo shoes for a wedding this fall, he saw this ad and so we're going to take credit for it. And then all of the other things, right, that went into that purchase decision that I made, they don't get attribution, right? The AI tool where I went and, you know, asked them about brands for tuxedo shoes, the YouTube videos that I watched organically, the subreddits that I read, which I did read a lot of subreddits, right? The browsing that I did on Blue sky, where I follow the. Who's this, this guy who does great men's fashion information, none of that gets counted. But Google and Facebook know they can take credit for showing that ad. And so, you know, a lot of people who turn off their ads find that 95% of the same sales still come through, and they're like, wait, what were we paying For. You're telling me we're only paying for that 5% of Lyft, but they were taking credit for all of these. That is the world that we're living in. This world where the zero click everything also breaks. Attribution modeling. Right. And it also breaks how we have to redo our digital marketing strategy, our marketing strategy overall. We gotta shift back in time to this test measure and use implied metrics rather than attribution.
- [13:03]
A
When we were prepping, you mentioned something about a San Francisco tour and some comedy stuff maybe to help people wrap their head around how someone might do something like this.
- [13:12]
B
Now, look, I think the San Francisco thing's quite interesting. So, Michael, you've been to San Francisco in the last few years, I assume?
- [13:18]
A
Yeah, yeah.
- [13:19]
B
So if you, you know, if you walk around the city, you will see an insane amount of outdoor advertising that is unlike any other American city I've ever visited or any European city. You know, most cities, when you look at outdoor advertising, when I say outdoor, what I mean is not just billboards, but also posters on sides of buildings.
- [13:38]
A
And stuff like that, too.
- [13:39]
B
Yeah, stuff that people paint on the sides of buildings, posters that they put up on telephone poles, polls, like anything that's called out of home advertising, bus stops, all that kind of stuff. Even the sides of buses. Those ads in every other city tend to be for products that almost everyone would be interested in. Health insurance, doctor's offices, toothpaste, television programs, movies, a lot of those. Like normal consumer things, because it's going to reach a huge swath of people. So you should try and only go for things that a lot of people might buy. But in San Francisco, it's not like that. It's a bunch of weird ads for companies you've almost certainly never heard of. Even if you're deep in tech like I am, I didn't recognize 80% of the ads that I saw. They were for AI companies, startups, you know, tech businesses that many of whom are not public companies, have, you know, fewer than $10 million of revenue. And you have to think to yourself, like, what? What is going on here? Why are these companies spending so much money on outdoor advertising, which is a very expensive form of advertising, and why are they spending it in a place where they know they can't measure return? And the answer is, because they're trying to influence a very small number of people who live in or visit San Francisco in the general Bay Area, including venture capitalists who might invest, private investors who might put money in pe, fund managers, directors of Engineering and IT who of course visit the Bay Area or often live there. And these small number of people, maybe they make up, you know, 5% of San Francisco Bay Area's population at any given time. That's who they're trying to influence. And they believe that it's more effective to buy outdoor advertising. That is a waste for, you know, 95% of people who see it just to reach those folks. And they don't care about the attribution, which is actually pretty smart. Once again, the tech companies, these tech startups are sort of leading the way in terms of what smart spend. Throwing 50 grand at the side of a bus might seem insane compared to putting 50 grand into Google, you know, AdWords or Google Ads, but it actually might be more effective.
- [16:00]
A
Yeah. And it's funny because as I think about it, there's probably a lot less competition too because when you're, when you're on the socials or in search or on YouTube, you're seeing ads everywhere you go. But when you're actually in, ideally, if you know that there's a building next door where there happens to be a bunch of influential people working and you can put a sign outside that building, man, that's going to be smart marketing. Right, so how do we need to be thinking now then? Because not everybody that's listening obviously is going to put physical ads in front of buildings. But I'm just curious what your thoughts are because so many people obviously are selling online products to online audiences. Any thoughts on, like, how we should fundamentally approach this in a world where clicks are effectively heading to zero?
