In this episode of The Visibility Brief, Yext SVP of Marketing Rebecca Colwell sits down with Chief Data Officer Christian Ward to unpack something many marketing leaders are quietly feeling: the way we've measured visibility for the last 20 years isn't working the way it used to.
Impressions spike and dip without reason. Rankings change by the second. Clicks are disappearing. And as AImodels fragment search across multiple platforms, the idea of a single, stable "position" is starting to disappear.
So if rank isn't reliable, and clicks don't tell the full story, how should marketers measure performance in an AI-driven world?
Drawing on the latest Yext Research and a new competitive scoring model inspired by Elo rankings, Rebecca and Christian explore how visibility should be measured going forward — and why the brands that win will be the ones that understand competition, context, and consistency, not just position.
The episode breaks down:
Why AI literacy is becoming a competitive advantage
What model fragmentation means for brand visibility
Why being "chosen" matters more than ranking#1
How Elo-style competitive scoring reframes measurement
Why having more local competitors raises the stakes for your data
Why structured, consistent data remains foundational
If you're a marketing leader trying to make sense of fluctuating rankings and disappearing clicks, this episode will help you rethink visibility — and build a more resilient way to measure success in a world where AI sits between you and your customer.
Episode Links
Transcript
00:00:00.080 — 00:00:30.600 All right, marketers, buckle up. Christian Ward is joining us today and we have a lot of ground to cover. We start with the rising bar for AI literacy and the growing momentum behind Claude. Then we move into what model fragmentation means for brand visibility. And of course, Christian saves the best for last. A preview of the latest and greatest from Yext research. Our Elo study, which gives marketers a far more accurate and dynamic way to measure search rankings. There's a lot to talk about, so let's get into it.
00:00:38.320 — 00:25:35.830 Hi, Christian. It's great to see you. Hey, Rebecca. How are you? I'm great. It's been a couple of weeks since we've had you on the podcast. Awesome to have you back. Thank you. Looking forward to it. Yeah.
Do you have anything fun planned for spring break? You know, I think we're mostly just taking it easy. Uh, all the kids will always be home for a bit. And I know that we're going to travel a little inside the state to see family and friends, but nothing crazy. Oh, amazing. You deserve the break.
I wanted to dive in today to talk about AI literacy because at your prompting, I have significantly up leveled my own literacy. So you filled this reputation inside the access for constantly pushing people to learn more about AI, not just, you know, not just read about it, but actually use it. I'm really curious, um, why does this matter so much to you personally? Like what's driving that conviction? So generally speaking, I'm relatively curious by nature. And in in that process, I've seen such benefit in people adopting some of these. It to me it is it's eye opening. But also, again, I'll just be as a as a dad. When you see your child sort of break through in something, it is enormously fulfilling. And one of the fun things that doesn't happen a lot in our adult lives. Right? And so this is one of those situations where people that are starting to try and adopt these things. I can see them on my team or at the company or even in, you know, other people outside of work. When they adopt them, they're realizing that they're getting time back. And time to me is just the most precious thing. I think it's the only real asset that we should measure our lives in. And to me, it really does do that.
Now, it doesn't mean you don't necessarily fill it with other things, but it really can speed up your ability. And I think for most people today, if you're not thinking about incorporating these today, they will be part of your career trajectory. And so I think one of our duties at Yext and in our lives is helping other people understand the mechanisms by which they could advance their career or their goals. And so I have found these tools to be extraordinarily good at that. And I think they're getting better every day. And so that's really why I'm pushing people. It's not I there's no secret agenda. I just look at it and I think it's really a benefit. And I think I think it's worth the time. Do you think there's a minimum bar for literacy in today's world. So interestingly, this is one of those scenarios where as people are learning it, the actual usage is getting easier too. So so you're almost coming at the same point from different angles. So the people that started doing this two years ago had to put enormous scaffolding around it to just get it to do basic things. Now the scaffolding is coming down where the tools are better on top of it. So I think that's a moving bar is basically what I would say in terms of what the minimum literacy is. But when I look at it, my my statements, most people is starting it and using it in the chat experience. That's a great onboarding. But it's it's so limited to what it really can do. If you start getting more into the cloud code world or you get into Codex, even code work was an eye opener for my team because again, I think they were very nervous about working in terminal and all these things that generally speaking, a lot of these people don't do. It's more of an engineering thing. Co work was sort of opened their eyes to, oh wait a second, This is very different than just chatting with it. It's been exceptional.
