Yext, Inc. (YEXT) Earnings Call Transcript & Summary

My name is Rohit Kulkarni. I'm the Internet analyst here at ROTH MKM and super excited to have Mike Walrath, CEO of Yext. I'll let him do the quick intro. But what I would say is, Yext is a company that has pioneered search on the enterprise side for many, many years before all the things that are on new forms of [ services ] that are happening in the last 3 to 6 months. So it's a very exciting time for our company, Mike. But would love to get a state of Yext right now as to -- just to -- like how I would write it in ChatGPT. Explain Yext as you would do to a 5-year-old.

Okay. Thanks for having me. It's great to be here. So the best way to think about Yext is, as a next-generation digital experience platform, which marries innovative content management system with various forms of natural language processing and AI technologies to make sure that a business has the ability to deliver authoritative answers to questions across a myriad set of digital experiences both on their own properties and importantly, also on third-party sites. And there's a lot of sort of technology that underpins that. We've been, as you know, making a very large investment over the last 4 or 5 years into generative AI and large language models for both the creation of the authoritative Knowledge Graph, but also for the generative capabilities. And it's not been an easy time to be making that kind of investment because I think there was a lot of question about whether or not it was going to actually ever arrive. I think we can definitively say that it has arrived.

Awesome. Fantastic. I think that's -- the way I would like to talk next 20 minutes is, let's start with very big picture, long-term things than medium-term and then near-term -- great near-term things. So generative AI, I think it's something that is -- you've been in tech for a long, long time. I think it's one-in-a-generation thing. And one of the most important aspects of tech that it is affecting in search, and I feel Yext is a search company. So maybe talk through generative AI, ChatGPT and all of the large language models, how should investors think about the impact of this technology on Yext's business? How do you view the opportunities, challenges? And what do we expect from Yext over the next 2 to 3 years when you think about monetizing this new technology?

Sure. Yes. So I think that there's a reason why generative AI and search are being talked about a lot together, and that's because the 2 technologies actually complement each other really well. So I think we've all seen ChatGPT and other demonstrations, particularly in the last half a year or so that show the power of generative AI. But at some level, there's an element of a parlor trick here, right? It's a very impressive one, but the AI doesn't know what it's saying, right? It really doesn't, right? It's a statistical model of what word comes next. And so one of the examples that I use all the time for this is, if you ask -- if I ask ChatGPT to write a biography for myself, it will write a biography. And actually, it will do it over and over again. It will be different every time. And every time about half of it will be factually correct, and about half of it will be entirely made up, right? And that's because ChatGPT and other forms of generative AI do not understand the difference between fact and the most logical word. And so the best example of this is that it will almost universally say that I went to the University of Virginia, right? And I didn't go to the University of Virginia. I went to the University of Richmond, but in a high -- there's a very high probability that the word combination University of will be followed by the word Virginia, when you have any connection to the state as opposed to Richmond. So the reason why that is really important when it gets paired with search is that -- this is why you're seeing so much interest from Microsoft and Google in these technologies is because you need ways to use the generative AI for what it's really good for, which is understanding the question and understanding the nature of the answer that's being looked for, but then you also need -- you need to tie that back to technologies that enable a fact-based approach to providing the answer, and this is one of the things that obviously, it's getting a lot of knocks for today. So it's obvious how this is going to play out in the consumer market. What I think is newer and I think what people are trying to figure out is what does this mean for the enterprise? Because you cannot allow generative AI to deliver facts about your business because it will hallucinate facts. It will make things up. It will -- and because that's what it does. It will get better at this over time, but the best way will always be to use the generative AI for what it's really good for, which is understanding the question and elements of delivering the response stylistically in a conversational way but constrain it by allowing it only to answer questions that have been -- or the authoritative answer has been structured and provided by the business itself. And so for the enterprise, this is going to be the fundamental shift in how they think about their digital experiences is that you want to enable AI and things like chat and things like review response and content generation and things like that, but you want to do it within a constraint that it won't chat about things that you don't want to chatting about. You don't want it opining on politics. For example, on your website because those become your opinions if it's your chat functionality that's doing that. And so I think those types of shifts are fundamentally changing the way that businesses think about how do I use these tools in a way that's safe. And we've seen it with companies banning the use of these tools inside corporate systems because it's not yet safe.

I guess -- so from generative AI and the adoption of that into enterprises and kind of the way some of the enterprises are trying to play with it, you have already launched a couple of quick products. They are still early, chatbot and content management systems. So maybe talk about like over the next kind of foreseeable future, what kind of use cases are you hoping to solve with these new products? And then I'll come back with a follow-up here.