- [16:43]
B
Look, the, the answer, I'm using the billboards as an example or outdoor as an example, right? But the reality is that in a zero click everything world, the only marketing that's really going to be effective is zero click marketing. This is marketing that, right? That does not require a click to have influence. And if you think about how you're influenced in your daily life, a massive amount of it is not through. I saw something and then I clicked on it and that's how I expressed my interest or remembered it. A ton of the marketing and branding and memory and, you know, positive associations that you build are someone emails you, you have a nice email with them, you see something in your Reddit feed or as you're scrolling through Hacker News or through TikTok or through LinkedIn and you're like, oh yeah, you know, SparkToro, like there's Rand's company. Like, oh, that, that sounds interesting. Maybe you're listening to a Podcast, there's no clicking here, right? You're listening to my voice or you're watching me on YouTube or whatever. There's nothing to click. Maybe you'll go and type in, you know, my company's website or whatever. But generally speaking, the influence is happening, even though there's no attribution. And I think that's a smart thing. This is how I recommend people invest in marketing. You have to find the sources of influence that reach your audience. That's two things. It's both the platforms. You know, Reddit is popular with this group. LinkedIn's popular, that group. YouTube's popular with that group. And it's the specific publications or influencers or sources of influence that are on those platforms. A lot of people Listen to Michael's YouTube channel, right? Subscribe to his podcast, get his newsletter. A lot of people are following this. I mentioned Dverk there, right? Which is like this fashion influencer guy on. I think he's on threads and blue sky mostly. He's got quarter of a million followers on those two platforms. He posts a bunch of stuff about, for example, like what. What people in the Trump administration wear and comparing that fashion to, like, the Reagan administration and then telling you what tuxedo shoes to buy. And. Right. Like all of this detail and information that requires no click whatsoever but is clearly influential. And so my advice for marketers is go figure out the sources of influence that your audience has and then find ways to be present in those places with a message that resonates.
- [19:10]
A
You know, it's funny, as I think about this example of going back to television, before there was the Internet, right? It was the top shows like the Seinfelds and the Friends and depending on what area you're in, super bowl, obviously. And all these shows, right. Like, people that were in marketing understood that if they could be somehow affiliated in some way with something that has. Attracts a lot of people or someone, or some show like this is where athlete endorsements and all these kind of things historically have worked out, out. Right. Or actor endorsements. This is something that's been around forever. And the idea that maybe we need to go back to fundamentals. We need to be thinking about, all right, maybe we need to spend money. Because I think marketers have been kind of programmed in their mind that everything is free now in the digital world, right. And we're going back to an era where actually you might have to spend some money. You want to talk about that a little bit?
- [20:01]
B
Well, I mean, I want to say two things. You Might have to spend some money, but a lot of influence, as long as you don't require attribution, is free. Earning attention and awareness through an email newsletter or a podcast or a YouTube channel, or submissions to Reddit or a great social feed. Those are free, right? Even a blog, right? Like, not entirely free, but close to it.
- [20:26]
A
Yeah. Word of mouth marketing, getting your customers to talk about your stuff. Like Apple. People love Apple and they talk about Apple all day long, right?
- [20:32]
B
Well, and until they don't. So this is, I think, one of the toughest things. Marketing requires a product that people want to talk about. Do you remember this old. I don't know how, you know, into like the Y Combinator scene. You used to be Michael, like back in the old days, but early on in Y Combinator history, Y Combinator is like a investment fund that's very popular in Silicon Valley. And they, you know, launched a ton of companies, including places like Reddit. And they're sort of worshipped in the tech space. And back in the day, I used to worship them too, because my first company was raised VC and was sort of in that universe. And there was this essay by Paul Graham, their founder. Actually, I can't remember the name, but I think the message from it was, make something people want. Like, that was the big takeaway. And for years, I think for more than a decade now, like tech founders, software founders all over the world have said, oh, well, they failed because they didn't make something people wanted, or they succeeded because they made something people wanted. And it's this. It's this binary. It sounds dumb, like, of course you have to make a product people want it. But what I would say is that advice isn't wrong. It's not bad. It's just incomplete. Especially today, it's incomplete. The real challenge is not to make something people want, is to make something people want to talk about, to share. If you can make a product like Apple does, right? Or like, I'm obsessed with Madeline Miller's books, for example, right. I think she's just a tremendous writer. And I want to talk about that, right? It's the trombone I want to play. When I walk into a cocktail party, I'm like, oh, my God, have you read Circe? It's amazing kind of thing. And making a product that people want to talk about is a superpower because it amplifies everything that your marketing does, right? Instead of, hey, we got 10 customers. One of them might help share what we do. It's, oh, we got 10 customers. Four of them might share. If six of them might share, you might need to do no