So I, I decided to go down the cloud path at your urging. Um, I did a little live coding over the weekend. It was amazing. I never I mean, I haven't coded since, like the HTML days in the early, the early knots. What struck me was how easy it was for me to make the shift from ChatGPT. I think it took about 30s, and I was concerned because I had all of these, I don't know, um, custom instructions set up, and I thought it would it would take the system a really long time to get to know me, and it didn't. So basically the switching cost was zero.
Yeah. No, no, no barrier to entry. Um, what is the implication of that for the LMS and the I don't know, the arms race that they're in right now. Yes. Um, so I've often spoken about something called the technology acceptance model. And it's, it's, it's I believe it's from like 1980. It's, it's old. But Fred Davis, a famous research scientist, wrote the paper and it's evolved a lot. But the reason why I bring it up is he sort of adopted the theory of Klan behavior and said, look, humans are going to adopt stuff like technology for really two reasons how utilities, how much utility does it have and how hard is it to use. And it was it's really those two things now. It's it's evolved. There's privacy concerns. There's all these other things at work here. But what you described, I've been pointing out for really a couple of years now is I think the the perceived ease of use of this technology or the cognitive burden is effectively zero. Like, I could argue, getting a toaster, like a basic toaster, a basic toaster is very easy to use. But we went and bought one of these Breville things that air fries, and the booklet is this big. So to adopt that is a lot of work at the family. Not ridiculous, but it does a lot. This is one of the first technologies and by it all other technologies You can adopt almost anything just by speaking. And so I think for these companies, as I've said, I think trust is the currency of AI. If you don't trust it, you're not going to use it. And I think the switching costs are effectively zero. And so you could say the memories and this stuff. But there was even a paper that just came out last week from research scientists saying, you know, we're putting a lot of this prompting and all this stuff. It really doesn't need a lot of that. It's actually quite good at this point. If you just talk to it right from the get go. It's like I said, I think the line of its benefit and how easy it is to use is advancing very rapidly toward us rather than us chasing it. And that's a big change. That is not the classic model for any technology adoption. It's much more that it's adopting us in that that's different, that we have not experienced that as humanity. This is very new, new, new land. It is very different.
I'm curious Because we've seen such a rapid shift from ChatGPT to Claude. Maybe it pivots again next week to something else, or next month. Is this something that marketers should be monitoring and like? Does it really matter which model wins?
Yeah, it's a great question because as you know, we're doing a ton of research at UX research on which models who's doing what, what sources they using. What I would say is this. You've seen a lot of rise with people talking about Claude versus ChatGPT versus the others. Um, they're all kind of in slightly different niches right now as to what you use them for. For example, I still don't think anybody beats nano banana, which is so funny to say, but Google's Gemini model for creating images. Now ChatGPT has one, but I don't think it's quite for the same level. And it's sort of. But that's not to say like the arms race. As you point out next week, there could be a new one. I actually think Nano Banana two just came out. And so you go through this process where they keep one upping each other, which is kind of fun, but it's exhausting. And I would tell everyone, we're all exhausted. We we all feel it. It's sort of this, this, this burden to stay informed. I think for marketers, probably the most important thing is getting their teams to understand that they should be using these tools in their marketing process. Um, the part about them showing up that really right now there's not as much consumer use, let's say, of Claude, far more consumer use of ChatGPT. But that is a fraction of Gemini, right? Because Google can turn any search into Gemini. So you're starting to see there's I think Gemini and ChatGPT for most marketers is where they absolutely should focus. Perplexity is much smaller, but it's an interesting model. Claude, on the other hand, as you well know, I think it's really for the marketer as their internal tool is absolutely phenomenal. So I think there's these lines right now, but I can't promise I actually think these lines are going to continue to shift in the sand as to which one is more important or less important.