Yes. Thank you. That way I'll actually remember it. So I think -- we have launched a couple of things recently. There's been a lot launched in this area recently. And part of the -- this sort of site guys around OpenAI has -- it's made it very easy to stick a UI skin on top of OpenAI. We're going to see a lot of this, right? And it goes back to -- that's a very dangerous thing for a business to get involved in because unless you have the underlying authoritative answers attached to it, you're going to get all these bad answers. So we're going to see tons of products that come out, but very few of them are going to be architected from the ground up around this idea of a structured Knowledge Graph, which is what has been fundamental to Yext business for the last 10 years plus. And so a couple of examples. So we launched chat into a beta last week. And what differently, one of the thing that differentiate Yext Chat is that we're using GPT as the way of understanding the query, but then we're using the connection to Semantic Search and the Knowledge Graph to deliver the answer. And so that ensures 2 things. It ensures that the answer will be authoritative and ensures that the scope of the answer won't be beyond that, which the business wants an answer. So if you're a large enterprise, you don't want it responding to certain queries that are going to be reflected by your brand. The second one you mentioned is content generation. Content generation is a really interesting opportunity. When you have a large structured data set like we have with Knowledge Graph, which is our content management system that holds all of the facts of a business and actually brings facts from other systems in to become the single source of truth. It creates an enormous opportunity to create content. So as an example, if you had 100 hotels in Southern California and inside the Knowledge Graph, you had every fact about -- and a piece of information about those hotels, a really interesting use case would be to ask the generative AI that's connected to your knowledge system, write me a blog post about what do I do on a vacation in Southern California. Write me 10 blog posts about what to do on vacation in Southern California and make them all centered around things that you can do at my hotel properties. That then becomes content that becomes indexable and searchable in the broader world. And by the way, every one of those is going to link back to any one of your 100 hotel properties in Southern California. The implications of this technology for business are absolutely endless. And it's like the world is kind of -- has woken up to the opportunities that exist here.

Awesome. I guess maybe in all of this, clearly, you have the Knowledge Graph, you have the relationships with a lot of enterprises, and you have long-term tenure with some of those relationships. So maybe talk through maybe that's the secret sauce, but in your point of -- in your point of view, what is the secret sauce that I think Yext can succeed in this new world of generative AI?

I think it goes to the evolution of the business, right? So we started the business with listings information, which was a really tricky problem about structuring location-specific information that could then be disseminated across a network of hundreds of websites. So if you search for a business on Google or Yelp or TripAdvisor or a number of others, what you're going to most likely find is a result where the -- all of the details of the information have been delivered more often than not by our platform, and they've been verified and updated by our platform. And every time you change anything in the Knowledge Graph, it's going to be updated. So that was our first digital experience solution. It's the solution we took the company public on. And what we've been doing since is adding additional digital experience solutions around page creation, site generation, reviews management and analysis. And then obviously, search over the last 4 or 5 years and now more of these generative technologies, all of which are basically underpinned by the exact same thing, which is once you structured the Knowledge Graph and you have your authoritative information either living natively inside the Knowledge Graph, the CMS system I've been talking about or being pulled in through custom or prebuilt APIs from all the other systems of record in your business, you can now activate these things very easily because the content is already there. And this is the future of content management. It's a completely different business than what we've typically had, which is a sort of static, very specific content management approach.

Fantastic. I guess switching gears from AI to something more over the last -- looking back 12 months, maybe that's what I would love to see. Just for the audience who are new to the story, talk through kind of what -- how things have changed at Yext since you became the CEO. And mostly from a management standpoint, personal standpoint, and then we'll talk also about the go-to-market.

Yes. So obviously, we've seen a lot of change over the last 12 months. It's felt at times a little bit longer than just 12 months. You've been along for this ride for a long time. So you probably felt longer than 12 months for you, too. I think as happens to a lot of companies, when you're -- when if growth begins to decline, it becomes a very difficult situation to manage, we were dealing with naturally declining growth at some level. And then we're also very exposed early on to -- because of our location-based sort of routes, we're very exposed to retail and hospitality during the pandemic. But equally importantly, we had made this enormous pivot, assuming that the market was going to be moving towards generative AI and the need to marry generative AI with search. And frankly, we are just way too early to it. And we are way too early to the level of commercial effort that we put against it. And so what we saw was this tremendous output of energy against convincing businesses that this was a really important thing for them to do, which wasn't backed up with the financial results. And that made the last few years, particularly difficult. And so when we made the management changes that we made a year ago, we really reset the business around, we need to get back to the basics of core efficient growth first, right? And that means restructuring the business, we're a smaller business. We're more than 20% smaller from a headcount point of view. We're obviously much smaller from a cost base point of view this year as well. And we're obviously also growing slower. So we're now in a position where we've made the hard changes that we need to make to be able to reaccelerate growth. The difference this time, and we've talked about this a lot publicly, is that fundamentally, we're going to reaccelerate growth using a go-to-market machine and the data around that machine to make sure that we have the information we need to grow our sales and marketing at a rate which lags our growth as opposed to leads our growth. And it's just a different philosophy, but it's one we're very committed to.