marketing at all. You can just sit on your butt and wait for your product to take off with every human being on Earth. That's what virality means, right? It's a virus. It's spread. It's like Covid, right? It's just, oh, no, we can't contain it. Suddenly everyone is talking about, you know, whatever, the Barbie movie a couple years ago that's making a product people want to talk about. And unfortunately, one of the problems is that product folks and executives and, you know, C suites, they like, get together in their, you know, boardrooms. They make something, then they hire a marketer or an agency, and they're like, okay, now you sell it. Come on, man. They're kind of making you do your job with one hand tied behind your back. If you had the power to sit in the room where product is being designed and say, whoa, whoa, whoa, guys, what if we made this a little more fluid? What if we ease the processing fluency of this brand? What if we made it something that contained a natural viral loop, right, where it automatically attracts more people? Because by bringing more people into it, you get some benefit from that. Your job gets way easier. And sadly, marketers are almost never let into the product room. So my first advice, if you can possibly be in the room where the product's getting built and designed, get in there. If you can influence that, if you're a founder and you can influence that, bring marketing into your product discussions. And then the second thing I would say is when you're building these zero click marketing plans, you have to know your audience deeply, and you cannot just focus on your customers. This is one of the other. Like, Michael, you know this better than anybody, right? But there's like, if we were to draw the Venn diagram, right, There'd be like the group of people who's in your audience, people who are listening to this podcast, for example, right? Maybe one in ten of them would have any interest in what SparkToro does, which might be great for me, but I need to speak to all of them. Like, my job is to be interesting and useful and valuable to all of them, not to just focus on the 10% or 1% or whatever it is that cares about my product. And too many marketers are told by their executive teams or by their, you know, their clients or whatever that they're just supposed to market to customers. And that's a terrible idea, because if you market to your broad audience, you will have far greater reach. Amplification, brand recollection. There's probably a hundred thousand products and brands that you can name if, if you could think of them and you'd be like, oh yeah, I've heard of that. I know that because they spoke to their audiences. Even though you've probably only bought maybe a thousand brands, products, very powerful stuff.
- [25:15]
A
Is your marketing falling behind? In my 15 plus years of tracking marketing trends, I've learned to spot the signals early. And right now, the pace of change in marketing is rapidly accelerating. What used to take years is now happening in months. AI capabilities that seemed impossible last year are already outdated. And I'm concerned that marketers who aren't prepared for what's coming next are going to be in a bit of a bind. The tactics that you're using today, the strategies that you spent years mastering, even the way you think about your audience, it's all evolving a lot faster than everyone realizes. This is exactly why social media marketing world exists. When you attend our 2026 event, you'll be positioned to lead AI strategy conversations in your company and for your clients. You'll be able to master tools and tactics before they become mainstream. And you'll also build authentic connections with future ready marketers that will have a big downstream impact on your business. Don't wait until everyone else figures this out. Join the marketers who want to get ahead and stay ahead. We have a really big sale going on right now. Visit social mediamarketingworld.info and grab your tickets today. Let's navigate this change together. We've been talking about this zero click challenge and one of the things that you mentioned was AI having really low click through rates and often not attributing information. You at SparkToro have access to lots of insights and information and research reports and all that kind of fun stuff. And there's been a lot of concerns in the marketing world that AI is taking jobs or going to take jobs, and a lot of it is because it's happening already at the development side. A lot of developers are getting laid off and marketers kind of feel like they're coming next, even though maybe they don't yet have that experience. But what's your thoughts? Do you believe that AI will replace marketing jobs?
- [27:18]
B
Nope, I don't think that's going to happen at all.
- [27:21]
A
And did you always have that sense or no?
- [27:24]
B
No, no. Like everyone else, I've been scared of it for years, worried about it for years, and sort of watching the data for a long time. And it is, it is the Data that has convinced me that that's not the case. I want to be clear. Oftentimes when I say macro trends are X or like, you know, whatever, AI is not taking developer jobs or won't take marketing jobs, I don't mean your boss or your CEO might not decide, hey, I want to fire my marketing team and replace them with AI and see how that goes. What I mean is in the US economy as a whole, if you look across the percent of people who are employed as marketers and you know that space or developers, which seems to be everyone's go to example with AI, you will find that if you look at the sort of post Covid, the bump and slump that followed Covid's hiring patterns, you cannot look at that data and statistically find evidence for rising adoption of AI leading to lower employment for these types of professionals. You can find examples, you can find anecdotes, but you can't find macro data. The macro data, everyone who's analyzed it. There was another big article this morning where someone had broken down like a big economic analysis and couldn't find any evidence that developers were getting hired at a lower rate.
