Episode #229:Rebecca Colwell, SVP of Marketing at Yext, explains why AI search adoption is surging and how it reshapes visibility. She shares research on usage and click behavior. She outlines three strategic shifts for marketers and shows how trust and accurate brand facts drive citations.

“If you’re not visible in that first moment, you miss the entire opportunity. The goal isn’t just to rank for a keyword, it’s to be cited and trusted by the model so you appear in the answer.”– Rebecca Colwell

Marketers are entering a new era of visibility, where AI-generated answers replace search results. Rebecca shares Yext’s findings on how LLMs are changing buyer journeys and what steps leaders can take to earn trust, measure citations, and ensure brands are visible when AI provides the answer.

Follow Rebecca Colwell on LinkedIn 

Follow host Steve MacDonald on LinkedIn

[Transcript]

[00:00:05] Steve MacDonald: Welcome everyone to the C-Suite Sales and Marketing Perspectives podcast. I’m Steve MacDonald, your host, and today we have a super interesting conversation, a guest on with Rebecca Colwell. Now Rebecca, you’re the SVP of Marketing at Yext, and I’m gonna have you explain a little bit what you do at Yext, but.

[00:00:21] Steve MacDonald: You have been a marketer to marketers almost your entire career. So you understand us in ways innately that maybe sometimes we don’t even understand ourselves, right? So I love the direction that we’re going here because we’re gonna be talking today about how the world is shifting on us in a number of different ways, and the impact of AI and the LLMs is a huge part of that.

[00:00:48] Steve MacDonald: But how we’re having to reinvent our ABM go-to-market based on this new era of AI and what we need to be thinking about, what we need to be doing, and the fact that the urgency to do that is actually here right now today. I would love to have you, if you could, expand a little bit more on your background.

[00:01:09] Steve MacDonald: You live in Portland, Oregon. You’re a huge fan, I can see, of greenery and plants behind you.

[00:01:17] Rebecca Colwell: Yes. Yes.

[00:01:17] Steve MacDonald: Those are real. Mine aren’t in the background. I wish I could take credit for that, but if you could just maybe explain a little bit more about who you are and then we’ll dig right in.

[00:01:28] Rebecca Colwell: Sure, sure, and thank you so much for having me. It’s wonderful to be here. Yes, I have been in marketing for the entirety of my career and in tech. I got into tech very early. It was still a pretty nascent industry at the time, but what I absolutely love about the space is that it is constantly evolving and changing, and that’s really fun—you know, really fun and interesting to grow.

[00:01:49] Rebecca Colwell: And I find that marketers that I work with as well, like tech marketers in particular, are constantly evolving. And yeah, I’m excited today to talk about some of the evolutions because when I started in marketing, email marketing was still becoming something we did in the workplace.

[00:02:05] Rebecca Colwell: Like my first email marketing campaign I did through a mail merge, which I don’t even think a marketer today joining would even understand. Like, what the heck is a mail merge? But, you know, I saw the introduction of social media into our marketing toolkit. I saw Google become, you know, Google eats everything in the marketing department, and now this shift to AI and LLM visibility.

[00:02:27] Rebecca Colwell: And so it’s really fun, and I love working with marketers because I think they’re always on the forefront of change inside their organizations.

[00:02:35] Steve MacDonald: Well, let’s get right at it because I’d love to just ask you if you could kind of paint the picture for us—why an AI search is dominating and starting to take over at a rapid pace.

[00:02:48] Steve MacDonald: Google, and then therefore, not as much SEO strategies, but LLM strategies, right? In reinventing how we create visibility for our companies. And I think you’ve done a study recently at Yext that I would love to have you share a little bit from what you found out.

[00:03:05] Rebecca Colwell: Absolutely. So yes, at Yext we did a study of over 2,200 people across the U.S. and Europe. And we asked them about how they search for answers across traditional search platforms like Google, as well as these emerging AI search platforms. And what we found is that across the board, 75% of them are using AI to connect searches more than they were a year ago. So very, very rapid adoption.

[00:03:31] Rebecca Colwell: And if you map that against the time it took other technology platforms to gain prominence, ChatGPT has reached 100 million daily active users in two months. It took Instagram two and a half years to get to that threshold, took TikTok nine months. So even like, if you think about the next fastest thing, ChatGPT did it in two months, TikTok did it in nine, Facebook took four and a half years. So it’s just the pace of acceleration in adoption is unlike anything we’ve seen before in marketing.