Okay. Okay. I think you recently reported earnings, gave next year's fiscal year outlook. And here comes the hard question. So far, we've been just talking at 30,000 feet level. But I think the key question, I think, for investors is, yes, there is clearly a bright and shiny object, which is AI, which is in the next 12, 24, maybe 36 months. But what happens to the core business? What happens to the core business that's making like generating what, $400 million in revenue. That's not a small jump change. So how does that grow? And what gives you confidence that, that can grow not just single digits, but how it used to grow solid double digits. When does that happen?

Yes. So I'm not going to predict when that's going to happen. So anyone who's just here for that, you're going to be disappointed. I think it's really important to say that the sort of the core business or the legacy business, some people call it is the current business, right? And the down -- the upside of having made this enormous investment in this pivot toward generative and AI and large language models in every element of our business is that this isn't a new thing. We're not chasing something -- we've been doing this for 4 years. We've just been doing it with limited success because commercially, it's been very difficult to have this conversation with businesses. So that conversation gets easier every day because it's been described by many as the most important conversations that businesses are having because they see this as a fundamental shift to how they're going to -- and a fundamental threat to how they deliver information to customers, employees, partners and all of those things. So when we think about growth, we see no shortage of opportunity, but we didn't see a short of opportunity 3 years ago either. We think it's a -- We think there's a lot more momentum around the opportunity now. The way we'll think about it is, it's really simple. There are fundamental metrics we can measure and manage. They're going to tell us how does this -- how do we know these businesses ready to grow and that we can put the leverage into the market against it. One of those is sales productivity. So we've talked publicly about we've seen a significant increase in sales productivity over the last couple of quarters. And that's partly because we got smaller and it's partly because we're doing a better job with our customers and because the opportunity is getting better. The second one, which is going to be incredibly important. And by the way, we've hired Tom Nielsen to run revenue who's very adept at building these productivity-led machines. The second piece is that productivity alone and approaching the kind of max productivity curve doesn't -- alone doesn't tell you that you're ready to grow. If you see maximum productivity and you expand your sales resources, you're -- we may or may not see the growth, and that goes to the second really critical metric around qualified pipeline. So you need to have both productivity approaching its sort of terminal limit, if you will, and also qualified pipeline accelerating in order to know that it's time to that you can apply leverage in the market. And so by bringing in a new Chief Marketing Officer 6 months ago, Raianne Reiss, who came from Amazon, who is incredibly adept at building these demand generation machines. We've really set ourselves up to be able to have the 2 key metrics that would allow us then to accelerate the growth of the business. The best news is that there's still headroom to grow the productivity and that growth is very close to free, right. So as we get better at this, what we'll get is you get the kind of the first bit of growth at a very low cost because you don't actually have to add any capacity to get it. Then once you're adding capacity, you really want to make sure the demand environment is robust, Otherwise, you get diminishing returns, which we've seen, and we don't want to go back there again.

Okay. So I guess with regards to -- you tried to address this a little bit, but with regards to just the sales productivity, maybe talk about -- just the overall go-to-market strategy. I know that there was -- you have changed the way you think about direct versus indirect. You have -- the way you think about consulting versus and so on and so forth. So maybe just level set, where are you with how the go-to-market is evolving? And what are the KPIs that you feel that we need to track from the outside?

Yes. Well, so I think the best indicator that we're -- and I've said this many times, the best indicator that we're seeing traction is going to be our direct ARR because it's the best measure of how we're doing selling our software and answer end value-add services, although we've deemphasized some of that 2 businesses. And so we have a reseller channel that represents a little less than 20% of our business, and there are more difficult market environments in those channels. So while there's opportunity there, we're less focused on it today as a leading indicator. The thing about rebuilding the go-to-market is that it's a series of coordinated motions that have to be a choreographed, right? So you can't just have sales. And this goes back to what I was just talking about, you got to have marketing and sales in alignment with each other. And then with the moves that we're making to shift more of our services to the RSIs and the ESIs and the partner ecosystem, that also feeds in very importantly because you're delivering value to customers through relationships with third parties, but then that value actually, that flywheel that gets created is around the RSIs and the partner community delivering demand back into your system because the way they grow their business is by you selling more to their customers, right? So that they can deliver the services that they're really good at around it. And so when you take all those different pieces, it's very complex. But the metrics that we -- and we try to do [ as good a job ] we can show in these metrics around. ARR and breaking out the ARR, which we started to do this year from direct and reseller and then obviously gross retention and net retention are important metrics there, which we try to make as visible as possible.