- [28:49]
A
So is it possible that they were just over hired during COVID Is that kind of what you're implying?
- [28:53]
B
I think that's a truth, right?
- [28:54]
A
Or what the data could be implying. Yeah, exactly.
- [28:56]
B
Yeah, yeah, everyone agrees that happened.
- [28:59]
A
So what does the data tell you about the impact of AI on marketing jobs and what's your thoughts on it? So we already know that you don't believe it's actually going to, at a macro level, take jobs. But what kind of insights do you have that maybe you can share with us specifically when it comes to marketing.
- [29:15]
B
And AI, I think one thing that is absolutely true is that it will be much more difficult to get a job and keep a job if you are not someone who is relatively familiar with and skilled in the use of and understanding of AI. That seems entirely true. But you know what, that's not very different from like 1991 when every job position said, hey, do you know Microsoft Word and Excel and PowerPoint? You had to know those things and 10 years before that you didn't have to know them at all because they barely existed. I don't even think PowerPoint existed at all. And now suddenly every single job, you know, white collar job in the US required knowledge of those things. AI is pretty similar to that. Do you understand how it works? Are you familiar with like the tokens? Do you know how, what a large language model is and you know, how they build them and what they're based off of and how you can influence them, especially for marketers. Right. How to influence them is really important. Or if you're in the. On the technology side, like, do you know how to use them to debug code and sort of be more productive and whatever, get your variables in order or write documentation using them? You're going to need to know those things.
- [30:27]
A
Yeah. It's funny because I have a new study that's going to be coming out sometime this fall. I haven't really figured out the date yet, but. And it affirms what you are saying. We've surveyed, I don't know, 900 marketers and asked them a million questions, questions. And one of the questions was, do you see it as a threat? And we asked on a Likert scale. Right. AI is a threat to my marketing. You strongly agree? Strongly disagree. And the whole thing in between. And most marketers that are getting educated on how AI works see it less as a threat and they're seeing it more as a opportunity. They see it as a, finally, an enabler for them to basically do their career better. Right. To write more persuasive copy, to perhaps do data analysis more quickly, to automate boring, repetitive tasks so they can make time for creative or analytical tasks. Are you finding this to be true from everything you're seeing?
- [31:18]
B
Yeah. That certainly comports with everything that I'm seeing as well. I think that AI is a software tool. People don't like it when I make this analogy for some reason, Michael. But I think it's a lot like Excel, you know, if you go back to 1975, right, and there's just thousands and thousands of, like, people in financial fields and engineering fields and a million other fields who are sitting around drafting tables with slide rules and, like, you know, trying to do all this manual hand work. And then you fast forward 10 years later and every single one of them is on a computer and using a spreadsheet program. You know, whether it was Lotus or Excel or whatever it was, Excel underpins, like, the entire financial industry of the world, right? Like the global economy all runs on Excel, you know, but for some reason, the media hype around it wasn't. It's going to replace jobs and like, oh, my God, people who don't know Excel, it just sort of happened, right? It was just. It felt natural over that decade. And I think the decade from, you know, 2021 to 2031, for whatever reason, like, you know what it is, even though I Say, zero click, everything. I think the media found that their clicks were declining until they said AI is going to take jobs. And then suddenly, like your aunt in Missouri. Oh my God, Rand, did you see that? AI will probably take your job.
- [32:37]
A
Yeah, yeah. You know, it's interesting because one thing I think we would agree on is that AI has changed consumer behavior. The way people gather insights and information has, I think, fundamentally shifted. I know that a lot of people are kind of going to chat GPT as their first way of getting insights instead of Google search. And I can't not talk about this. Because of your background in search marketing, in your professional opinion, what has AI's impact had on search from your lens?
- [33:08]
B
I try not to inject my opinion when I have good data. So instead I'm going to share with you good data rather than like my personal opinion.
- [33:15]
A
Yeah, I will add my opinion on your data. How does that sound?
