Loading summary
Transcript69 lines
- [00:00]
Jordan Cooney
The Voices of Search Podcast is a proud member of the I Hear Everything Podcast network. Looking to launch or scale your podcast, I Hear Everything delivers podcast production, growth and monetization solutions that transform your words into profit. Ready to give your brand a voice? Then visit iheareverything.com welcome to the Voices of Search Podcast. A member of the I Hear Everything podcast network, ready to expedite your company's organic growth efforts. Sit back, relax, and get ready for your daily dose of search engine optimization wisdom. Here's today's host of the Voices of Search podcast, Jordan Cooney.
- [00:43]
ChatGPT visits have grown 90% year over year, from 3 billion to nearly 6 billion visits. This is a fundamental shift in how people find information, products and services. While most enterprise teams are optimizing for traditional blue links, hundreds of millions of discoveries are happening in AI responses. Your SEO strategies, built around specifically driving traffic from Google links, is becoming less valuable. While daily users never click through to your site. The question isn't if the shift to GEO will happen, it's whether your brand will take advantage. So how do you transition from SEO to GEO before your competitors leave you behind? Jordan I'm Jordan Cooney and joining me today is Josh Bliskell, head of AI strategy and research at Profound. Today, Josh and I will reveal why marketers need to jump on GEO strategies now and share practical frameworks for enterprise teams that need to make the transition successfully. Josh, welcome to the Voices of Search podcast.
- [01:46]
Josh Bliskell
It is great to be here. Thanks for having me. Yeah, I mean, it's incredible to be here because the space is moving so fast. It's just an incredible space to be. And for a little bit of context.
- [01:56]
Jordan Cooney
Yeah, please.
- [01:56]
Josh Bliskell
I head up AI strategy over Profound. Profound is a year and like three months old. We're the platform that enables marketers to create content and actually start to tangibly improve, understand, not only defend, but grow their visibility within AI search, through citations, through visibility, through bot accesses, indexes, clicks, human referrals, all that good stuff. So we are right on the forefront of geo. Excited to be here.
- [02:21]
Jordan Cooney
I mean, it's crazy because Profound is like a year and some change old, right? And just I think this past week ChatGPT became three years old. So like, you know, the toddlers are starting to crawl and it's. And it's amazing to see how fast the industry's changing and how much new information is becoming accessible so quickly around just AI discovery in general. Let's talk a little bit about Profound right off the bat. You know, what does Profound's billion citation research reveal about which brands are winning, which brands are losing in this AI discovery race that we're in right now.
- [03:00]
Josh Bliskell
I think at the core the data right now is showing us that it's about context. Answer engine optimization is about providing context to models in a way that the models can digest, understand and respond back to users with we are in a high utility content world. The content that wins is not necessarily the same content that traditionally ranks in SEO. There are different ways to start optimizing and there are ways that you can start building out frameworks and strategies to start thinking about how to appeal, how to offer these answer engines context, not only in terms of informational searches, but also as we start to see with commerce, with agentic browsing. So there's a lot of different areas here, but by and large, if you want like the one word, context is king and context is gonna dictate success in this space in the next year for sure.
- [03:49]
Jordan Cooney
So we've gone from this content is king to context is king. Right? And somewhere in there is probably the intent or the purpose or the drive behind what the user's looking for. Right? Because this response world, it's a response, right? You're getting something back from these LLMs. One of the things I'm curious as you guys look at your data and you look at your insights is when it comes to refreshing context, how important is it to be constantly invested in providing new, relevant, updated messaging as a brand to stay contextually relevant in these responses?
- [04:27]
Josh Bliskell
I think I'll simplify it to this way. I think the reason that answer engines have created such a bias towards the present that they have is because answer engines inherently, they have been hazed all the way through 2023, early 2024, with all of these great questions, who's the current president? You know, what's going on with this? What's going on with that? The answer engines themselves are trying to find the latest information. They're incentivized to find things from December 2025, from January 2026, looking at every source and trying to wait, is this going to be the absolute most updated, the absolute most factual piece of information that I can give to the user. So when I think about content creation itself and actually thinking about how we invest and optimize and refreshes and refactors, I'm thinking about one is this going to be a space where the answer engine is going to try to drive towards immediate frequency? So if I'm talking about corporate credit cards, for instance, right. I Want to know what the monthly new joining offer is or what the monthly bonus points are?
- [05:28]
Jordan Cooney
Sure.
- [05:28]
Josh Bliskell
That's going to be really important. The answer engine is going to drive to that and you can see that in the query fan outs. Right? The query fan out should dictate a little bit of that strategy as well. Right. If you can see ChatGPT is saying Best Corporate Credit Card December 2025. Well, there you go, you've got your answer. This is a monthly cadence kind of problem. If you're looking at something like gingerbread houses in Bolivia, where can I buy a gingerbread house kit in Bolivia? You can probably have one exceptionally high quality source that's probably not going to be usurped or overridden for some time if you're going exceptionally long tail. So there's also a bit of influence there in terms of how much content's being created in the space. Like if I'm working with someone in the running shoe space, even if running shoes aren't changing day to day, there is still kind of this influence that the mountains of ugc, the mountains of content there will have. So simplified two axes, right? It's going to be the freshness of how much like the literal freshness weight that ChatGPT places and then the actual amount of content and actual, you know, volume of information is coming out in this space. So really high. Like cell phones really need to be fresh on the tech specs, things like that. But also a lot of people are talking about it very cutthroat space.
- [06:39]
Jordan Cooney
So let's, before we get into the whole fan out query piece, because I think that that's a really important part of the discovery of what marketers need to know. Right now I do want to keep talking about thinking about patterns, data sets that we consistently are seeing and we're hearing about components specifically around like Wikipedia, Reddit and the frequency of seeing information from those and these kind of these additional sources that are building back to your original point around context. They're building context around a brand, their products, their solutions, the what, what they offer in the market. Tell me about how you guys are thinking about those components, these patterns that we're seeing around these sources of information.
- [07:25]
Josh Bliskell
It's a diverse information portfolio and what each of these different sources of information really represent besides being domains. To me, Wikipedia, Reddit, these really represent classes of information that when ChatGPT increases Wikipedia citations, it's not because they have some secret love affair with Wikipedia, it's that they're trying to index more heavily on pure factual Peer reviewed encyclopedic knowledge. When they index heavier or not on Reddit, they're testing how the influence of UGC impacts the quality, the factuality, the perceived value of their responses. And so when you think about Reddit, I don't think you should think and see the little Reddit icon. You should think and see kind of just the concept of UGC. What is UGC's role in these answers? So we know Reddit's sitting at around 2, 3% and actually kind of declining right now in search at this point. And we know that Wikipedia is sitting at around 7% of all ChatGPT citations. These classes of information are pretty durable things, even if they're fluctuating to and fro, because those elements are going to be essential in these responses throughout the next few years. I mean, if you're going to build a rag system and you're going to rely to build this trust with users and really hope to try to encapsulate the cultural zeitgeist of like, what do people think about these running shoes? You're going to need some of these pieces of information. Same thing with Wikipedia. Wikipedia offers a great deal of factuality about many of the world's largest companies. So it's almost kind of like this legitimizer. What are some of New York's biggest or best startups? ChatGPT is able to go and index directly on that, look through Wikipedia and find out what the legitimate sources and startups are. That's my perspective. At least think about these as classes of information.