The only one I can definitively say absolutely focus on is Gemini because they already have billions of users on search, so they've got this ingrained ability to turn it on. But yes, I think marketing I think it's not just about how you're showing up and what your brand is saying and all about your business. It's also the actual adoption of it by your teams. So when we think about visibility, I'm monitoring my visibility across all of these different platforms. Um, let's get into the visibility question more directly, because there's a conceptual shift that's happening that I think marketers are still trying to to process. Yes. In the future, it's not about rank. It's about being chosen. And that is that that one is still hitting me really hard. So I'd love if you could help us unpack a little bit of what that means. Well, if we if we go to where we came from, which is I put in a keyword or a query and I get a list of all these different elements. We've all lived in this world where while the higher you rank, the more likely you're going to get the click, and that. That's fine. That we've all lived under this regime of of of marketing for, for 20 years. And where you were in that rank really determines. So if you're on page two, that's really tough. And you really did all these things and it was search engine optimization and it became a massive industry. I don't think that there's a lot of correlation between that process and being chosen, but the difference here is you now have different models, you have different algorithms, you have different source citation sources. You have memory attached to every single AI. And so it knows what you're into versus what I'm into. It knows all of these things. So now it's a matter of the AI is almost sitting in front of the human and choosing which of those are going to be shown. So it's a whole nother layer that you have to get through as a marketer and as a brand. And so the best way to do that is to make sure that anything a human might be asking of their AI, you've provided that information to the AI so that you can be part of that dialog of I'm being chosen. It's going to be a lot less about sort of just this one algorithm, and a lot more about putting much more data and knowledge about your business out there. And the best way to think about this is I did a a conversation in front of a client, and I was showing them picking a financial advisor. And I always joke about this one because it really blew my mind, which was the it came back and says, you got to work with Charles Schwab like you've got to. And I was like, really like that. And it was so like, this is it. This is the one. And I asked why. And if we went down this long list of, well, they have this financial planning and they'll do estate and they'll do all this stuff. And I said, yeah, but everybody like Merrill has that. Everybody have these. You're telling me things that don't, don't really make sense to to why you're so vehement. And it finally, after probing and probing, I ask everyone you should try this. Keep probing it and say, seriously, is there any other reason why you're recommending this? And it comes back and says, well, I know your daughter goes to the University of Florida and that you also want the University of Florida. Here's my mug. And they basically go, um, all three of the principals of the Schwab office near you. They all went to Florida, and they're big Gator boosters. And I thought that might work out well. And I'm like, Holy cow. So now this is what we would call cross context memory.
So it knew that in a different context, but it applied it. And we're not it's not forthright about how it applies it. I showed this to a client the other day on on shipping and logistics. Same thing. It applies it, but it doesn't always tell you it's applying. So the reason why I bring this up is that means that by those three principles, putting their affiliation with the University of Florida on their website, that was what got them chosen. That's a totally different thing. And I look, I think it's really going to make marketers have to take a step back as to what their strategy is. It's not just content, it's that you've got to get the honest knowledge of everything that might come up. And again, I hate to say this, but for healthcare, for professionals, Realtors, what sports team, what country club, what all that stuff does matter and you're going to want to get that data out there. Well, what what is so concerning about this is that you, as the consumer, didn't have visibility into the factors that were driving that recommendation until you asked over and over and over. I as a marketer, I'm certainly not going to have that visibility. You know, getting chosen has historically been a bottom of the funnel concept, right? The market where I generate leads, I nurture, I have visibility, I can tweak, I can test, that's all gone. Yeah. Yeah. It's that is such a good point. And I drilling in a little bit because when you think about as a marketer, a lot of we always think about top and mid funnel. And yes, you want to get all the way through, but to some extent the value you're presenting the deal potentially or the coupon or whatever it may be, some of that last little conversion, that choice. Uh, it that that gets much deeper into the product as it fit all the needs, things like that. But now what we're saying is, is all of that's happening at the same time. Things that like, way down here might have mattered there, there, there being the AI is using that knowledge way up here. And so I do think that that is a big change for how marketers have thought about. It's also a very big change, which I've seen a lot of, I think very smart people recently talking about this, which is, look, your brand is so important and being visible, so important.