[00:04:04] Rebecca Colwell: And I think it’s because as consumers, we’re a little frustrated with the Google experience. I mean, when I first started, I don’t know if you remember like Yahoo and directories, where you’d get online and you’d have to scroll through different folders of information and try to find what you were looking for.

[00:04:22] Rebecca Colwell: Google flipped all of that on its head, right? And said, hey, there’s this little box. Enter a keyword, find what you want. And we’ve all gotten really used to that. But the challenge is that when you search on Google, you don’t get an answer. You get ten blue links where you need to go and do additional research to get the actual answer to your question.

[00:04:43] Rebecca Colwell: And an LLM like ChatGPT doesn’t give you a link. It gives you an answer, and it’s synthesized—it has gone through and read through a variety of trusted data sources and then come back and said, this is the answer to your question, not more work that you need to do to discover what you’re looking for.

[00:05:00] Rebecca Colwell: So I think the rapid adoption of LLMs is tied to the fact that we are getting what we’re looking for, which is an answer to a question.

[00:05:08] Steve MacDonald: Yeah. It’s interesting because I had never had frustration going to Google like I have now. So I’m like, don’t just give me more work. Give me the answer.

[00:05:20] Steve MacDonald: And I know that Google has Gemini, right, and the Gemini AI search, and the answer can come up at the top, but there’s also a reason why they don’t always want to put the answer at the top. And if you could explain that a little bit, because it’s just like, well, why doesn’t Google just adopt their—they have an LLM, why don’t they just change their model, right? But there’s a reason behind that.

[00:05:45] Rebecca Colwell: Oh yeah, absolutely. It’s their business model, right? It’s the ad model. So the majority of marketing departments invest most of their ad spend in Google—Google Ads, right? And so the way it works is if I want to show up in a search for orthodontists near me in Portland, I can invest my time and energy in trying to show up organically.

[00:06:05] Rebecca Colwell: I try to get into that map pack, and if I can’t do that, I’m gonna pay for an ad that, you know, answers the question Invisalign near me or orthodontist near me or something like that. So Google has a vested interest in collecting pennies or dollars for all of those ads, and that only happens on a click, right?

[00:06:23] Rebecca Colwell: So if I, as a consumer, come and answer a question and then don’t click a link, Google doesn’t get paid for surfacing the answer. So as a marketer, what you’re likely to see with the AI overviews—so you’ll probably notice there’s an AI mode now on Google—when that AI overview is present, click-through rates drop from 16% to 8%.

[00:06:46] Rebecca Colwell: So, they’re cut in half, and this is concerning, you know, for Google and for marketers. So what we’ve observed at Yext in our research is that those AI overviews are being shown for searches that Google doesn’t think it can monetize as much, right? So I’ll give you an example. If I go to Google and say emergency vet near me,

[00:07:10] Rebecca Colwell: it is gonna monetize that search. It’s gonna have paid ads at the top for the closest emergency animal hospital, the map pack, organic search results, and ads because it knows I have very high intent to take my pet to the vet. If I ask a question like, my dog is acting lethargic and I think it got into something and ate this, do I need to take it to the vet?

[00:07:35] Rebecca Colwell: There isn’t as much intent behind that question. It’s more of research, and so it’s likely to show an AI overview to answer that question because it wouldn’t have gotten the click anyway. So yeah, what we’re seeing is that impressions are down a lot, but the high-intent clicks have not dropped as much as we would’ve expected because of AI overviews.

[00:07:55] Steve MacDonald: So the bottom line is that more and more and more people are using LLMs to search. So therefore, we can’t just stay with our tried and true SEO strategy. We have to evolve or we’re gonna devolve, right? We’re gonna be left behind, is the bottom line. So I’d love to have your kind of, just your thoughts on if a lot of us really haven’t started

[00:08:25] Steve MacDonald: an LLM optimization strategy, like how do we even think about that, right? What do we need to do as marketers to start to shift and do it quickly?

[00:08:36] Rebecca Colwell: Yes. Yeah, that’s a great question. So there are three big changes that we’ve observed.