- [09:07]
Jordan Cooney
Better yet, I mean, I know you're from Chicago. We should have the debate about pizza at some point. That's the real debate that's in Reddit right now, right? Is it New York pizza? Is it Chicago pizza? I mean, this Chicago thing really leaves a giant amount of food in my stomach every time I eat it. But we can talk about that later.
- [09:23]
Josh Bliskell
Absolutely.
- [09:24]
Jordan Cooney
So really quick though. A few months ago you put up a LinkedIn post about something that I think is really foundational and key to everyone's learning when it comes to understanding how these AI engines are really reshaping what users need in terms of the Internet or the web, right? And that is there's a huge percentage of AI prompts that are focused on generative activities, right? We're talking about like, write X, create this, do something for me, right? Versus this informational path of like, I'm seeking knowledge, I'm seeking a product, I'm trying to shop right now for the best Black Friday things and it's just too many listicles and I can't scroll anymore in the death scroll of social media. So tell me like, tell me why this is important. Understanding that this is a tool that has a different set of utility that goes beyond just search and discovery and that it is something that we use as utility in much of our daily actions.
- [10:26]
Josh Bliskell
Here's the deal. Search has moved from being, you know, the hallway. Search was the hallway, right? It has moved from being this hallway where you pass through it to get to these rooms to being the room itself. Search is where things happen. Search is the party. I don't need to go find the party. I'm in the party. I'm on chatgpt.com yeah. So you know, it has moved from being the means to the end to the end itself. And in doing that we've proven a few things. One, that the future everyone's trying AI. Health care wants AI. You know, everyone wants, the military wants AI. We know that in the search industry AI is an improvement to the ux. Users enjoy searching through conversations. So regardless of what's going to happen with ChatGPT, regardless of if perplexity is going to come out with an awesome new thing, conversational search is here to stay at the very minimum. Now what we think is really exciting about that is that these conversations go from being about discrete, like I am directly asking for information on how to build a marketing plan to now I want to create a marketing plan. How can I do that? Well, here's how to do it. Awesome. I got 15 new leads that came back next week. How do I track all that? Oh, you should try HubSpot, Zoho, Salesforce. Boom, boom, boom. These, these problem conversations. People start with kind of a problem that they have or a need that needs to be solved and then eventually through the process of needing to solve this problem. The thing that we find is much to brands, you know, jubilation, much to their happiness. 47% of these conversations actually feature unprompted product recommendations. People actually come in and say I need a new desk setup, I work from home, my back hurts and I am just done. Oh, have you considered, you know, Herman Miller? If you want to spend some Herman Miller makes great chairs. There you go. So it doesn't, I don't think it throws off. It's not like this anti commerce machine. A lot of people are like, does that mean that like people aren't shopping? No, it means that people's shopping journeys start with what's like really core to them. You know, this is my need, this is what needs to be solved. And then you have to think about again, context, contextualizing your brand, your product, your service in that gap, in that hole, we've got a problem. Fill the hole. You know what I mean? That's the exciting part for brands. We've never been able to do quite in that directive a way before and have those conversations.
- [12:47]
Jordan Cooney
And this is a great segue because you started earlier talking a little bit about fan out query and why that's so important to marketers today. And let's dive into that in this context of what we're talking about, which is how marketers need to better understand the relationship of filling the void, of answering the real problem that a user has, of being part of that discussion thread, that conversation thread of problem to insight, to actual CRM. This is a platform and a software that I need in order to track my leads. Right? Like that is a, that is a flow from problem to solution that many marketers don't understand. And fan out query is at the central, like, backbone of helping us fix that. Why is fan out query so important and why is it transcending the way we think about keyword research and content identity identification and development?
- [13:40]
Josh Bliskell
Because you will, you will put a 10 paragraph question into ChatGPT and ChatGPT is going to write three to five word statements. I'm going to say, I'm 28, I want running shoes, I live in Brooklyn, I'm training for my first marathon. I love running in the rain. Ready for this? Best men's running shoes. Best marathon running shoes, running shoe store, Brooklyn. Yeah, that's the authoritative, I.e. the actual queries that are being run on the search API. And so without knowing that, and I see a lot of marketers, you know, it's actually our saving grace. Like there was this terrifying moment before everyone kind of knew query, fan out. Where marketers were like, do I need to personalize for every single person that's going to experience my brand? Thank goodness. Not because query fan out helps us flatten down to what's actually productive again as judged by the LLM. Note, I didn't write best men's running shoes. You know, best marathon running shoes, running shoes store, Brooklyn. The LLM arbitrated and said, all right, I love that you're 28, not important. I love that you love running in the rain. Not important. Training for the first marathon, that's important. And you want to try those shoes on in person. That's also important. Okay, noted. So what we're going to look for here is we're looking for the authoritative actual queries and that's where we need to intersect our content with.
- [14:53]
Jordan Cooney
Right?
- [14:54]
Josh Bliskell
We need to literally aim our content to hit those fan outs because that's the actual search, that's the actual SERP that's getting drawn back into the answer engines and that's the pool of citations the answer engines can choose from when deciding if your brand, your product or service is going to show up. So again, the thing about that is that you can then have. This is where you kind of test do Personas matter for my brand, I'll set up very complicated memories, very complicated Personas and see. Oh, actually this like 45 plus buyer Persona, the query fan outs are actually the same as the Gen Z Persona. That's crazy. When they're looking for corporate cards, all that matters is actually the income level and their credit score. Wow. That's what ChatGPT does. I don't have to sweat, you know, best credit cards for seniors, best credit cards for, you know, 40 plus, best credit cards for first credit card for Gen Z. You don't have to make this piece of content so very cool. Anyway.