But to some extent, if the information about your brand isn't present and and that consumer's asking in question, it doesn't matter because you're not even going to be chosen to be in the discussion. So this is a new level of of of when we talk about visibility and content, structured data and things that the buzzwords around how to do this, I just I advise everyone to do the same thing. If you're listening to this, take your brand, take your product, take your business. Take your service. Put it into each of these models and ask it. What are the 500 questions that I'm not thinking of that someone might ask about my about my business or my category. What are the 500 questions? And then start writing out your answers to all of them. Because believe it or not, three of those questions are going to determine ten sales in the next month. And you just don't realize it because you won't even be in the dialog at that one answer. It brings me to this measurement problem that I think has been really underappreciated as well.
We are, as marketers, sort of attached to these metrics, and that we've been very comfortable with for a really long time. Those are going away as we're losing visibility. But I'm also kind of questioning if the metrics we've been using have even mattered. Like up until now. Right. The concept of rank like, does rank actually mean anything? And so you and I have been chatting a bit about this. We're launching an index, a new metric, sort of as a replacement to ranking.
Can you tell me a little bit about, first of all, what the problem was that we were trying to solve with traditional surf rankings? Yeah.
So this comes from our data science team, and they've been working on this for a very long time. So the report's just coming out. And basically what it is is when you think about how things are ranked today, uh, if you think about basketball teams or sports betting or anything like that, they follow a very similar process to what's called ELO. Rank named it for a physicist in chess, where basically they were trying to understand if you have all these people playing chess all over the world, they're going head to head who's really ranked high. So you've heard about this, and it's how you get into the different statuses. In chess, it's called Elo. So our team actually took that and start applying it to head to head competitions in classic Google Rank. Because what we have found is number one, Google ranked, we all know is very volatile based on certain parameters. So for example, I know anyone who's an SEO knows this. You'll be talking to a customer or a client. You're like, well, I'm googling myself and I'm number one, like, yeah, because you're sitting in your restaurant. Like, like, that's not that. You're broken. The experiment. So what you have to do is normalize for as many things as you can so you can control. But then what ELO does is over time, it's essentially the idea of if you have one brand going up against another, so you have a Starbucks going against Peet's Coffee or whatever it is, they only go head to head in certain locations in certain markets. So you don't say, oh, you're out ranking. That's way too simplistic and it's way too volatile. So the report that's coming out, I want to say it was 20 something million queries where we do head to head and build ELO ranks. So you can actually look at in classic search. And we're adapting it for AI to being visible the same way where you basically say, how am I doing on average in these models with this type of keyword or these types of questions? So you can get a better statement and a better sort of stable way of looking at it. Because when we launched Scout, we had all this data. But you can immediately start to feel that it's very hard to know how you're doing in your tactics and in your day to day work to get better visibility. So we launch this as a way to analyze it. So all of our clients are now starting to see this data. But it is the research is absolutely fascinating. It shows that if you actively manage your data, meaning people that work with us, we know what they're doing. We can actually see it. But the beauty of Scout is I can see everybody they compete with, even if they're not clients yet. So I can see how they're doing. And it is so highly correlated. It's something like six points higher on average in Google Rank over time as you as head and head to and every head to head race on millions of of of head to head competitions. And so you get this understanding of what I think we've all wanted a very clear metric on. Yes, it is partially based on rank, but it's really based on a head to head competition of two brands against each other in all of these models. And so I can't wait for people to read this. I think it's it's it's groundbreaking in the approach. It sounds like it's muting some of the noise that comes from the volatility of how we've been measuring things before. Yeah. And I think to your point, look, when we say how useful are these metrics. Uh, I'm always reminded of the famous marketer of, you know, half of my advertising works. The problem is I don't know which half, right. So the reality I think we all get comfort from metrics, but the reality is, is how useful they are tend to play out over time. So if I say to you, oh, you know, here's your rank, that's not really telling me, what is it that rank relative to time and my competition and I mean all the time I want to know everybody. And so what you start to find out, for example, many of the brands we work with are global brands and a lot of the CMO suite, they almost exclusively care about the other global brands that are up against, because that's the budget they're fighting. Like, you know, you're talking about the Titans. What's absolutely hilarious is if you look over time and you widen the aperture to the small players. The small players are killing them in these markets, but they can't see it because they're not looking at it and they don't have the data. Well, now we have the data. So it's so much fun to say, listen, you should think about that head to head in the markets where you're going head to head with those brands. But a lot of times the people that are beating you are mom and pops. And interestingly, some of the reasons they're beating you in these AI and other models is because they sponsor the local youth YMCA soccer teams, and they've got a ton of content about being in the community. And you're so worried about your global brand presence, you're not thinking about the actual customer journey that starts at the location and builds out from there. And so I think that's a huge mistake that so many people are doing right now. They're doing these like top down and took me wrong. I think that's important. But it's not that it's not really how the consumer's journey goes. So that's more me or you, where you are, me where I am, and the memory about us. And the questions, though that way are very different in terms of who's beating whom in rank. We have said for a long time that people compete on a local level.