[00:08:42] Rebecca Colwell: The first is that, or the first big change I would recommend is, number one, you have to think beyond Google. So it’s still a really, really important platform, but expand your field of vision beyond that and think about your visibility and sentiment on platforms like ChatGPT, Perplexity, Claude, Grok, and social media platforms like TikTok, you know, and then forums like Reddit.

[00:09:06] Rebecca Colwell: Search is no longer a singular destination, and I think we might be the last generation that uses Google as a verb. The second shift is that we need to shift from thinking just about SEO, which is links and keywords, to influencing answers. So there are a lot of different terms—GEO, AEO, EOI, like AEOI (Answer Engine Optimization)—but who knows what it’ll be like.

[00:09:24] Rebecca Colwell: But here, the goal isn’t to rank for a keyword; it’s to be cited and trusted by the model so that you show up in the answer and you help form that buyer’s first impression. And so, happy to share some tips for how to do that, but the idea is you want to be visible in that answer.

[00:09:48] Rebecca Colwell: And then the third shift is that the funnel, as we know it, is changing dramatically. I think we think get the click, come to the website, capture the lead, nurture them through email marketing, nurture them through, you know, other tactics, you know, convert the lead.

[00:10:07] Rebecca Colwell: Increasingly, that entire education journey can happen in an LLM conversation, and AI could be having that entire conversation on your behalf. It’s still sourced or cited by information that you provide, but that entire funnel could be entirely compressed. And so if you’re not visible in that first moment, you can miss out on the entire opportunity.

[00:10:33] Steve MacDonald: So I’d love to ask if I am taking away the right takeaway from this. For one, we have to start creating content that will be cited inside of these LLMs. Yet again, an entirely new reason that we need to be creating good content, right? Already. On top of the fact that, like Forrester said, hey, ABM marketing is all foundational in content, right?

[00:11:04] Steve MacDonald: It all starts with content. So we’ve known that, right? But sometimes talking to the rest of the C-suite and talking about content, when over 60% of content that marketing creates, sales never uses. So, you know, like this content conversation needs to be reframed inside of our own companies, right? And so you’re helping with what is foundational now—LLMs.

[00:11:29] Steve MacDonald: But here’s an impact, I think, from this LLM as well, but it’s also because we’re losing trust with our buyers at the same time. And you and I talked about this before—like, go to the most recent LinkedIn Edelman study, and it talks about how the 32,000, the Trust Barometer study, 32,000 respondents—61% think that we’re being intentionally misleading, right?

[00:11:54] Steve MacDonald: And so that’s fueling this self-service B2B buyer journey where they don’t want to talk to us until the very end, and then they already have their top two or three selections in mind. So content is becoming a much, much more important part of the sales process because people are looking outside of us in the direct conversation to make their decisions, right?

[00:12:19] Steve MacDonald: And so all paths are pointing back to—we gotta be a lot better at creating content than we ever have been, even though we’ve always known it’s been ultimately important. 

[00:12:30] Rebecca Colwell: Yes. Yeah.

[00:12:30] Steve MacDonald: That’s my takeaway from what you’re saying. Is that a good one? Or?

[00:12:34] Rebecca Colwell: It is. Your point about trust is a really, really important one.

[00:12:38] Rebecca Colwell: So if you wanna show up an answer to a question on an LLM, the LLM has to trust that it has accurate, recent, consistent information about your brand. It does not wanna hallucinate. It does not wanna give the wrong answer, right? Because the switching cost—let’s say you get a wrong answer on ChatGPT—you could jump to Claude or Perplexity or any competing platform in a second, right?

[00:13:00] Rebecca Colwell: So they want to give you an accurate answer. So, um, it’s all founded on trust. And so the question is, how do you deliver content to an LLM that it trusts? And it comes down to a couple of things. The number one most important thing is that your brand information is consistent. So let’s talk about something really straightforward, like business hours.

[00:13:24] Rebecca Colwell: Let’s say you’re a coffee shop, right? And your business hours are likely listed on your Google Business Profile, Apple Maps, Yelp, Yellow Pages. There are hundreds of long-tail sites that also have hours listed. There’s probably a coffee shop–specific one. So when somebody says, “What’s the best coffee shop near me that’s open right now?”