- [15:46]
Jordan Cooney
Yeah, but historically, right? So as SEOs and as, as, as now, I consider myself a, on the older spectrum of the SEO, I'm part of that Persona, I guess you might say. Oh no, it, I, I used to think about long tail strategies, right? Let's create these programmatic plays and let's get every single color, every single size, every single deviation of, you know, user that I could possibly think of and create an individual page that targets that specific user. And what you're saying is with fan out query data we can better justify where we are going and where we are producing content because not everything applies to our end customer or end user. So let's, let's dive into that for just one more second because I think fanout is just a massively underserved topic right now in our industry because I actually think it's one of the most important utilities that we can utilize in our strategy for doing great content work, doing great website work. One of the things I've noticed recently specifically in ChatGPT, not so much in Gemini, but it is when you go do a prompt, you're Starting to see ChatGPT showcase the fan out queries that they are using to generate this response. You're seeing them like, like cycle through them like very quickly. And I've seen that recently here because we're in the, we're in the holiday shopping season as we are recording this episode, Josh. And it's wild to see, like the activity and there's how quickly. And I'm just like, sometimes I'm sitting there looking at myself, I'm like, did I really write that prompt? Am I that dumb? Because ChatGPT did a much better job of concisely making the question that I'm putting in. So let me ask you, how is it that we get to this data and how do we get to this data accurately and quickly? Specifically, fan out query data.
- [17:35]
Josh Bliskell
Oh, my goodness. I actually just posted a LinkedIn piece about this last week. First of all, it's in the network logs. So in ChatGPT's network logs, you go to your URL, you type something into ChatGPT, press enter, go, hit your URL bar, there'll be a little code, It'll be like 20 something characters long. It'll be after the slash, it'll be kind of gibberish. Copy that. Open. Go to your inspect element. Open. You can tell I'm doing this from my head because I do it all day long. Go to inspect element. Go into the network tab of inspect element. There will be a filter there. Paste that code in, refresh the whole page. You will see in the Fetch XHR requests category type. So it'll be like a little orange icon. It'll be the same id. You click that. If you command F queries, you can literally see the fan outs there. Presented for time immemorial, you'll be able to see the fan outs, but you can also see when I'm doing demos on the front end, Perplexity is awesome. They actually just. They make the UI very clear. They're like, here's the fan outs. Like amazing for ChatGPT, sometimes they don't show it. I'm like, I'm like demoing to like C suite and I'm like, all right, trust me, there's fan out there. I'm not going to go into inspect element for them. But perplexity does it great. So when we're doing this data grabbing, what we're really doing is we have a profound. At least we have a network of insanely complicated network of proxies, thousands of different ways that we can represent a prompt and we go to the front end of all these different tools and so we're actually able to just capture the raw network logs as it exists, as the user experience would basically be. Right. So if you go to chatgpt.com and you asked a prompt, profound system would work the same way if you put that prompt into profound system itself. We do that billions of times. We love scale, we love sample sizes here. So that's basically, that's the methodology behind it.
- [19:12]
Jordan Cooney
Yeah. And I want to just one more layer of this onion because I think there's something of tremendous value not only for our audience, but for the entire marketing community at large, which is, candidly speaking, this is probably one of the first real data sets that we've been able to truly identify when it comes to ChatGPT cycling something back to marketers. Right. I mean, this is. And I've been, I've been a staunch critic of this. I'm not going to deny I've been a staunch critic of visibility in AI because it is a farce of us really truly understanding what's happening inside of these LLM responses. And candidly, I think it's not a great directional tool for a marketing operator, a performance driven operator, to go in leverage right now here where we talk about Fan out Query and we start to look at this, this, this set of data. There is a lot of science in terms of what the LLM is actually doing when it comes to a specific prompt and why that's important. And then you can also create trends over time looking at if that's changed, if that stayed the same, if it's consistent. Because that's going to ultimately indicate to us as marketers where we invest in content, why we need more of that on our website, why we need to be more authoritative in Reddit or other places longer term. And I think that's one of the major key elements of us as marketers getting smarter about doing our work. So this layer onion question. Sorry, it was a very long winded way of getting to the question, sorry, Josh, is tell us how Fan out Query is going to be used in 2026. How are we going to start to implement this data in our daily marketing tactics? And why is this discovery you posted on LinkedIn? But so many are gravitating towards a vital way of us doing planning when it comes to organic marketing.
- [21:12]
Josh Bliskell
I think, I think first of all, I think I'll push back a little bit there. I think visibility, I think visibility is interesting once you are calibrated. I think that is the key. If you can agree with that, I can agree with you there. Because if you're just going to go out there and pick 50 things at random to track and you're saying that's my visibility, what you really need, you need a few different elements the fan outs are really helpful, but I think in 2026 we are going to see prompt volumes collaborating with fan outs. We're going to take the top search prompts about you, your space, your industry. Profound has this data. This is the coolest thing that no one talks about. Everyone is like heuristically, culturally in the SEO space. Like, as we all know, nobody has the data. I cry myself to sleep every time I hear that. You know, an angel loses their wings every time someone says Profound doesn't have the prompt volume data we have. We have over 150 million individual prompts that were getting piped into our system every single month. And the cool thing about that is it's interregional, it's intermodal. There's tons of different models we're looking at. And so when we're looking at that now we can say, okay, these are the top most asked prompts about your space, about corporate cards. Now we plumb those into the actual system. Now what are the fan outs for the most frequent prompts period? Now what you've got is you've got actual volume confirmation and then you actually have the simplification of that real volume. So now I know these are the questions people are asking and then these are the three to five, you know, key phrases or pieces that I need to orchestrate my content to win towards. And maybe that's going to be something that, you know, the rest of the SEO world is totally not tracking. Maybe there's a word that they're into, maybe it's monthly, maybe it's changing the word corporate cards to corp cards that, like something so benign could change the way you're going to start to structure your content. I'll change my title tag, right? Don't mind if I do. If I'm thinking About this for 2026, prompt volumes are going to be at the core of everyone's strategy. Everyone wants to know if we don't get clicks, if attribution is as hard as it is, which it is. Attribution is still completely broken.
- [23:10]
Jordan Cooney
It's very broken.
- [23:11]
Josh Bliskell
I need some volume confirmation here. I'm not going to be comfortable making this kind of work make sense up to the C suite. Unless I know these are the number of people who are searching this prompt. And I think there's a multi billion dollar startup sitting out there just to solve this attribution problem of people copy pasting the name of the product they saw that was visible in ChatGPT into Google and then transacting there what we know from data is that Google. For people who are using ChatGPT to evaluate products, Google's basically relegated to the navigation and transaction layer. ChatGPT is the actual consumer decision journey. That's where it's happening. Like I said, come in with the problem, contextualized answer, feel like you have emotional buy into that actual product that solve, then go and transact on it, make the decision.