Okay. Okay. I guess just from sales and marketing and go-to-market, I guess, from towards more on the product side. I know you're going to have an Investor Day at some point in early April. So we'll probably look at new products or other things over there. But just big picture, since you became the CEO, for example, can you talk about what are the new product priorities that you have implemented? And in your opinion, kind of which are the new products or kind of changes to the products that are more significant in your standpoint?

Yes. So I mean we talked already about a number of the generative things that we've launched -- but I think one of the things that gets lost and is really important is the pace of innovation on the Listings business and the reviews and Pages business. So we've had fundamental improvement in the overall automation of the Listings business. And those have been major focus in the last few releases. We've completely relaunched our Pages business, which is effectively website creation to enable third parties to use it as more of a development platform. And we're really building the whole thing as much more of a development platform, which I think takes a lot of the friction out of selling into organizations who see themselves as builders. So as a digital experience platform, we're -- there are really key components of the architecture that we can deliver. But many of our customers, they want to use their own resources to build on top of these things. And I think it's a really important piece of consolidating the different point solutions functions, which our competitors aren't able to do.

I see. I think just innovation velocity, that's something that is, in my opinion, pretty important leading indicator. You're obviously doing a lot on AI, but also automating all the things on the digital experience is also something in my book, an extreme positive. I think another aspect is obviously pricing. Anything that you want to add with regards to how you're viewing pricing or price elasticity or pricing leverage that you may have in the model?

Yes. So I mean we feel great about the margin structure of the business, and we recently updated our views on that. And so I think when it comes -- because we have such a breadth of solutions, not all these solutions are going to be priced the same way. And so we're going to be very pragmatic about some elements of these that are going to lend themselves really well to sort of a licensing subscription structure and some are going to lend themselves much more to a usage-based structure. And so I anticipate that we'll be evolving on these fronts. And as we continue to add new products and solutions, we'll look for the best, most customer-friendly way to price these different solutions so that we can deliver value.

Okay. On the -- switching gears on the macro side. I know we've talked about this very recently on the earnings call, but I guess the most follow-on question has been how to peel macro versus micro headwinds that you're having. So maybe talk through some of the macro headwinds, be it by vertical or by geography that you feel are still weighing down on the business and some of them may be actually reducing to some extent?

Yes. So I mean, I think it's more overall macro than it is any geography. We probably saw the headwinds, and we mentioned this last year show up a little earlier in Europe than they did here in the U.S. But really, what we're seeing is we're seeing increased scrutiny on expenses coming out of corporations of all sizes. And that's -- look, that's the bad news. And I think everyone is talking about it, right? The procurement cycles are longer. The spending is being reduced overall. And then we're doing it in our own business, right? We have a major initiative to reduce the number of software vendors that we work with and create leverage with the ones who bring the most value. So -- which creates a real opportunity because we have an opportunity. Our platform is much broader and much more mature than most of the point solution competitors who we face. And so even in light of this, I think the upside for us is being able to go in and talk about consolidation of roles and functions within -- or I'm sorry, a functional technology inside an organization. But I have to tell you, like our outlook for the year assumes this isn't getting any better, right? And it assumes that if anything, there may be even more pressure as businesses focus on cost cutting and become more bottom line focused. And we're just going to be conservative about it because there's really very little upside to not being conservative about it.

Okay. I think what you've done with regards to being conservative from a cost standpoint, I think that shows up in the numbers. I think the way the profitability of the business has almost inflected with soft revenue growth is something very highly, highly commendable. I guess we are almost running out of time. Maybe just one rapid-fire prediction. On the search side, does Microsoft GPT gain share versus Google?

Boy, that's -- I don't know. Look, I ran that business. It was a big part of running that business at Yahoo! for a while, and it's really hard to compete with Google. I think what GPT and Bing are going to do together is really interesting. And I think if nothing else, everybody knows I'm a golfer, it will be -- it might be something more like the Live PGA TOUR thing where a lot of pressure is placed on the incumbent. For us, we're agnostic to it. It doesn't matter. We want to see more of these models. We want to see more of these. We see them as -- they're becoming commoditized very quickly. That's a good thing for businesses, and it means that there's going to be a lot of options. I don't know how to play the -- who's going to win the search for us.

I think you did a pretty good job. Thank you, Mike Walrath.

Thank you.