Right. It's really interesting to see that that actually plays out in the data. Is there any other interesting findings that came out of this report that that surprised you? Were the rest of the team?
Yeah. So first off, there's two. So we also just released a global citation update. So I think that was on 17 million citation just for Q4. So we could analyze all the citations to understand what sources are really ranking really well in these four different models that we're analyzing. What I would say is I won't do this justice, but in the reports, we've now actually started adding calculators at Yext Research.
So I've always going to say what's interesting, the best way for anyone to see this is you can go in and select like your industry, your your Geo. You can you can make selections and see the data and download it yourself. So so we're sharing a lot more of the actual raw data in these reports. But one of the things that I think is absolutely fascinating is if you think of all these studies, they all have a bias in terms of where did you run the study from? What's the geo? And so one of the really interesting things that we're finding is that different industries in different geographies, not just like Europe, US, but just different parts of the United States, a lot of this is the competitive, um, we're calling it basically the market competitiveness factor. So if you're in a place where there's a ton of accountants and you're an accountant, I know this sounds logical, but it's so competitive there. You actually get a bigger boost by by managing your data, like getting more data out there, more structured everywhere, you get a bigger boost than in less competitive areas. And so to some extent, if you're in the town, you're the only accountant, you're fine, you're ranking great, you're you're fine. But we would expect that if you manage your data, you might see an even bigger thing. And what happens is Google expands the search radius to make sure they're bringing back at least 20 results. So you might be up against people that are 10 to 20 miles away. But in Midtown Manhattan, all of you are within three square blocks. So it gets it's a very interesting look at these ranks based on this distance, competitiveness level by industry, by company size. So there's just so many great findings in this. You can slice the data probably 32 different ways, but I would recommend everyone go take a look and read it. But um we're going to do a lot of follow ups on this. This taught us so much about how to think about long term the value of doing these, uh, this, this approach to data and data syndication. I love it, and it makes perfect sense to I mean, even as an athlete, if you're in a less competitive level, less competitive league, the bar is a little bit lower. And the higher the the more intense the competition, actually, the tighter the competition becomes as well. I remember my physics teacher in high school saying, to change your Act score by a point. Once you get into a certain level, it's the difference of answering one question. That's right. Correct. You're incorrectly. Yes. Fascinating.
Um, so, Christian, remind me, where can folks find this research? So Yext.com/research. And realistically, all of the reports are linked to the calculators. So as you read the report, you can also jump in. We also have a neat way of explaining the methodology on the ELO. There will be an animated sort of walkthrough of how we're doing it and why we're doing it this way.
But ultimately, I think our goal with the next research is to just continue pushing the envelope on. The beauty for us is with Scout data, which looks at all of the competition in any given area to one of to whoever the target company we're looking at. We're gathering so much data at scale that we can now share it and really start to give very pointed recommendations. And so if you're a Yext client, you can actually now get this data specifically for you, and you can speak to your account representatives about that. But as an industry, I think we all have to really look at what is this effect. And on top of that as AI and search, because I think really what's going to happen for the next 12 months is we're going to start talking, stop talking about search and AI, and it's going to merge, right? Gemini is already starting to do that. You're starting to see some of the search behavior happening in ChatGPT, where it looks like a local pack, like they're going to merge into like the best experience. And as that happens, we better all be prepared as to what our strategy is to measure and to have a way for us to target the tactics and the strategies that are working.
Well, thank you so much, as always, for bringing your fantastic guidance and advice. And, uh, we will see you again soon. Excellent, excellent. Great to see you and thank you, I appreciate it. That's a wrap on this episode of The Visibility Brief. If you found this useful, subscribe, leave us a review or send this to a colleague who needs to hear it. We'll see you next time.