- [23:57]
Jordan Cooney
Let me ask you this question because you brought up prompt volume and I think prompt volume is a really important piece of this puzzle. I mean, I know it's a segue from fan out query and I think, I still do believe centrally that there's a big, big piece of the marketing strategy that, that, that relies on that. But let's go back to prompt volume and let's go, let's go dive into that for a second as we think about prompt volume. One of the big differences that I'm aware of right now is that LLMs are a much more fragmented market than search has been because search has largely been dominated by Google and, and many of the, let's just call it developed nations will go with the World bank definition for now. And then there's a handful of other countries where there's other search engines, South Korea, China, Russia. But the reality is that like largely Google's been this central theme and historically there's been this great exchange of Data since about 2002 where Google had created this lovely thing called AdWords. And what companies did is they would pay money to this company called Google to have ads in the search engine. And in return, Google would give them data, lots of it. They would give them at one point data on the actual query, they would give them data on the user, they would give them lots of data. They've since declined the amount of data. We could talk about that some other day. But like the amount of data has been a central part of how we create really clear what was historically called search volume data. And now we are talking about prompt volume, but we don't have this exchange. There is no real monetary exchange with ChatGPT. There is no real monetary exchange with Gemini, Google's version of this and the like. I mean, there have been a couple of attempts. So yes, we can argue that some companies have tried to build an advertising platform. Perplexity, notably, notably has been very perplexed lately. But anyway, that's a whole nother conversation of what's going on with advertising and shopping and how that is transforming inside these lms. It's obviously going to come, right? I mean we all know it's going to happen. But that exchange of data I think is so central to quality, you know, data points that we can use as marketers. How are you, Josh, or profound thinking about this backbone of prompt data? How are you thinking about it in terms of the vacuum that is what we actually have from a provider versus what it is that we are aggregating or collecting and sampling? And then the last piece of this, because I know this is a giant mountain of a question, so I apologize now in advance if you need to break it down. But like the other piece of this is there's so much uniqueness in terms of the multiple LMS that exist, multiple user specific needs or personalization as they used to call it, an SEO or search. And then to add on top of that the multi region language complexity that we're facing with that almost none of these LLMs have actually been able to solve yet. So, so mountain of a question, sorry in advance.
- [27:15]
Josh Bliskell
I think I have a few different things. I think the number one, the number one thing to realize about prompt volumes is that the way that Profound procures the prompt volumes is through we have over 42 different data panels with over 5 different true sources of data across the world. It is a consumer data panel of opted in users. So just like a Nielsen, basically we're going out through things like clickstream data, those kind of elements, very common kind of data collection method. But of course the same way that to a portion semrush is kind of creating their own data. You know what they did it with keyword volume initially we're going to go out there and we're going to build this map of prompt volumes and the focus of that is based on what people are entering into ChatGPT versus what they're entering into Google. It's very similar to traditional keyword volume database creation. So if you're comfortable with semrush hrefs, great. This is that just for actual prompts. So beyond that though, I think the, the strategy here, for me at least because of how multivariate this whole space is, I use prom volumes as a means to an end. Prompt volumes are good only to get me to the fan. Like this is how core fan out is. Maybe I'm going to glaze you up a little bit, but fan outs are the core. Everything else exists in service of the fan out. So long as fan outs are going to be at this central level of the process because the prompt tells me the user problem and I can see full conversations, I can do conversation modeling, which is helpful. So the prompt gives me the context for the moment. But then if I look at the fan outs, then I actually can understand authoritatively. So let's just say I'm asking a question in Spanish in the U.S. now, if ChatGPT takes that question, that prompt, and I know that there's some discussion out there that ChatGPT right now is translating, literally converting different language prompts into English, if it takes that, you know, prompt and converts it into English, I'm going to move even on that strategy where I'm trying to really rank for a Spanish speaker, if it's going to get me cited, I may move to create an English version of the page first and then maybe create a Spanish version of the page second because ChatGPT may be citing that English source first. So the real it's, it's really like a multi gate system where it's like, all right, you pass the first gate, you've got the volumes. Second gate is what happens with the prompt once it actually goes into the system, and then the third gate is what actually gets cited, right? So you're looking at how each of those things end up adding circuitous turns and unexpected little caveats to your expected system. So I think the sin here is if you look at the prompt and you think, well, that's it, that's the prompt. We simply must track that because we don't know how it changes through time. We don't know how the actual prompt goes through the system and ends with the citations. Which is why when you have the prompt volume, that's all well and good, but see what comes out the other end. See the citations, see the responses.
- [30:11]
Sponsor Voice
Time for a one minute break to hear from our sponsor, Pre Visible. So you're looking for SEO help and you got a couple of options. You could start replying to spam from agencies that claim they can get you to rank number one on Google. You can pay an hourly rate for a consultant who will inevitably nickel and dime you with hourly charges. Or you can work with a cookie cutter agency to quickly launch a strategy less project with low success rates. None of those sound very good now, do they? Well, that's where Pre Visible's integrated consulting model comes in. Pre Visible draws From a collective 40 years of SEO and digital marketing experience to unlock your organic growth opportunities. They build custom solutions that combine strategy, technical expertise, content and reporting to effectively operationalize SEO for your business. Pre Visible's four stage approach ensures that your SEO programs thrive by starting off with a strategy first. Approach, then they support you in your efforts to create quality content, help you identify technical issues, and most importantly, they'll work with your cross functional teams to integrate your SEO strategies to make sure that your SEO budget actually drives results, not just your agency's bottom line. So join brands like Yelp, eBay, Canva, Atlassian Square, all who rely on the SEO consultants at Pre Visible. For more information go to Pre Visible IO that's Pre Visible. P R E V I S I B L E I O.
- [31:39]
Jordan Cooney
I think one of the pieces to this that I'm curious to get your opinion on is the personalization aspect of this. Right. And I know you shared a personal example earlier. Hopefully you're not training for a marathon because it's terrible for your knees. But you know, I don't want to sound like your mother, Josh. Right, Good. But on this note of running shoes 28 like the rain, there's also a ton of other contextual nature that an LLM can use in a much deeper way that a search engine probably couldn't or could only use maybe historical reference like, oh, this is a user who loves visiting New Balance. So I'm going to try to serve up New Balances in the shopping feed. Right. And we know that Google's been doing that for a long time, but LLMs have this unique capability where they can actually know a user's much deeper intentions or desires because they're building context. Right. So I may even at some point have prompted to say something like I want to try something different from New Balance. Even though New Balance is something that I visited over and over and over again, Google would never get that. Right. The Google search results would still show New Balance options. Right. Whereas the LLM would be like screw New Balance, we're moving on. Right. You know Brooks, it is. So the question for you is how. How do you think about that component of personalization when we think about both prompt volume but also let's take it the deep a step further and talk about fan out query as well.