[00:13:47] Rebecca Colwell: And ChatGPT goes out to answer that question, it’s gonna search across all of these sites. If it sees the same answer everywhere, it’s going to trust that that’s the right answer, and then it’s gonna say, “Great, you open at 6:00 AM.” If it sees different information across three different sites, it’s not gonna trust you.

[00:14:05] Rebecca Colwell: It’s not gonna cite you in that answer because it’s not confident, right? So, when we talk about content, I think it’s also shifting away from how we think about content in forms of long-form blog posts or white papers and things like that, but down to the atomic unit of brand facts about your organization and ensuring that information is consistent and accurate, invisible in all of the places that an LLM is searching.

[00:14:35] Steve MacDonald: Because we could almost look at that as like, those are our About pages and things like that, and you’re like, yeah, okay, it’s there. But we don’t spend a lot of time on those. I mean, we write a paragraph about us and we’re like, that’s pretty good.

[00:14:46] Steve MacDonald: Okay, right. We focus on other areas of content. And you’re not saying that blog posts and white papers aren’t important, right? But what you’re saying is, start paying attention to the things that are important, that are foundational to you in a business, right?

[00:15:03] Rebecca Colwell: It’s helpful to understand how an LLM answers a question.

[00:15:07] Rebecca Colwell: So Yext research just did a study we published this week, and we analyzed 6 million citations. A citation is like the little source of information that AI references to answer questions. So, by the way, you can do this experiment right now. Go ask ChatGPT or your favorite LLM a question, and then at the bottom it will have a little link that says “Sources.”

[00:15:31] Rebecca Colwell: You can click on that and see all of the things that it’s cited. You can also ask it, “How did you answer this question?” So, I’ll give you a recent example. I asked, “Who is the best orthodontist in Portland, Oregon, where I live, that accepts Cigna insurance and offers Invisalign treatment?” So if we break this question down, when I looked at the citations, the first part of that question was “best,” who is the best orthodontist?

[00:15:58] Rebecca Colwell: So, to answer that part of the question, ChatGPT looked at review sites. So it looked at industry-specific review sites. It looked at Google ratings, but it was saying, “Let’s crowdsource to find out what other patients have said about the best orthodontist in Portland.” To answer the question “Is this an orthodontist?”

[00:16:16] Rebecca Colwell: It looked at Google categories and categories on, you know, medical directories and listing sites. “Portland, Oregon” referenced maps, Google Maps, Apple Maps, those sorts of things. And then to answer the question “Does it accept Cigna insurance? Does it offer Invisalign treatment options?”

[00:16:34] Rebecca Colwell: It looked to websites and local pages, so that brand’s website. And what we found is that 40% of the sources that were cited were brands’ own websites.

[00:16:49] Steve MacDonald: Ah, now when you say websites, you mean multiple sites or pages on their website, or—

[00:16:56] Rebecca Colwell: Correct. Like detailed pages on their websites, right? So let’s say you are a multi-location brand. You probably have a specific page for every single one of your locations because maybe you have different services. If you’re a restaurant, maybe you have different menu items, business hours. But if you are a Nike, you have a dedicated page for every product that you sell—the colors of the shoes, the sizes, the features, are they great for running or whatnot.

[00:17:23] Rebecca Colwell: So the great news is, if 40% of the sources that an LLM is referencing come from your website, that means you have control over that story, right? So you just have to give it the information, right? Like, you have to have those pages available with as much accurate detail as you can. So that, if my local orthodontist said that they accept Cigna but they didn’t call out that they offer Invisalign, they may not show up in that answer.

[00:17:50] Rebecca Colwell: So it’s like detail, detail, detail. Get down to the nitty-gritty facts and structure them in a way that LLMs like, which is short, sweet snippets or chunks. You’ve heard the word chunking, right? And in natural language. So it will then take that and write an answer that sounds natural. But try to avoid keyword stuffing and some of the things we’ve done in the past for SEO, and just write a natural-sounding answer.

[00:18:18] Steve MacDonald: I’ve got a big question. So you gave a lot of great examples of kind of the world of B2C, right? How does this apply in the world of B2B, right? Yext is a B2B marketer, so you gotta be thinking about, right, for the Yext services, how do I optimize for LLMs too? And the audience here is all B2B, you know, go-to-market companies.

[00:18:45] Steve MacDonald: How does this translate into the B2B world and what we need to be doing? 