- [33:17]
B
Yes, that would be perfect. Okay. Okay. If anyone wants. We published all this stuff on the SparkToro blog. And I work with a company called Datos, which is our, our clickstream data provider. We were talking about the Nielsen TV families earlier. Michael Dados is. And all clickstream providers are essentially like that, but for browsers and like what people visit on the Internet. Right. So they, they essentially have millions of people who opt into a panel and then they, they say, you know, what do all those people visit in, in what order? And you can get lots of data out of that. So, for example, they looked at cohorts in the clickstream panel of people who started using AI tools like ChatGPT or Gemini or Claude in January, February, March of this year, and then what happened over the next three to five months. And what they found was surprising to me, actually. I did not expect this. I thought it would go the other way, that once someone started using AI tools, they saw a quick spike in the number of Google or other traditional searches that they ran. And even three to five months later, they were searching more than they had been in the past two years. So AI tool use suggests that you will use traditional search more than you did before, not less.
- [34:35]
A
Let's discern what that could be. Is it possible? It could be affirming whether it's true or not, by doing a Google search as a backup data source. What's your thoughts? Is that possible?
- [34:44]
B
It's totally possible. Also, AI gives you a bunch of, you know, brand like, hey, what tuxedo shoes should I buy? And they're like, oh, you should check out. Yeah, you know, whatever it is xyz. So I go to Google and I search for xyz.
- [34:55]
A
I see, right?
- [34:56]
B
Like, because Chachi BT doesn't have a shopping link, you know.
- [35:00]
A
So just so we're clear, the data that you've seen shows that presumably AI usage has increased, but so has Google search. Is that what I'm hearing you say?
- [35:11]
B
That's exactly right. Google had their biggest search growth year in the last six years. Last year, when AI was taking off at the. At the highest velocity it's been at, it's now. The growth is now slowing. Right. Which you can see with all the people worrying about the AI apocalypse. And while it was growing super fast, Google was also growing. I think Google grew searches something greater than 20%. And you could see it in Google's earnings results. Right? Like the market was like, oh, damn, Google's, you know, taken off. And what I want to be clear of, because anytime I bring this up, somebody jumps in the comments and says, I barely use Google anymore at all. I use ChatGPT for everything. Rand is wrong, friend. Like, look, broad data is not your experience. I'm not saying you don't do that thing. I'm saying if we take 10 million people in the United States and we look at every URL that their browser visits, this is what we find. Right? Like, at the broad scale, it's just like saying 65% of grade school teachers are women and then some guy saying, I'm a man and I'm a grade school teacher. No one's saying you're not, dude. That's not what statistics is.
- [36:21]
A
I think that a lot of marketers are thinking what I'm about to say next as to why this might be true. And I know you kind of. I think you know where I'm going with this. Even though it's true that people are using Google search more, isn't it also true they're not scrolling past the AI thing at the top? The summary?
- [36:41]
B
Yes. Thank you.
- [36:42]
A
And that's probably why they're using Search more, because it's just as convenient as ChatGPT. Right?
- [36:47]
B
Remember what we started the show with? Zero click everything. Both things are true. Google can be searched way more than it was two or three years ago. And there can be way less traffic available because Google has reduced the amount of traffic it sends out to a significant degree. The amount of traffic that is available is lower. And this is why zero click marketing matters so much. Because if someone searches for Google and they only see the AI overview or the instant answer, or, you know, Whatever. Even sometimes Google will give you so much information in the little description below a website that you don't have to click on anything. They've already told you the answer you were looking for. And so you know what, your job, if you're a marketer, is to influence that page. What Google is showing people is what matters. That's your new homepage. That's your new point of conversion and point of sale. That's where you have to do marketing rather than how do I get a click from Google over to my website and then I'll do my marketing.
- [37:49]
A
Okay, before we get into this, because I definitely want to get into this AI search optimization or whatever the heck we're calling it. This is really important, I think, for people to grasp Google. Just as a reminder, everybody invented the GPT, okay? They were the ones that wrote the white paper that ultimately OpenAI ended up using to develop ChatGPT. Google's the one that has some of the most advanced stuff out there that's just finally catching up, right? And Google also has distribution and data centers and more money and they're publicly funded and all this kind of stuff. So Google, we might be saying, remember when there was chat GPT, just like we remember, what are those search engines that used to exist that wouldn't even talk about anymore? You know the ones.