- [33:04]
Josh Bliskell
Yeah, I mean I think that there are certain I I would call them like basically prompt modifiers, things that completely change the intent of the conversation. And, and that's really where the we have a Personas tool. But basically I'm going to stick away from talking too deeply about profound but we have the ability to model. You know, you love this brand, you hate that brand, you have this need, you don't like that thing, you're training for a marathon, you're just getting into running. Like those things basically we can model very easily and we can even model things like I don't want to see New Balance shoes. Like we can, it's, it's quite simple to do that and to throw that in the actual response. The thing that I think is important here when we're thinking about that hyper personalization is really following out the actual fan out to completion. I think there's a need to actually get analytics on when we think these buyers are going to be doing these things and when we see them doing it with prompt volumes or we don't see them doing it with prompt volumes. If they say, you know, if New Balance comes to me and says, you know, I think, you know, a lot of our consumers are just, or maybe Nike, maybe this is better. Nike's kind of the standard. Nike comes to us and says, I think our consumers just want to try something different. What's happening? Like how do we keep them? Or maybe Apple, everyone wants to switch to a Samsung or a Pixel. How do we work on that narrative? I think the first thing you have to do is to start analyzing those prompts and those fan outs like is it a done deal? Is the fan out, say Google Pixel? That's a very different situation and we see that a lot. By the way, research coming out there on that branded fan out. What's the percent? There's a lot of branded fan out where guess what, you didn't have a chance and you didn't know it because the fan out was already saying Google Pixel in the beginning. So you've got to see where the LLM is already driving that user to go based on that actual high intent. So what I do a lot of the time for those things. Those, like those prompt variables, those prompt modifiers that just completely, you know, sideline all sense of like linear decision journey because let's face it, most consumer decision journeys just aren't linear. For those caveats, you've got to build those into your model. You've got to see what those citations are. It's the same thing. Wouldn't you want to know if people say I have an iPhone 16 and I want to switch, what does get cited? Does an Apple forum get cited? That would be amazing. Wouldn't that be awesome? Like Mac reviews, like other Apple users maybe talking about how they switched and didn't like it or things like that. So there's a lot of strategies to build out there, especially when you're considering what those kind of consumer behavior ticks are. But it's hard to predict it other than with the prompt volumes and I think the prompt volumes are a, I don't think they're like 100% there yet to be able to predict what those, like those, those unexpected things are. But I think I know the marketing team at any of these companies has a pretty good inclination as to what kind of problems and what kind of, you know, questions that normal buyers asking.
- [35:47]
Jordan Cooney
And this is a great segue because you brought up marketing teams as SEOs, as content marketers, as AI marketers. Let's just call them AI marketers because everything's AI now, right?
- [35:59]
Josh Bliskell
I love it.
- [35:59]
Jordan Cooney
What are the conversations that we need to be having with our partner teams with PR with social with community teams with. And if we're a small company, let's, let's take the small company where like I'm a sole marketer, I am the VP of growth and I do all marketing and I have a whole toolkit of swords, knives and bows and arrows to do my marketing stuff. How do I think about my own conversation about working within these influential channels that have an outsized level of importance inside of LLMs?
- [36:36]
Josh Bliskell
Well, every surface that your brand inhabits now online now trickles down to the end goal of visibility whether or not you want it to. Like I don't. If you're on a PR team and you're like, well my, my PR stuff isn't meant to help my brand visibility. Like unfortunately it can be factored in to determine if you're visible or not. It is going to be used as part of the surface. So it's like you can either create and have that knowledge innately that like okay, everything we produce is now going to be factored into this like all knowing search system. Or I'm just going to keep doing the thing I was doing and not really link in the hard question I get. I'm glad you asked it this way. People ask often, should I be talking?
- [37:12]
Jordan Cooney
Of course.
- [37:13]
Josh Bliskell
This is. Search has gotten so much more horizontal in the last year.
- [37:17]
Jordan Cooney
Yes.
- [37:17]
Josh Bliskell
You know, social teams, PR comms, affiliate paid, all these teams need to talk to each other. A PR piece that was really nominally meant to just kind of announce a product could contextualize the accuracy of how that product appears in ChatGPT for a year. Could be a multi billion dollar product if you're a VP of growth. I'll go back to the small company example a little bit, then maybe we'll bring it a little more macro. If you're a VP of growth, you have to think about things along those horizontal lines. Nikki Lam, she Had a great piece at Brighton last September, which is just, and we've said it before, but search everywhere optimization. And I just think that's the people who are talking about it even earlier in the year. It's just so true. It really is. About these multi surface, what I would basically say is you're really building consensus. These multi surface consensus building activities where the accuracy of how you're going to be represented, the coverage that you have in multiple different areas. Are you working on your affiliates? Are you prescribing to your affiliates how to keep narratives the same? Is your social team on Reddit aligned? If you're a VP of growth, are you inhabiting Reddit appropriately? Are you thinking about a Wikipedia page? All these different services now can help you be more visible in AI search and thereby drive real clicks, things that necessarily weren't necessarily revenue adders, you know, marketing channels. Not nearly. You know, if we're looking at like PR for instance, PR is awesome, but a lot of brands and a lot of companies don't realize how many clicks it can drive when it's actually applied in an answer engine context. Because there's just often so much rich context about products, about services, about releases. And guess what's great with pr? It's all fresh, it's all recent, it's all now. So if you need that context, all these people doing deep thinking and deep research prompts, those prompts, you know, it's around 5% of users, they have the time, it's a small percent, but they have the time to actually dive in and know where your products are coming from, know what your products are doing. Otherwise, if you're doing the instant responses, what you're going to see is listicles start to win. Things like that, you know, for most things contextualizing your actual product, where it sits in the industry, here's a listicle. Here's the top 10 products, here's where things go. The reason that works so well is because these answer engines are lazy. They want to have someone do the legwork in the room of saying, I don't want to figure out every corporate card. I love this source. It was published three days ago. It's got every single corporate card on there. It's got every welcome offer, it's got all the different interest rates. I'm going to pull the top three from here. Someone already had a framework for analysis. I'm going to throw it in there for deep research. It's going to go to your press page, it's going to go to your Product pages, it's going to go to your landing pages, it's going to go to Reddit, it's going to look at reviews. That's where you know you will win or die based on the information that you have specifically. And forget about it. If the user has like a hyper contextual answer or a hyper, hyper contextual question. It's my very first credit card. Again, that's one of those modifiers that like, you've got to be cognizant. How does it actually start to model it out there? How does that journey differ? How does that set of citations differ? It's very cool.
- [40:19]
Jordan Cooney
All right, so I have like three questions, two very practical ones and then one that's maybe a little personal. So. Okay, I'm not going to give you the mountain. I'm going to give them to you one at a time this time.
