In this episode of The Visibility Brief, Yext SVP of Marketing Rebecca Colwell is joined by Garrett Sussman, Director of Marketing at iPullRank, for a refreshingly human conversation about how people really search — and how AI is amplifying the psychology behind it.
Marketers love to talk about algorithms, platforms, and technical optimization. But what if the biggest variable in discovery isn't the system at all — it's the person behind the keyboard? How we frame questions, what we believe going in, and the bias we carry with us all shape the answers we get back, especially in AI-driven search experiences.
With a background in psychology, literature, and technical SEO, Garrett brings a unique perspective to the table. This episode explores how identity, bias, intent, and trust influence search behavior — and what brands can do to show up accurately, even when they're not in control of the interface.
The episode breaks down:
Why search behavior is as much psychological as it is technical
Why two people can search the same thing and get totally different answers
The rise of conversational search
Why AI answers feel more trustworthy than links
How LLMs filter, summarize, and simplify information
Why reviews, forums, and social proof now shape brand visibility
What brands can influence in AI, even when they can't control it
If you've felt like discovery is getting harder to predict – or harder to influence – this episode will give you a new lens. It's not just about ranking for keywords anymore. It's about understanding how people think, how AI interprets intent, and how brands can adapt by focusing on trust, structure, and a holistic, human-first approach. This episode will help you rethink visibility from the inside out.
Episode Links
Transcript
00:00:00.160 — 00:00:29.000 Search platforms are changing fast, but one constant is that every search starts with a human. A human with a question and how they ask that question can dramatically impact the results that they see. In today's episode, we are delving deep into the psychology of search and why it matters for brand visibility. I am so excited to geek out about this topic with my friend Garrett Sussman, Director of Marketing at Apple Rank. Let's get into it.
00:00:36.960 — 00:06:36.060 Hi Garrett, I am so excited to have you here today. Rebecca, thanks for having me. Let's. Oh, there's so much to talk about. Let's dive in. We're gonna get so geeky. Um, before we get into the really fun conversation, um, I would love for you to. To give our guests a brief overview of. I pull rank, the incredible agency that you're with and your role there. Okay, so I've been there for the past five years. I'm working remotely in Northern Virginia, started by Mike King, who actually was a search engine land marketer of the year back in 2025. We're doing a ton with AI search content strategy. We are a boutique marketing agency that serves a fortune 500 and mid markets. And so we are very technical SEO, very AI search, very content strategy in terms of what we're delivering to these clients to make sure that they're visible on ChatGPT, AI mode, perplexity, you name it. Excellent, excellent. Well, it's really interesting because you mentioned technical SEO. And as marketers, I think we are very quick to get into the mechanics of search, the platforms, the algorithms that are changing. But we often overlook the people behind the keyboard who are doing the search. And that's why I am so excited to connect with you today, because you start from a psychological lens and you study how people think about the language works and how the questions are formed. So I'm curious. What originally got you interested in this?
You know, it's funny. 11th grade. I'll take you to the scene. I had the most amazing teacher, Kate Winton, and I took this psychological literature class. So we're reading like, Crime and Punishment by Dostoyevsky or The Man Who Mistook His Wife for a hat. And I just thought it was so interesting, like sociopaths out there and how they think. But then you start to to realize that we don't all think the same. Um, and that origin story kind of led me to always being interested to how everyone sees the world differently. Um, and then my, you know, entering the world of SEO and ultimately AI search, you start to realize that when we are searching for information, nobody does it alike. It's really hard to understand what is search behavior, how it's changing, and how that ultimately impacts businesses. We do tend to assume as marketers that everyone's a monolith. Like everyone's just going to sit down and search in the same way. But I imagine if I shared my search history, it would be incomprehensible to someone else. Like, what was she thinking? Who who would share their search history? I mean, we're in a bubble. We live in a bubble and we'll get into this. But like, there are so many, um, kind of biases that we're blind to that influence, the way that that we actually look for a new toothbrush or a new doctor or a new, uh, you know, sports score, um, that we're looking for. All of it is very much informed by our identity. And so it was like, as marketers, we kind of need to know that. That's so true. I'm curious. We'll get into the biases shortly. But when you were studying psychology, psychology in English and college, is that right? Yes. Right. Yes. Are there components of what you study that you find to be especially Applicable today in your current role? Yeah, I think that we are very much influenced when you're studying about identity. Um, for instance, and we'll talk about this whole idea of like confirmation bias. But we as people don't love to be challenged. We want to find information that reinforces our existing beliefs. And so that very much influences content strategy. When we are trying to appeal to an audience, like, who are we trying to appeal to? Are you trying to be everything to everyone, or are you trying to appeal to people who already align with their identity with your brand? Or are you trying to, like, convert people and change their perspectives? Which is really hard to do, and there are so many different social psychological triggers that influence whether or not we we can do that or people are willing to do that, that I think can help you, you know, dictate where you invest your time with the content you're creating and how you show up for search queries, etc.