- [38:28]
B
Oh yeah, Lyco, Salta Vista.
- [38:30]
A
Yeah, yeah, we might, yeah, like AltaVista, we might say, remember ChatGPT? Remember those days? It's like saying, remember Clubhouse or whatever? You know what I mean?
- [38:38]
B
Yeah, yeah, Dig. I always think of like how every techie in the US was like, oh, I used to go to Dig everybody day.
- [38:44]
A
Right? But here's where I'm going with this. It's very possible that Google could just flip a switch someday and say, hey, we're going to show you the AI overviews only. And if you want the traditional one, you have to go change your settings. If they do that, they're exactly the same as ChatGPT at that moment, if not possibly even more powerful.
- [39:02]
B
Yeah.
- [39:02]
A
And I'm just curious if you've seen any data that shows that search traffic from Google is in decline. Do you have any data on that?
- [39:10]
B
Yeah, yeah, absolutely. So we, we calculate that percent of zero click searches, which is defined as essentially somebody not leaving Google's ecosystem. So for example, you might perform a search and then be like, okay, great, I got the answer to that. Now I'm going to perform another search, or I go to Google Maps or I go to YouTube and you're essentially not leaving Google's ecosystem. And then some people just close their browser entirely. All of that that stay inside Google's ecosystem, never leave. That's called zero click search. It's been rising from under 50% to almost. I think the last data we had we published was 65%.
- [39:46]
A
So 65% of people do not click. Is that what that means?
- [39:49]
B
Two thirds of people who search Google never leave Google's ecosystem.
- [39:53]
A
And I would imagine with ChatGPT it's more like 95% or something, right? Do you have any sense of that?
- [39:57]
B
I believe it's 99.9 8.2% of people who perform prompts, not searches. Right. Because we don't know exactly what, what a search equivalent is in, in ChatGPT. But perform prompts, leave ChatGPT's ecosystem directly from their browser. They might open a new browser window and search Google. And we, the clickstream data would consider that a different session. So we can't say for certain whether that's conflicting.
- [40:24]
A
But this is where it gets really interesting. We might say, do you remember a day when you had to go to a website to trigger AI? Do you remember when you had to go to google do or chatgpt.com to use AI? Because we're already starting to see right click AI integrated into almost the entire Google ecosystem. I would imagine the Microsoft ecosystem. We're already seeing AI audio sources where you don't even type anymore. Do you think we're getting to an era now where the AI will be so ubiquitous that we'll be able to call on it through all mediums and we won't even worry about typing in a URL. What's your thoughts on that?
- [41:05]
B
Here's what I don't like doing. I don't like saying I think this will.
- [41:08]
A
Is it possible?
- [41:10]
B
It's absolutely possible. Of course it's possible. It happens all the time already, right? Like every time Windows restarts on my computer, it's like, hey, can I shove some more AI into all your products? Sometimes it's useful, Sometimes I'm like, stop, Just stop. Slow your roll, bro. Whether it's possible or not, the question for me is more around what are people doing? I believe if you're a marketer, your job is not to predict the future. It is to be present in the places people pay attention with a message that resonates. That's your job. And so I think an inordinate amount of, and I mean no offense by this, like, if you're a marketing prognosticator, go for it. And you have fun with that. But I don't. I find value in, like, doing things that are happening now, rather than trying to say, well, in five years it will be this way, so I should do that.
- [41:57]
A
Okay, I love it.
- [41:58]
B
Everybody who's like a futurist, they've all been wrong.
- [42:02]
A
I like it. Okay, so now let's get back to where we were about to go, which is how can we influence the results as marketers that come from the AI systems, whether they be Gemini slash Google, whether they be Anthropic slash Claude, whether they be OpenAI/chatgpt, grok, whatever the heck they are. Because we are moving towards an era where there's a single best answer. That's what it seems to be, right? Like. Like these systems are providing what they think is the answer, right? Am I wrong?
- [42:37]
B
So when we did a large analysis and what we find is that when we analyzed AI prompt responses, the responses are usually, are very rarely one answer. Okay, so GROK is the only one where they try to give sort of one answer. And I think many people would agree that has its problematic elements.
- [42:55]
A
Okay, well, forget about the best answer. But there's a limited number of options that are presented to a user in an answer. You want to be amongst that, right? So how do we influence that?