- [40:30]
Josh Bliskell
Josh.
- [40:31]
Jordan Cooney
Sorry. So my first, like really practical question on this is as we think about these conversations amongst teams or even just companies that are trying to really reestablish their brand in an LLM response, what's the marketing strategy and advice that you're giving these, these companies and brands to help them start to build that credibility? Right. Whether it's it's a company that already has some negative sentiment or it's a company that wants to build a better story or narrative, how are you thinking about supporting those teams in making that, that effort?
- [41:10]
Josh Bliskell
It's got to be content, content, content initially. Like if you want the vz, like what you should do tomorrow is start doing content experiments, start looking at your network logs, start looking at how these answer engines are accessing your pages.
- [41:21]
Jordan Cooney
Love that.
- [41:22]
Josh Bliskell
A technical test would be great. Do a little test. JavaScript, Gemini Copilot, Microsoft Copilot, they can both now render JavaScript. So that's very interesting. Right? A lot of big blockers there for a lot of legacy brands I was working with. Oh, we've an outdated CMS. Things that FAQs populated by JavaScript. Not ideal. Right. But do that test, check it out. But then beyond that, releasing that very first list piece of content, or releasing that very first product landing page with a really dense table or a lot of information specifically geared about your offerings. In December, you know, really starting to hone in on specific questions that you think users are going to be asking right here and right now. It will help you understand what you need to be doing to win in your space and what you may have been doing already that might not be optimized for answer engine retrieval. Like the way I talk about this is that basically we are in a world where we're working out the same muscles, but we're doing totally different exercises. We're still working with URL slugs that didn't go anywhere. Schema is still here. I didn't say kill your schema. You're working on page content. All these things are still.
- [42:27]
Jordan Cooney
That would be very scary advice, Josh.
- [42:29]
Josh Bliskell
Terrifying. We're still using the same units. We're using them in slightly different ways. And so thinking about exciting ways to do that is going to be important. I mean, for SEOs, very clearly, this is a huge moment. CMOs, your C suite, everybody is looking for someone to raise their hand and say, I know about structured data and these scary systems that rely on so much structured data. This is actually an extension of what I already understand. I have the context, I'm here to play, I'm ready to go. And the same thing with E commerce, I think E commerce is this. I see engineering and product teams in a lot of these E commerce decisions. How do we get our product feed in there? How do we optimize our product feed? I'm like, where in your org could there be someone who specializes in structured data? Could it be the search team has some knowledge here and they're like, oh, yeah, yeah, that would be good. The search team has such a mandate right now to raise their hand and say, we can help with this agentic commerce thing. We can help with instant checkout. We can get you ready, we can tell you about how to do this stuff. There's a 5,000 character description field in ChatGPT's product feed. Where are my SEOs? Where are my SEOs doing that work right now? I love you all and I want to see, you know, I want to see hands raised in the world, in the industry. It's amazing. Right now the opportunities are just abounding.
- [43:39]
Jordan Cooney
Absolutely. And my next practical question, which is a great segue on this, is I want to know how profound is helping these teams connect these dots. And here's where the intention of my question is coming from. And I think this is a super important one because if you're at all connected to search, paid, organic, any way, shape or form, the noise around geo AI platforms tools is astounding. At this particular point, myself, having come from the SEO tool space and having been at search Metrics for many, many years and being a leader in that organization, but also being a voice in the SEO community and have worked with many of these tools, the amount of inquiries and questions that I receive about the tooling space around AI Discovery and LLM tools is astounding. It's baffling. So the intention is coming from there, Josh. But I really want to know how is Profound helping teams make these connections, make these in many ways seemingly highly complex things to common marketers. Things like schema and fan out query, like the stuff we've been talking about. Most marketers would look at us and be like, those guys are crazy. And so how are you guys connecting those marketing teams and departments? What is it that you have that is uniquely valuable in the market that nobody else is hearing or seeing about and is helping build those bridges?
- [45:16]
Josh Bliskell
The crazy thing that we have right now is we have a full orchestration suite complete with workflows. And so what we really want to do is we want to say, you know, there are basically two views of the profound platform. There is like the C suite view where like I just want to know my metrics. I want to understand what we're doing, how we're defending, what's up, what's down. Vaguely just generally like two sentences. Why? Why? What's going on? And then we have the actual. I call it the nerd view. I do. We have profound the tool. We have refound the tool. And profound the tool has got so much data in there. This stuff that you're talking about granular gets you right into the responses, lets you see the fan outs, everything. The thing that we're doing with our orchestration is saying to whatever level you're comfortable, you can now customize flows that do this analysis for you and give you simple actions to take direct opportunities, create content for you, start to think about distribution, give you action plans for improving or thinking about how you're going to loop in other team members. I think if you're thinking about, you know, I want to jump into AEO and I think this is going to be super overwhelming, though I don't really understand it. The thing to do is to really spend a few, like check out the starter plan or even if you don't even want to do that, go to Google Sheets, open a Google Sheet, go to chat. I'm going to kill the starter plan revenue right now. Go to Google Sheet, open a page, literally search 10 queries in ChatGPT, grab the sources, look at the brands that are mentioned, go into the network logs, look at the fan outs. It's all there. The fan outs are there for you. You can see them, grab them, monitor them, come back the next day, do the same thing, put a bar Chart together, put a line chart together. Look at those things, edit them, edit your pages. Look at what's going on. If you look within the next few weeks, you may see some changes. If you're starting to change those pieces of content, you don't even really, you don't even need a tool suite at that level. You just need to understand, you know, basic marketing. Once you want to get advanced about it, though, you could say, hey, I want to win for men's running shoes. I'm Nike. I'm going to come into the platform. I want to win tomorrow. Let's do some orchestration. I put together a few of these analysis nodes in our workflows tool. Boom, boom, boom, boom, boom, out. We have two or three different content plans, a few content briefs, a defensive plan, and basically how we're going to actually structure the release of this content. It's really, it's unlike anything I've ever seen before from a marketing tool standpoint. Because what it does is it says the only outcome right now is whether or not the bot clicks are being driven directly to you. And the thing about it is that you have to rely on a few authoritative data sources, like fanout, like prompt volumes. And those data sources can be analyzed relatively cleanly on the fly in a programmatic way. Once you have the baseline set up, once you say, hey, I'm this brand, I want to improve citations, I want to improve human referral traffic, I want to understand, even I want to understand my sentiment. Once you set your goals up, it's really easy to start thinking about what might be missing, what might be here. You might ask, why is it so easy? The reason it's so easy right now, I think, and this is my philosophy only, is that we're still in the kind of first turn of optimizations in this space. We haven't reached the point where everyone's armed up with the same data, where everyone's doing the same strategies, right? In SEO, we're at the 10th dimension here. You know, if everyone listicles right now win an AI search, let's just. And they do. But let's just say, hypothetically, listicles are completely dominant. Fine. Next year, everyone has listicles. What happens to that strategy? As it proliferates, its effectiveness declines rapidly. Right? We all know this. We're all SEOs. We understand how these strategies end up proliferating. They're like little hacks and they're gone, right? Awesome. Then we reach the second layer of where, okay, maybe there's something we do within the listicles. Then that gets us a little more juice versus a regular listicle. Or maybe it's listicle plus something else. Or maybe it's a hybrid of something. Right? We're still in turn one where people haven't yet even consolidated onto the few things that are very clearly working in this space. Once we do that, things start to get more indirect. It starts to become about combining variables and testing multivariate, multi arm bandit statistical tests. It gets really in the weeds right now. Blue ocean. It's really actually calmer than most people would have it believe. It's a really good time to start adopting and learning. It is a lot to me, like bowling with the bumpers on.