Absolutely. You know, it's interesting. We when we we just got through doing our annual planning for the year two. And when we think about search, there is a component in our plan where we say we are capturing search for demand that already exists. Like people already believe that they need a solution like ours, and we want to be the one that they find. And then there's this whole other category of marketing work we're doing, which is educating people to change their perspective so that they, you know, we can we can move them down the path. Very, very challenging to do two year points. Yeah. And there's so much nuance depending on like in SEO. We're so classic for knowing for saying like it depends on a business and industry a situation. But like every different type of search has a unique journey to it. Some are really easy and we don't think twice about it. Some are purely emotional. Some are like really deep analytical searches. Um, you know, whether we talk about classical marketing funnels with, you know, informational, the top funnel all the way down to comparing and choosing the business versus conversational search, which is so much more complex, where we use natural language in order to discover and evaluate and discover value. There's actually a really interesting study done by Google, um, where they determine this concept of the messy middle, where it is more and more like that, where people enter to search at a trigger point and they find themselves going back and forth through this loop of evaluation and discovery and decision making and exploration to ultimately they make a purchase or they just bounce all together. It's it's really tricky for us to navigate and to lock in
00:06:37.660 — 00:06:41.980 that messy middle is so true. I have so many follow up questions. Garrett. Okay,
00:06:43.300 — 00:16:15.979 let's start with when. So every search starts with some kind of question. Right? And we've got to meet people where they are. So the way that that question is framed can have a huge impact on the path that that user goes down, whether they find a really straightforward answer or whether it opens up different paths of discovery. Let's start with the mindset that I might have when I come into search. Is there a difference between me knowing I need something very practical? A I think you said toothbrush earlier, right? Versus I might be researching a vacation that I might be interested in taking. Like, what's the mindset difference between what type of question am I be asking? And even like the channel that I might be asking it on? Yeah. I mean, we talk so much, I think, in our industry about intent and there are such wide ranging types of intent. You know, classically we think about the idea of, um, informational, commercial, transactional, navigational. Like so many people go to search just to find a website. It's, you know, the the hilarious classic example is how many people actually type in google.com to search something in the first place, not knowing that they can use their browser bar to to do the search. And now we're adding Gemini in there, which is all the can of worms. But I think that you start you need to think about what is the search intent. It's that basic. Like if I'm reading a book and there's a word I don't understand, I'm going to search for the definition. And that's really easy and nothing search. If I'm looking for a restaurant to go to with my family, I have all sorts of constraints in my head. And historically we use a. I think the term from from one of the Google reps is is keyword ease, which is the idea of okay, historically, Google wasn't good enough to understand a natural question of saying, I want to go to a restaurant that's open until five. That's going to be moderately priced, that, you know, has good parking and is good for kids. No, you know, I'll type in restaurant near me, kid friendly. And so the expectations of the results are, um, you know, put put us down a path and then we refine, you know, Google we talk about the whole like Google search. Uh, zero click is Google is trying to keep us on the search engine results page and have us find everything there without even ever going to a business. Now, do you remember the days of using Google when you would combine different things? So you could you could say, you know, this thing plus this thing, exclude results that don't have this phrase or do have this phrase. It's like we were learning Google. Easy. I don't know logic in our searches. Google, Google us to do that. It's great. Oh my goodness. The tools like you become like this expert searcher. And it's funny, it actually made a comeback recently where you have the negatives. Um, because the idea of like putting your search in quotes versus not quotes will give you different results historically because you want the exact, you know, the exact phrase using a negative to remove something from the results. And now people are so frustrated with AI overviews that they started writing minus AI in their search, because they didn't want to see a generative AI result in the traditional Google search result.