- [43:05]
B
That's exactly right. A number of options are presented. AI is sort of trained to give multiple responses, and that's because the training data, right, the stuff that is trained on, often gives multiple responses. And you absolutely want to be in there, right? If somebody says, hey, what's the best software for. For abc? And your company provides software for abc, like, oh, man, you. You really want to be in that list. And it would suck if you weren't. And the way to influence that. There's. So there's. There's two things to consider here. People who are in search marketing generally recognize that there's rag, which is retrieval augmented generation that is mostly or. Or largely served by what traditional search engines do. So even I don't know if you saw the report right, that ChatGPT is essentially scraping the crap out of Google for a ton of its responses. And they use is it serper.dev like this company in Texas to do it. And so a lot of, like, people have said, you know, the, the Scooby Doo mask meme where you pull off OpenAI and you're like, hey, it's just Google underneath for retrieval augmented generation. That is what's going on. And then there's the Standard AI types of responses, which are powered by essentially the language model underneath and the training data. And that training data is essentially giant corpuses of text that come from books, they come from TV show transcripts, they come from YouTube transcripts, they come from probably every word on the Internet or close to it. And then they have weightings, right? And so they'll say, oh, well, Wikipedia is more important than, than rand and Michael's HouseOfText.info, which, by the way, I registered right before our chat, so look forward to joining you on that. The problem, the problem is that these systems change so rapidly, like, so fast that the advice from six months ago is out of date in a lot of ways. I saw a bunch of experiments, literally six months ago, January, February, time frame, where search marketers were showing me some very cool things, where they would essentially submit a single thread to Reddit, have a couple of accounts comment on it, get less than 10 upvotes on that post, and within 30 days be one of the answers inside ChatGPT's responses, if you framed your question correctly. So Reddit is obviously very important to OpenAI. That looks like it's diminishing. It looks like they're, oh, no, we were overly reliant on Reddit. People figured it out. Like, how do we do more of these things? I know this is overly simplistic, but I want to be clear that the only thing that seems to consistently work, that's worked sort of since the beginning of AI and the beginning of search engines and still works well today is be present in all the places your audience pays attention. If someone is talking about, I don't know, like, orthodontical surgery software, and your company makes orthodontical surgery software. And, you know, that's the thing that people are going to ask ChatGPT about, whatever phrasing they choose, or they're going to type it into Google, or they're going to go to Perplexity, or they're going to use it in their agent that is powered by Gemini. If you know that, and that's your world. When publications of all kinds talk about orthodontic, you know, surgery software, you want to make sure your brand is on the page, preferably close to that word. That's not rocket science, it's just pr, right? Like, it's, it's just good pr. And I will say I think I have found the easiest way to influence PR at scale without a huge amount of cost and lots of pitching is social media, because journalists and writers and authors and people who run publications big and small of all kinds. I'm not just talking about, like, New York Times. I'm talking about, about, like, Orthodontists Monthly, you know, and like, the editor for that is on Blue sky. And, you know, like, she's following a few people in the industry. If you get, you know, one of your pieces in there right in front of her eyeballs, she'll be like, oh, yeah, click. And then she includes your name in the next publication. And then that feeds into the AI systems and it feeds into Google Search.
- [47:24]
A
You know what I love about this? For those of us that have been around for a little while and have a little bit of gray hair, this is kind of encouraging because we are going back to very basic fundamentals of PR advertising. Understanding, influence, understanding how to get exposure, brand recognition. These are things that the pendulum is shifting back to, really things that we have long since forgotten about. Do you agree?
- [47:48]
B
Yeah. I mean, look, just look at the attribution versus measurement conversation. Right. We're going back in time and pr. PR is something that I think digital marketers, with the exception of, like, link builders, largely dismissed. Right. And they didn't play that game. What I will say is the devil's in the details. Right? Like, oh, Rand says just be present in all the places people pay attention. Yeah. But, you know, it's. It's actually pretty hard to figure out what publications are influential in your field and figure out which ones might be used by AI.
- [48:17]
A
Well, do you know any companies that help with that kind of stuff? Random. We're getting near the end where it's totally acceptable for you to talk about this, like, tell everybody what you do because they may not understand. Yeah.