[00:18:50] Rebecca Colwell: It’s a great question. So, let’s say someone wants to say, “How does Yext compare to an alternative solution?” Right? It is going to also cite information sources. So it may look to our website to see if we have a compare page. It’s gonna look at product descriptions, feature descriptions, those sorts of things.

[00:19:08] Rebecca Colwell: It’s gonna go to competitor sites and look and do the same analysis. It’ll look at review sites like G2. And so it’s that same fundamental concept of providing trusted information that LLMs can access, and ensuring that trusted information is structured in a way that the LLM can read.

[00:19:29] Rebecca Colwell: So one of the challenges that a lot of brands have is that they create really beautiful pages designed for people, but they’re really, really big and heavy and hard for an LLM to digest. And so in the same way that we have thought about metadata to feed Google and search engines, we need to think about schema markup on our pages that tell LLMs, “This section is describing my products. This section is describing services or pricing or availability in a market.” And that increases the likelihood that you get cited because they can find you and understand you.

[00:20:07] Steve MacDonald: It’s, you know, the way I think about that is we’ve all used like ChatGPT before, and you can upload documents, right?

[00:20:14] Steve MacDonald: You can upload documents as source documents or context documents. But ask ChatGPT itself, is that as good as having that data—a Word doc or something you uploaded—is it as good as having it in a vectorized database where it’s actually been organized and the information’s been prioritized correctly?

[00:20:36] Steve MacDonald: And ChatGPT people will tell you themselves, “Oh yeah, we can skip things if you just upload a document. Like, we can not see things that are important.” And that’s exactly what you’re talking about: set your website up as a database for these LLMs so that they can easily read and understand that database that you’ve put up about your company.

[00:20:58] Steve MacDonald: And that’s like—you talked a lot about data as an all-important underlying strategy here. Like, what aspects of it that you haven’t already talked about do you wanna make sure that we understand about kind of our data strategy behind all of this?

[00:21:15] Rebecca Colwell: It’s a great question. And it’s really interesting because, as marketers, we love data.

[00:21:21] Rebecca Colwell: It just—typically it is performance data, right? So it’s time to get a little introspective and think about our brand data. So, I’ve already touched on the importance of having really, really accurate, consistent information everywhere. But the other thing is the recency and frequency of your updates.

[00:21:37] Rebecca Colwell: This has already been proven on Google, but we know that online reviews impact how you rank on Google. So the more recent reviews you have, that sends a signal to Google that people are still coming to your business and engaging. That same signal is picked up by LLMs to say, “Oh, you have a recent review, a recent post on social media. You’ve recently updated the content on your page.”

[00:21:58] Rebecca Colwell: So I would just say, make sure to not let your information get stale either. This isn’t a one-and-done exercise. It is a constant exercise in keeping that information accurate. And the other thing I would stress—this is really, really important—when we think about your data, if we were to look at a restaurant, for example, a restaurant chain, I don’t know what your favorite one is, but we could pick one for you.

[00:22:25] Rebecca Colwell: You’ve got locations, hours, menu items, ingredients on those menu items, open jobs, specials, you know, all of that stuff, right? It’s not just providing that information, it’s showing the relationships between those items. So you might say, “This location offers these menu items,” right? You might not have the full menu everywhere.

[00:22:49] Rebecca Colwell: And so it’s also just helping LLMs understand how all of your information connects. We call them entities—how they all connect to each other to tell a bigger story. And LLMs are great at following those paths and understanding how to piece it all together.

[00:23:04] Steve MacDonald: So, by the way, to answer your question, I have four teenagers, and my favorite restaurant is Chili’s.

[00:23:09] Steve MacDonald: Their “Three for Me” meals—because I can go in and buy my family a dinner without having to spend 150 bucks—and like, oh my gosh. So aside from that, I digress. So, you know, this is so interesting, and I want to give an opportunity for you to answer the question: What is the most important question that I haven’t asked, right?

[00:23:34] Steve MacDonald: Because you’re the expert in this, I want to give you space here to say, okay, if we’re starting this LLM optimization strategy, what else do we need to know? What else do we need to be thinking about?

[00:23:48] Rebecca Colwell: I think the very first thing every marketer needs to understand is how they are showing up today.