That would do the trick. So those operators, they've deprecated a lot of them. Some of them still exist, but I feel like, you know, it's joking as as a as a elder millennial, um, we're like the printer generation where it's like we used to figure out all these hacks to make things work, where it's like, you know, people, people who, you know, are younger. I mean, I don't want to stereotype entire generations, but like, let's say the Greatest Generation can't, and the newest generation can, and we're in the middle, like having to figure out tech for them all. Yes. Yeah. We've we've navigated a lot of changes, that's for sure. Um, it's interesting because the next generation, we've all learned to speak Google and search in this really kind of unnatural way. Now we're moving to more of a conversational search. And I am curious if you find that people are asking different types of questions when they go to an LLM that they would never ask Google. Yeah. And Sundar Pichai, you know, has has gotten a lot of criticism. Google is like great but evil at the same time. You know their their their do no evil model has has gone away. But he's argued that AI search is additive. And I do agree with that. I think that you you know, as we were saying with keywords, we were being trained. I know your colleague Christian, we're just talked about this a lot. We've been trained to get the output that we're looking for by working within the capabilities of the system. Ever since, we're seeing much more developed algorithms like Bert that understands natural language or mom. And this is like ten years ago now, um, it can understand the context. And so we can use that natural language to get the answers that we're looking for, which ultimately changes search behavior. Um, so we're seeing more and more of that. And I think it's accelerating in people's comfort in using natural language when they search, as opposed to depending on the way that we're used to with the keyword ease. Absolutely. Well, let's talk about the language itself. So you have often said the actual words that I use in the search, or the way I frame the question, can have a dramatic impact on the results that I get. And that might be because of like a bias that is embedded into that question, whether I'm whether I know it or not. So can you give me an example of two questions that might seem to have a similar intent, but have a hidden bias built into it? Absolutely. So two examples. The most classic example that I, that I lean on is the idea of coffee with, um, coffee has been known over time to have health benefits, but also you don't want too much of a good thing. And so if I go to a search and I frame my question is why is coffee good for me? I will get a certain type of search results if I get a if I have another question. When I say why is coffee unhealthy for me, I will see search results that reinforce the confirm that idea that it's bad for me. The same thing with take like electric cars. If I wanted to look at look for electric car and I say why is an electric car unsafe? I'm already phrasing that with a preconceived notion that electric cars are unsafe. So I will start to see results that tell me that they are. If I have something that's more unbiased and neutral. Uh, what is the the safety expectations of an electric car, then I might see more balanced results and I can do more research on my own. And then I can come to my own conclusions based on authoritative results. And so I don't think people are aware the extent to which they bring their own bias to the way that they search. And then also Google is by nature editorial like objectivity for many of these queries is not a thing. When I'm asking like, what is the best hoodie out there? There, there's no like this is the answer. It depends on who's asking the question and what matters to them. So it's really tricky to to navigate. It's interesting you gave we'll go back to the coffee example. Those were both you know pretty stark examples of a bias. Is coffee healthy or is coffee unhealthy? Um, but even if we tried to word a perfectly neutral question like what is the nutritional value of coffee or something like that? Like having been on the other side of this, trying to write surveys, even asking a question that doesn't have bias, like, how is your experience today? Or did you have a good experience today can lead to a different response, so it's shockingly difficult to ask a truly neutral question. Oh my goodness. Yeah. And that goes back to some of these psychological principles like the idea of anchoring. You know, we see it in numbers all the time where if I say a random number, you know, when you're price hunting, that will influence whether or not you think the price of something is too expensive or inexpensive? It doesn't make logical sense. It doesn't feel rational. And yet, you know, if we see the number, you know, 9999 just anywhere in any context, and then we go and buy something, that's for a couple hundred bucks. We're like, oh, a couple hundred bucks is nothing. But it's a it's all subtle. And the same thing I think happens with, um, with sentiment analysis. To your point about the good versus bad, like you. Not that we can be hypnotized. I don't think that's that's a like a weird psychological dark art. That's that's a pseudoscience. But I think you can seed, um, influences and anchor perceptions. Absolutely, absolutely. And I think this is going to be especially relevant, relevant within an LLM. Right. Because at least with Google, you ask a question, you get a list of links that you can click through and then hopefully decide for yourself which of these sources are relevant and what makes sense. But the MLM is doing the filtering for you, right? And it's giving you this response. And I'm curious if we are