- [49:27]
Jordan Cooney
So my personal question on this track, this is where I think things get kind of interesting for us. And I hope it opens a new door for you, Josh, which is, by the way, as a fellow Midwesterner, this is a Midwestern question. All right, so I'm gonna ask how mom's doing. I'm gonna ask about the turkey soon. All right. But you're sitting in Champaign, Illinois. You graduate from the University of Illinois in advertising, I believe, if I'm not mistaken, from our planning meeting. And how do you see yourself right now? Like you're sitting there graduating and are you really thinking, oh, I'm going to be having conversations as an AI researcher to different marketing teams and different marketing stakeholders about how to use AI. I want to know how that journey came to be and why you're enjoying having this conversation with these various stakeholders.
- [50:20]
Josh Bliskell
The most fun thing you can do in your life is make discoveries, in my opinion. I dislike helping enable other people's frameworks. I really, I go towards the question marks. I have this incessant need to figure things out. So when I, when I initially graduated from U of I, you know, shout out Illini out there. The mandate for me was to try to go head up, like insights or strategy at an ad agency or maybe do some digital marketing. And I jumped into product. I jumped into HubSpot, rather, on their SEO team to get just a little more experience in data and analytics and really think about data first. Marketing in advertising, they teach you copywriting, they teach you creative executions. It's very qualitative. It's very tasteful kind of work that you're doing. You're trying to make an ad, you're trying to sell an emotion. With SEO, you're almost in the exact opposite situation. I thought that was such a wonderful opportunity. I was like, I want to immerse myself. Because at the end of the day. It's really easy to look at a marketing strategy that's based around data, based around actual tests and statistical tests rather, and say, this is working, this isn't working. It's very hard to look at advertising and say this generated X dollars. And I think as we look at the evolution of marketing, things are moving in that direction anyway. They were moving in that direction even in 2018. I mean, obviously. But once we got into SEO, it became clear for me at least that SEO, at least in 2020, 2021, 2022, it was going to be pretty tracked. It was going to be, it was established, the strategies were known, we were holding on, the updates would come, we would adjust. The updates would come again, we would adjust. It was okay, nothing was crazy. ZeroClick was kind of like, there's going to be a lot of featured snippets and that might be eating into our clicks, but okay, we still got, I mean HubSpot, we've got great blog content. It's going to be great. And so ChatGPT comes out and I was appointed to head up. Why don't you just stay on top of whatever this chatgpt becomes? That was the mandate I got in early January, which was so lucky by the way. I had played with GPT2 and stuff. I always joke with my wife, my wife was peeved at me. I was in my stats class, I was sitting with GPT2 just generating, just nonsense jokes, whatever I could make it generate. She's like, you need to focus on your stats. We're going, this thing can talk. It's garbage, but it can talk.
- [52:37]
Jordan Cooney
You don't understand, honey, this is good jokes, this is material, this is stand up material.
- [52:43]
Josh Bliskell
But I will be 100% honest and say I never thought that would be my work. I will be the first to say I thought it was a neat toy, a neat tool. And then when ChatGPT came out, that is when I realized everything was about to change. Because my aunt, my grandmother, everybody was talking about it. And it's those moments. The same way with bitcoin. Once grandma starts talking about bitcoin, you better hold on. Things are gonna change in the world. Nothing, no shade, grandma, but like it's going to change. So as a whole movement of society and as things are moving, AI is getting bigger. I head up. I launched with David Groeschel, really talented marketer. He now works, he runs his own pickleball company. Very cool guy. But he and I start the marketing, the AI marketing solutions team over at HubSpot Awesome. We're building automations. We're built like if you got, if you downloaded content from HubSpot, I would write the email that would ping back to you with AI that was personalized to your exact journey that you had taken on HubSpot. I saw you looked at this event planning piece after looking for marketing plans. Do you need something about like measuring leads after an event? Here's a blog post. It was awesome. Really cool stuff. Very like that was early days of AI and the SEO team, after I had left them, came back and said, you know search and you know AI, do you want to build HubSpot's AI search creator? So they brought me on back again to collaborate very closely to build AI search creator. And that's early 24. By late to mid 24, this was taking off in a pretty big way. Profound had come out of stealth and I thought, well, this is it. I mean I was pretty good at AI search over at HubSpot and I was like, I want to do this. I want to just lean super hard into this. We are not going back to blue links conversation. The ui, the ux, it will change fundamentally and with that, marketers are going to need to adapt. That's my mountain. I gave you the mountain back. That's a long answer, but it's true. Totally unexpected.
- [54:31]
Jordan Cooney
I love the story and I really appreciate you sharing.
- [54:33]
Josh Bliskell
That's.
- [54:34]
Jordan Cooney
That's awesome. So last question here. In terms of just general direction of what's happening with AI discovery, I have to ask this because we're in the middle of holiday shopping and it's really about just agentic commerce and how fast this is moving. And here's the brass tacks. I think this is a testing year for all the LLMs. They are screwing around and seeing how consumers are really going about shopping so that they can be ready next year to make a ton of money. And I think that we're seeing really crazy iterations and changes. Even my team, we've been playing around with a. We have, we have a test that we're just about to release where we tracked what ChatGPT was responding with prior to Black Friday. So last Friday. And then we're just, by the way, we're just right after, we're recording right after Black Friday and this past Black Friday and it is like night and day. It is just crazy, crazy, crazy. Different responses, totally different fan out queries. Just. They're just adapting so fast as they're seeing users behave differently because the immediacy and the desire In a moment of shopping that is something like Black Friday changes by the second.
- [55:46]
Josh Bliskell
Yes.
- [55:46]
Jordan Cooney
So anyway, tell me how you're seeing agentic commerce. What are the key foundational elements that are required for marketers as they not only think about the rest of this holiday season but even next year and they plan for 2026.
