Yelp Inc. ($YELP)

All right. Good afternoon, everyone. Cory Carpenter, Internet analyst at JPMorgan. Thanks Yelp for joining us at the conference again. And I'm going to kick it over for the safe harbor.

Thanks so much, Cory, for having us at the conference. We'll be making some forward-looking statements during the conversation today that are subject to risks and uncertainties. Please refer to our SEC filings for more information on the risk factors that may affect our results.

All right. I got a ton of questions. If anyone on the line has one, I think you can submit it electronically. I have that in front of me or in the room, feel free to raise your hand. So let's start with just the big theme, at least from our perspective in recent quarters, you've talked about rearchitecting the company around AI. Maybe to start, could you talk about the changes you're making and just what your vision for this evolution is?

So we are absolutely reconceiving the entire Yelp experience around AI, but we're also transforming how we work. On the product side, we've invested heavily to build out what we call Yelp Assistant. That's a new experience. It's a conversational experience. We think it's really engaging. And one of the things that's clear about LLMs is people want the evidence. Why did the LLM tell me what it did? Because oftentimes, they seem to turn up results that are incorrect in some ways, erroneous or hallucinated or just inaccurate. And so that was a big theme for us is how do we communicate given the quality of our content that this result is relevant to you. So that's on the product. We did make an acquisition, Hatch, we'll talk more about that. We've built a voice AI product for restaurants that we think is best-in-class. There's a lot of things there. And then just I think we all recognize that there's this moment that came in December with that version of Claude that really unlocked an incredible amount of potential productivity. So we've embraced that as well, both on the product and engineering side, but also looking at the way that we can apply all of this to customer success, sales and certainly including the finance and people operations parts of the business.

So I know I mentioned earlier, we're going to flip it. We usually talk about the advertising business first. This year, I actually wanted to start on some of the diversification initiatives that you've made. So I think first question there is just you've got a few new revenue streams you're building. Maybe talk about the broader strategy behind that and what you're doing there.

Yes. So for us, just for context, Yelp obviously is traditionally very much an advertising business. We have 3 components to what we call other revenue. One is licensing, one is transactions and one is subscriptions. So for other revenue, we have a partnership with DoorDash that actually grew 88% in the first quarter. So we're pleased with that. On the licensing front, something that we've talked about for quite some time is how Yelp appears on other sites and other experiences. One thing that certainly emerges is there's more players in AI search and particularly in that vertical which we serve, AI local search. And so we've been entering into licensing arrangements with other platforms, most notably and most recently, we signed an agreement with OpenAI. You can now find licensed Yelp content on meta.ai, on Bing, on Microsoft, Amazon Alexa. And for a long time, we've been in Apple Maps, on Yahoo! and many other platforms. So just this idea of the ubiquity of Yelp content across the Internet on major platforms has been something that we've been very focused on, and we've been able to enter into additional licensing agreements. In addition, we certainly enjoyed getting traffic back from those sites. So that's definitely been part of the strategy and our point of view is simply that given the Yelp content and the directory that we have, if you want to be in local search, you really do need to partner with us. So that's gone extremely well. And then most recently, we did make this acquisition of Hatch. Hatch is an AI lead conversion management product. And it works across all platforms. It's not unique to Yelp. It works on Angi and Google, Thumbtack and others as well, of course, as Yelp. And we're really pleased with that. March annual run rate was at $34 million -- revenue run rate was at $34 million, growing at 92%. We did acquire the business at the beginning of February. We're still going through that integration process. Things are going well. We are investing. We're bringing people over. We're bringing know-how over. They now have a voice product. We've helped to inform that part of the product road map. There's obviously our at-scale lead management side of the business. So that is all in service to the target that we've now provided, which is $250 million in run rate revenue and other by the end of '28. So we've wanted to diversify the revenue streams because we have been exposed to the cycle, and that's been certainly more challenging in '25, and that continued into '26.

So I want to ask about that run rate. So you did give the -- by the end of 2028, $250 million or north -- or at least $250 million. Right now, you're, I think, a little over $115 million, so more than doubling. Just help us with the building blocks, how to get to where you are today to where you're going in 2028? And then also what type of investments you need to make to support that growth?