00:16:17.500 — 00:20:54.020 being less critical of the responses that are coming back and taking that more as truth, as opposed to to thinking more critically about the content that we're seeing on an LLM. Um, are you like, I'd love to hear your perspective on that. I hate to be negative when it comes to the human race, but like we we are lazy and it takes a cognitive load to think critically. And I think there are many situations where people do just want someone to tell them the answer and not have to second guess it. And for a lot of people, I remember reading a research paper where it was people who are not familiar with generative AI results believe it's more magical and are more likely to trust the results. They don't know of hallucinations. They don't understand the black box nature of how it works. It's probabilistic under the hood. And so I think there's going to be audiences. There's going to be populations that do take whatever's put out at face value. You know, I feel bad I'm going to throw my mom under the bus. But I remember like, she was visiting me in D.C. and we were looking for places to park, and, um, the LM, she she asked it, she calls it her Chatty Cathy, and she's like, what? You know, where where can we park around the D.C. zoo? And it lists out all these places. And she's like, oh, we could go here, here or here. And I look on the map and I'm like, mom, like, these don't exist. This was this was a couple of years ago when the hallucinations were more rampant. But, um, but I was like, we we can't go there. These aren't real places. And so I think that there's a lot of education that needs to happen with the critical thinking element that needs to be employed for alums like I, you know, 5 to 10 years, it could be a completely different conversation. But right now, you mentioned, you know, a particular type of person maybe that is more skeptical. But I also think it is relevant to the subject matter as well. So the Dunning-Kruger effect, right. The the more I know about something more of an expert I am, the less confident I feel about it, because I know how much I don't know versus like maybe the less you know, the more confident you feel, right? I think this plays out in our ability to critique what comes back. So if I'm working with an LLM and I ask it to provide content for something, I am a deep subject matter expert in, I'm highly critical of what comes back because I'm like, that's not right. That's not right. That assumption is wrong. And I kind of dismiss it. But if I ask it about something I don't know very much about, like, I don't know how to how do you play cricket? It could be giving me so much BS and I'd be like, yep, that's how you play cricket, because I know nothing about that subject matter. It's such a good point. I'm I'm with you 100%. I think that is a really important distinction between like how we interact and and specific zones of genius, if you will. Right, right. So think about how this applies to a brand. And we have to shift from directly influencing our consumers, potential consumers to influencing the LLM to talk about us in the way that we want to be, the way we want to be talked about. Because if a consumer is not familiar with us, they may not challenge what the LLM is saying. Yeah, yeah, it's really hard. And that's why, like. Like, at our agency, for instance, we historically focused primarily on the website, but it's so much more omnichannel. Like there's this concept of like the narrative. And I know and locally you talk about a lot about sentiment. What are the channels that you own. What are the channels that you rent on? What are the channels that are earned and how can you influence that narrative? Because the LM will pull in from everywhere. And I think that's really scary for brands because it is even less of a lack of control. But the influence is there if you have and, you know, the whole industry of reputation management, like if you have a negative smear on your brand CEO, for instance, they did something like that can still pop up years later in these LMS. It's it's it's a really hard thing to, to protect yourself against, especially if it's a pattern. Right. If you get that same kind of negative feedback over and over again, or about slow service or poor quality items, it doesn't matter what you say on your website to to counter that, if user generated content and reviews are are contrary. Oh my goodness, I came from the review space originally. Actually when I got into SEO and reviews are such a