- [48:29]
B
Michael, just throwing out the largest softball you've ever seen and gave me this giant aluminum bat. Yeah. So, I mean, what SparkToro does is with the help of clickstream data from Dados and a bunch of other data sources, including a lot from LinkedIn and Google. We basically say, you describe your audience to us, whoever they are. You could say, hey, they're orthodontists in the Midwest, and they tend to run small offices with less than 10 employees, and they see five patients a day on average. And they try to stay up to date on their certifications, and we're trying to sell them software. Okay, great. You've described your audience, and then what SparkToro will do is essentially use LLM type stuff to produce a group of profiles that match your description and then analyze their behavior. So, like, people who match that description? What websites do they visit? What YouTube channels they subscribe to? What podcasts are they listening to? What searches are they typing into Google? What topics are they paying attention to in their, like, Google News feed? What subreddits do they subscribe to? Like, all those types of things? And which AI tools are they using, more or less than average? Which social networks are they on? Like, all that kind of stuff. A lot of the time when you pull up a SparkToro report, you will recognize a few things in there. You'll be like, oh, yeah, that makes sense. Like, that looks like my crowd. Of course they. I don't actually know the world of orthodontics, you know, but you're like, Orthodontics Monthly, great. I don't know if that's a real publication, but let's pretend it is. You might see that in there and go, well, of course they're in there. But then you'll see 10 other ones where you go, ooh, right. I would have forgotten about that. I should pitch that place. Or I know somebody there. I should try and get into that publication and let me look at the topics they care about. Oh, that topic, Actually, I have something really good on that topic. I should pitch them on that. And you know what? I bet I can do a podcast about that with this guy over here. He's got a big YouTube channel too. Like, you figure out your strategy. SparkToro is just demographics and behaviors of any describable audience on the Internet and then what they do so that you can reach them. Because it seems insane to me, this is why Casey and I made sparktor like that. You know, a few years ago, if you wanted to do this, you would hire a market research firm for like a hundred thousand dollars or more and have them run surveys. And surveys are terrible. It's like surveying people about what groceries they buy. Instead of looking at the store receipts, like, just. Just look at the receipts. So that's Partoro.
- [50:56]
A
Sparktoro.com is where everybody can check it out. Rand, if people want to connect with you, I don't know if you're active on the socials. If not, what's your preferred? You kind of mentioned a couple already. I'm assuming you said Blue sky and Threads, but I don't know what's your preferred platform and where do they find you and if they. If there's any other websites you want to plug, because I know you got some other things you're working on as well, Feel free to do that.
- [51:14]
B
Oh, yeah, I love to plug stuff, but if you happen to love 1960s Italy based cooking and foraging video games, I run a video game studio called Snack Bar Studio and you can check out the game we're making. I'm also on the verge of I don't know when exactly this episode will come out, but on the verge of launching another product called Alert Mouse, which is basically just a good version of Google Alerts.
- [51:36]
A
Where will they find alertmouse?
- [51:38]
B
Alertmouse.com okay, cool. Yeah. And then Sparktoro is where I blog and write and publish a lot of research that we've talked about today. And you can of course try that tool for free there too. I am active on the socials LinkedIn, especially in blue sky and Threads, and rarely I don't reply on Twitter, but I'll sometimes publish my research there too.
- [51:57]
A
Brian Fishkin, thank you so much for unraveling some really fascinating information for us today and getting us to to think deeply about where the world is going. We're so grateful for it.
- [52:05]
B
Oh Michael, thank you for having me.
- [52:07]
A
Hey, if you missed anything, we took all the notes for you over@socialmediaexaminer.com 685rand has also told me offline that he's got some industry reports that he's going to give us that he referenced in the interview that we will link to in the show Notes Again, Social Media Examiner.com685 if you're new to this podcast, follow us. And if you've been a listener for a while, we would love you to share this out on the socials and feel free to link to me. I'm most active on Facebook, LinkedIn and X. And do check out our other shows, the AI Explored Podcast and the Social Media Marketing Talk Show. This brings us to the end of the Social Media Marketing Podcast. I'm your host Michael Stelzner. I'll be back with you next week. I hope you make the best out of your day and may your marketing keep evolving. Catch you soon.
- [52:54]
B
The Social Media Marketing Podcast is a production of Social Media Examiner.
- [53:02]
A
This is the year to finally come to social media marketing world 2,026. Grab your tickets right now by visiting social mediamarketingworld info.