[00:23:54] Rebecca Colwell: So, is my brand being cited? What is the sentiment of that? Am I being cited everywhere? Am I being cited across every model for all of the queries or types of conversations that I wanna show up in? And answering that question has been really challenging for a lot of marketers. Google gives you so much information, right?

[00:24:14] Rebecca Colwell: We have certifications in how to understand Google Analytics. And so, I think a lot of us get uncomfortable feeling like we’re flying blind, but there are tools out there. If you are a location-based business, we have a free tool called Scout. So you can go to the Yext website, check it out, but what you can do is enter the address for your business and we will run an analysis that tells you how you’re ranking in Google, just to give you a baseline, and then how you’re showing up in AI search across the different platforms.

[00:24:40] Rebecca Colwell: And then we will show you your sentiment. So once you have that baseline of where you’re at for every single one of your locations, if you’re a multi-location brand, from there you can start to identify what you need to do to increase your visibility. And to do that, you need to understand a little bit about what we discussed earlier around what sources of information AI trusts.

[00:25:05] Rebecca Colwell: It’s ensuring that you’ve got a really robust website and landing pages with accurate, rich information about your products and services. You’ve got your listings optimized everywhere. You are looking at reviews for your business. You’ve got a robust social media presence. Stay on top of all of that, and then ensure that the information that you have on all those sites is accurate and consistent.

[00:25:31] Rebecca Colwell: So those are the three things that I would do: measure, publish to those trusted sources, and then structure your data to be accurate across those trusted sources. And that’ll get you really, really far in improving your visibility in AI search.

[00:25:46] Steve MacDonald: I love it. You know, our poor human brains can only take about—and remember—two or three things at a time. So if there was one most important takeaway that you wanted the CMOs, the CROs, the CEOs, B2B go-to-market, ABM go-to-market companies that are tuning in to this conversation, what’s the one thing that you would want them to take away?

[00:26:11] Rebecca Colwell: Yes, there’s a lot of misinformation out there right now about how LLMs work.

[00:26:16] Rebecca Colwell: So, recently the conversation was all about Reddit, right? Like, “Oh my God, this is the number one most cited source. I need to have my Reddit strategy in place.” And then OpenAI came out a week or two ago and debunked that and said, “No, not really, that’s not what we’re looking to.” So I would say,

[00:26:32] Rebecca Colwell: Try to avoid the hype around what’s happening. And just remember that LLMs want to provide the right answer. And to do that, they are looking for trusted, consistent information about your brand. So focus on getting the foundations in place and the rest will follow. So structure your brand information, publish it out to those trusted sources, measure how you’re doing.

[00:26:57] Rebecca Colwell: It’s a simple one-two-three process. And if the next shiny object comes out in the news—“Oh, it’s all about Wikipedia,” or “It’s all about TikTok,” or whatever—just, like, take it all with a grain of salt because it’s probably not likely to be the right answer.

[00:27:13] Steve MacDonald: Well, I have to say that the mission of this podcast is to bring different perspectives from CMOs and CROs and CEOs together from all of our years of experience, insights, and perspectives that are unique and different, that make us think about what we need to be doing.

[00:27:33] Steve MacDonald: Because at the executive level, we need to be thinking not only fundamentally about, like, for you, our marketing strategy, but we need to be thinking about our overall strategy—about how we lay the foundation for success for our business. And I think if we can all do that, that’s the rising tide that lifts all boats. And I thought you were an amazing part of that rising tide today. So I just wanted to say thank you for coming on, and I think we probably need a follow-up because there’s so much to this strategy that we have to get right. And so I’d love to have you back on in the near future as well, if you’re up for that.

[00:28:12] Rebecca Colwell: Absolutely. I’m sure we’ll have fresh new research to publish, and I’d be happy to share that.

[00:28:17] Steve MacDonald: Great. Well, is it appropriate that if people have questions, which I know they will, that I give a link out to your LinkedIn profile so people could reach out to you?

[00:28:25] Rebecca Colwell: Absolutely. Yes, please hit me up on LinkedIn.

[00:28:28] Rebecca Colwell: Send me a message, and you can also email me. So happy to connect.

[00:28:32] Steve MacDonald: Alright, well, Rebecca, thank you for coming on and sharing all this knowledge. Really appreciate it.

[00:28:38] Rebecca Colwell: It was a wonderful chat. Thank you, Steve.