00:20:55.100 — 00:24:55.709 interesting point of social proof because like Google for instance, like you have Yelp who's super litigious and they are will get rid of reviews, um, even if they seem valid. That's been like for small business owners, that's always been a major challenge, let alone like major brand franchises, whereas Google is less discriminating. Anyone can kind of leave a review, and they haven't always been really great at removing review fraud. And it's a very difficult thing to police in the in the first place, not to defend them, but like that is a really difficult challenge. And yet your reviews fundamentally influence your brand. And they should. But it's not fair. Yes. Yeah, it is really tricky. Um, you mentioned this mix of all the different things that can influence your visibility and sentiment. Are you and your agency placing more or less importance on, on certain channels than, than you were maybe a year or two ago? And so one thing that that I think is really important is audience research and understanding your industry and understanding what channels matter to your industry. And that's a big part of the content strategy. So while we do see the, you know, some of the data coming out from a variety of different, you know, tools, and I remember to reference Christian again, Christian more, did you know you did an awesome kind of citation review of the types of citations that were showing up in LMS? Um, Reddit tends to be at the top right now. And, you know, in February of 2026. Um, YouTube is is towards the top because we know we're seeing more of an influx of multimodal content. So it's not just text, but it's videos, it's data. You know, even recently, like for B2B, there's a push of LinkedIn content, which makes sense, like the social networks. And so you do see kind of a lot of these generalized platforms rising to the top in my opinion. But but that is shifting what what we see today isn't necessarily going to be the case for tomorrow. Are you seeing the same stuff? We study it from a very specific lens, which is how are LMS answering questions about local businesses. So we're not studying how an LLM might answer the question, what is the best baseball team in history? Um, but we might be answering the question, you know, um, where's parking near the baseball diamond. Right. To your point. So within that context, what we're finding is that the brand's owned properties. The website and their local pages are the most heavily cited because they are the authority on their own information, their business hours, their menu items, ingredients, those sorts of things. Um, and then it's followed by directories and listing sites, which would be Google Business profiles. Um, MapQuest shockingly shows up a lot. And so those long tail things be happy. Right. And and what's what's playing out there, I think, is that all of these sites have highly structured information and it is a quality signal. So if they are all saying the same thing then the LM can trust this is accurate information about your brand. If it's all over the place, you you know they're going to be questioning that information and you're either not going to show up in the result or there's just really high risk that they're going to pick the wrong piece of information. Um, so that's primarily what we're saying. And then reviews of course, influencing questions like who is the best or is this good? Um, that sort of thing? Yeah. No, I've seen that, and I and I completely agree with it because I remember coming up in the industry originally, like, citations were so critical. Like your nap, your name, your address or your phone number has to be consistent everywhere. And then there was a period of like, okay, does it only matter for, you know, you know, a certain amount of sites. And I feel like we're seeing this resurgence of the importance of citations and, you know, the different, um, directories and listings, you name it, because
00:24:56.870 — 00:26:04.810 you have to protect against the hallucinations. Like the worst thing for a business is to go to ChatGPT and ask for their phone number and just get the wrong number. Even worse, like some sort of scammy spam number like that could do some serious reputational harm. So I do think that in the world of AI search and llms, it's so critical to have that consistency. And if you don't, you need to fix it like yesterday, because it could also be in the training data. Um, you know, we think of the common crawl, and and that stuff is even harder to fix. You have to wait till they go and update all the training data. So, you know, it's like it's it's kind of like a, a, a school test where you've already failed. And so now all you can do is like prepare for the next test next year. Um, do your extra credit. Exactly. Do some serious extra credit. Yeah. You know, I'm curious to you when we think about how LMS are responding. We didn't even talk about the concept of memory and personalization and the bias that might come through there as well. So