- [56:00]
Josh Bliskell
Everything is going to come down in a year's time when we're thinking about commerce we're going to think the idea of web based rag for commerce is disgusting and gross. And what an incomplete and gross system that was. If I ask ChatGPT, I want to buy a men's belt from Macy's what's in stock. I do not want ChatGPT to go try to scrape Macy's website in the future. ChatGPT is going to go, it's going to use a couple connector, it's going to use an app, it's going to hit Macy's back end. It's going to say hey Macy's, what's in stock? Macy's is going to say this is the belts that are in stock. ChatGPT is going to say thanks, I'll show this to the user. This is the future of commerce. It is going to be database driven, it's going to be feed driven, it's going to be connection backend driven. It's these, these merchants are going to start integrating and we're all going to see those integrations coming to full fruition. Now what the user behavior right now. This is what I predict at least the more technical spectrum, right? Yep. More technical products. The more technical your product, the less likely you're going to use Agentix Shopping to actually transact and actually purchase in the model. I don't mean that you're not going to come to a decision. I for one have like decided my dream car Based on what ChatGPT has told me my dream car should be. It's like yeah, you're a Ford Bronco. I'm like great, thanks for telling me but I'm not going to buy that through, you know, ChatGPT checkout. Whereas if I'm going to ask can you please plan me a party, buy me some napkins, a key lime piece and get me some foldable chairs. I feel very comfortable just dictating that off. And so there is this kind of like need and when I say technical product I mean anything from like a car to dog food. I want to know the formulation. It's important. Credit cards, those kind of things are very technical products. They require a lot of Time and consideration. But for this kind of like non novel shopping, shopping is a chore. I see. That is where ChatGPT really should start to optimize in the short term. See if they can get users to just get comfortable with doing this purchase and buying it. Like for me, maybe I'm a little antiquated, but when I even buy concert tickets, I don't do that. Like I, I sit down at the desktop and I purchase them like it is a. I mean, I buy them on the desktop. It's a desktop decision. It's a desktop, it's a desktop. Same thing with planning.
- [58:00]
Jordan Cooney
It's a whole new Persona.
- [58:01]
Josh Bliskell
Josh.
- [58:01]
Jordan Cooney
The desktop buyer versus the mobile buyer.
- [58:04]
Josh Bliskell
I'm gonna sit down, I'm gonna buy my flight on the desktop properly. I can, I don't want to click anything wrong.
- [58:10]
Jordan Cooney
Damn it. That's in row G. I didn't want G, I need to be in row C. That's true.
- [58:17]
Josh Bliskell
So we're looking at these things and we're like, okay, the consumer is probably going to be non technical for the short term and a lot of these conversations and conversions are probably going to happen through back end systems. If I'm a brand right now, if I'm thinking I'm a three person plumbing company, what do I do? I think you don't have to have a fortune 500 resources to start winning in this space. Start thinking like a Fortune 500 in that I work with Fortune 500s all day. They're not all in the product feed. Two of them are in the product feed right now and one of them is half in the product feed. Anyway, we are looking and thinking, how do I translate my plumbing company into a product feed? Look at the product feed spec. So say like, oh, pipe cleaning $75. Here's the description, here's the thing. It's a service based thing. Here's how I would basically translate that into a JSON object so that when someone says, what's the best pipe cleaning? My pipes are eroded. How do I go there and get this, I'm in Minneapolis, Paul's plumbing shows up, boom, boom, boom. Because in the description field you said you're a plumber based in Minneapolis. We do full service plumbing, blah, blah, we're certified, we do this, we're board certified, whatever. We do all this stuff. This is how you can start thinking about it. You don't have to upload your product feed just yet, but you do have to think eventually when this starts to come through, how are you going to contextualize what is that going to look like for you? And it's important.
- [59:34]
Jordan Cooney
Absolutely. And I couldn't agree more with you. Data and feeds is going to be the central, most critical component to winning in commerce in the next year. But there's going to be moments in time. And these moments in time is where I think the creative components of commerce are going to come into play. Whether it be major events like Black Friday or for very seasonal businesses, things like Valentine's Day or other major selling events, the creativity aspect of commerce will come into a critical component when it comes to an LLM, being able to understand what you're offering and how it's different. Where will marketers, SEOs, content marketers play in that zone? What will they be doing in those moments to win?
- [60:20]
Josh Bliskell
Ooh, I think I'm going to loop back towards context. Like right now, if you want to win in commerce, you have to just contextualize your product well on the pdps, even though I think it's not optimal. We know that FAQs, for instance, we took a sample of 10,000 top performing PDPs. 10,000 sample of bottom performing PDPs. FAQ content was 848% more prevalent in our top performing set. FAQs. When you see a top performing PDP, there's an 8x, almost 9x higher chance that that PDP is going to have an FAQ. Wow. Which is a really big contributor to it actually being top performing. It's very interesting there. So it really is about adapting and accepting that. I don't think anything right now is out of the question for some of the content changes we're making. So long as it's helpful. I think the heuristic you should use is am I being pushed in this way? Yes or no. And then is this going to be long term? Does this basically help OpenAI create better answers or not? You want to be in lockstep with OpenAI. You want to think, if I'm OpenAI, do I look at that piece of content and say, oh my goodness, I wish everyone created pieces of content like this? It makes me ChatGPT's answers so good. It makes ChatGPT look incredible. It makes ChatGPT look smart. You know, data driven. This answer was awesome. And the source that provided it was flipping amazing. That's the mandate. It's like, how can we align ourselves? Because it's the same way with Google. SEOs are probably rolling their eyes. They're like, yeah, four letters, buddy, E, E, A, T, bud. That's the situation. I get it. I'm with you. But I think there's a wider world of marketers that have an interest in this space and everybody needs to get online with that because people will see 848% FAQs. And the deal is this. Not everyone knows that that doesn't mean just go out and start adding random FAQs. That means you gotta align yourself. I gotta say it. I get a lot of flack from SEO's. Like, it's very similar. Like, well, not everyone's an SEO who's jumping into this space.
- [62:06]
Jordan Cooney
True. Yeah, very true. Okay, that wraps up this episode of the Voice to Search podcast. A huge thank you to Josh Blisco from Profound for joining us today. If you'd like to get in contact with Josh, you can find a link to his LinkedIn profile in our show notes or on the voicesofsearch.com. if you'd like to visit his company website, you can visit try profound.com if you haven't subscribed yet and would like a daily stream of SEO and content marketing knowledge in your podcast feed, please hit the subscribe button in your podcast app or on YouTube and we'll be back in your feed every week. That's all for today, but until next time, remember, the answers are always in the data.