00:26:06.210 — 00:30:39.050 if I'm doing a search on Google and it knows nothing about me, for example, or a fresh, pure search, the bias might simply be, you know, in the in the context of the question itself. But let's say for weeks, um, the LLM has been learning about me and I consistently say when I go to a restaurant, I prefer that it's a local restaurant. I want organic ingredients. I want it to be locally sourced. I want a chaos menu, whatever it might be. So the next time I go to ask that question, all of those biases are already in the context of that response, right? I mean, is there a way to clear that out or remove the bias, or do we just need to think that that's the new status quo? I think, well, so we're used to that at local. Right. Because it's like there you there's an assumption, an expectation of a local bias in some capacity. For instance, like if we are looking for a grocery store recommendation, if you're telling me one's in Australia, like that's really not helpful for me. Um, so it's like we know in local, for the longest time, proximity mattered. I think that we will see more and more of that influence, too, because the thing is with prompts or engaging with an AI search platform. You have to provide context. To get the best answer. Like for instance, we're you know, I said earlier, if I'm looking for a restaurant and I have kids and I need something as kid friendly, it's a less cognitive load. It's a better experience for me if I can just say, what's the best restaurant for me? And it already knows I had kids and only recommends those and takes that into consideration. The question I think the issue comes in when obviously for a brand that wants to insert itself and is not a preferred, um, brand of the person they're trying to get in front of because they can't discover them, they're already giving them their preference. But like on the consumer side, it's frustrating too. I had an experience where I did connect my Google, um, AI mode to the personal intelligence, and now all of a sudden it can go into your emails and look at your emails for additional context and play that in and I asked for, like the best socks. And you could, you could, you know, change this to local as well. But I look for the best socks and it pulled. It's recommended two sock brands that I had purchased. And so you start to think, okay like is this a good experience. Is this what people actually want with memory where it knows I like this, it's confirming what I've already purchased. But in this context, do I want this recommendation? I think that's going to be the next frontier for for local brands to have to try to figure out is like, how do you bypass those preferences? You know, if it's if it's pure discovery, then you have to like kind of find an affinity to that audience. But if they already have their their preferences, can you bypass that? I don't know, what direction do you think we go? It's interesting because it's the premise of brand marketing, right? Build a brand that I am loyal to. And every time I'm in the convenience store, I'm going to pick Coke over Pepsi because it's it's a shortcut in my thinking. But this is the brand. They're probably the same. I don't drink either, by the way, but like, I think it's similar, right? If it's saying I have loyalty to a brand and the results are always going to bias to that loyalty, um, it could be amazing if I'm that brand, but if I'm not that brand. Yeah. How do you break through? Um, that is going to be, I think, a very big challenge for brands moving forward. And I think ultimately that's why marketing generally is becoming more holistic, where you can't just purely depend on organic search, um, or the old traditional ways of, of focusing just on SEO to show up in the results. I think that if you establish the, the, that brand visibility in all the other adjacent channels, whether that's social media or whether that's forums or whether that's other, you know, directories or just trades, that's going to be the best way for you to potentially infiltrate a topic that you really are not currently the preferred brand at. Like from to your point, do the winners keep on? Winning is a virtuous cycle for the market leaders? Or
00:30:40.130 — 00:31:30.480 are we going to see Google and OpenAI kind of adjust the way that they function to support the underdogs? The new the the new entries into the market? I think that's to be seen. I, I hope so. I hope so, yes, I do too. Well, it is to be seen, and I would love to invite you back to chat about it more as these things unfold. Um, I think the big takeaway here today was that, um, a couple, of course. But, you know, marketing needs to break down the silos and think differently and holistically about how to get discovered in this era, and it starts so much earlier than when a consumer puts their hands on the keyboard and it's in the mind of the consumer. And we really need to be tapping more into that great psychology research that, you know,
00:31:31.640 — 00:32:08.600 um, that we all dug into. So, um, Garrett, this is so great. Thank you so much for joining us and sharing your perspective. And, um, I would love to have you back again soon. I gotta come. I mean, how dare you not schedule a five hour podcast interview that everyone wants to listen to? But like, when we're talking, it's just you want to keep going because it is a fascinating time to be in our industry. And thank you so much for having me. This is so much fun. Thank you. And if you found this conversation useful, uh, follow the visibility brief. Subscribe wherever you listen, and we'll be back next time with more conversations about how discovery is changing and what it means for the marketers navigating it.