Absolutely. So as I mentioned, we're all in on AI and this AI tool opportunity is extensive. Certainly, we first want to operate Hatch and do that really effectively. We think there's additional product opportunities for the customers they serve. So I think this is very much a traditional opportunity to provide software solutions to businesses to make them more efficient or improve their return on their own investments. So we see significant opportunity there. We will continue to look for M&A opportunities. We don't have anything to announce today, but we're active. At the same time, certainly, we need to show that we can operate the asset successfully, effectively, continue to grow that asset and bring more product to market there. So that's part of the consideration that goes into doing the next transaction. And then beyond that, the licensing still presents a significant opportunity for us, and we think that there's lots of ways for us to deliver value to consumers even when they're not on Yelp. So those are really the core components of what give us confidence in our ability to grow other revenue.

One more, and then we'll go to advertising and capital allocation. So just as you're diversifying beyond advertising, how does that change your broader resource allocation at the company? And then how are you -- how do you expect AI to impact headcount and expenses?

Sure. So we've guided flat headcount for '26 with Hatch being a bit of an exception. We're certainly going to add folks there. But just in general, I think we see the opportunity to really drive productivity with AI. And over the next several years, we see significant opportunity. I think we're all learning how to apply this technology. And it's not just simply add technology and magic happens, although it is very impressive in its capabilities, you have to work differently. People have to think about their roles a bit differently and people have to adapt to the way that the tools are most effective in helping them to do the work. So that's still a bit of a journey. But clearly, when you see the capabilities that come with it, the intelligence that's associated with it, the ability to reach throughout an organization for data through various MCP servers in the enterprise that bring back information, that's really interesting. In addition, applying AI, I think we all are very familiar with the product and engineering use cases. But in clearly call centers, inbound calls, customer success, there's clear opportunity for call deflection and in some regards, in moments, even better customer service with AI. But what is certainly clear on the sales side is the opportunity for coaching. That's really emerged as a capability. We've built that ourselves, but I think you're hearing that from other companies. But this idea that the AI can listen and provide in real time or right after a call, suggestions to a sales rep on how to improve the conversations that they're having. We're having real success with that. Equally interesting is giving the manager tips on how the reps are engaging in conversations so that they can be more effective as managers. And that's rapidly emerging as a way to enhance productivity. So we're excited about that as well.

So just building on that, as we think of EBITDA margins, you've been working through this stock-based comp transition for a couple of years. I think it's fair to say '26 is a bit of an investment year for some of the initiatives we've talked about. But I also thought it was notable, you did say at earnings a few weeks ago that you expect strong growth in EBITDA margins over the next several years. So maybe help us with where you expect the cost efficiencies and leverage to come from? Maybe you just kind of answer that with that last question. And how does the margin profile of your newer revenue streams compared to the ads business?

Certainly. So again, it's going to take a little bit of time. But if you just think across -- and bear in mind, I think there's 2 camps or 2 sides to this. There's productivity and there's efficiency. And we definitely are lined up on the productivity side. We want to grow revenue, and we want to deliver great consumer experiences, and we want to deliver value to our customers, whether they're buying a Hatch product or they're advertising on Yelp. So the way we think about it is faster time to market, more output and really adaptability when you get feedback from customers, this is what we need. This is what isn't working. I'm willing to buy this if you had it. That cycle time is compressing very rapidly. And again, this is certainly not unique to Yelp, but I'll give you a quick example. We came out with a voice AI product for restaurants, which we call Yelp Post that picks up the phone, can make a reservation, it can answer questions about hours, location and things like that. And we realized that there's a much larger TAM if we could add food ordering to that. And so from the time that we realized, hey, food ordering is a real opportunity to where we are now, which is in testing with this product live at -- with restaurants, that is a couple of months. That was just not possible in the past. And by the way, food ordering, super complicated product. If you think about someone's going to call up, they're going to have additions, changes, they've got specific requirements. They need to send notes to the kitchen about what's being ordered and to be able to handle all those edge cases accurately because the person is going to show up, they're going to see what they're ordered, they will have paid for it and they take it home and maybe they don't notice, the possibility for being disappointed or frustrated is very high. So accuracy really, really matters. And as we're just talking about, LLMs are error prone. So you got to have a lot of know-how and actually a lot of training data to fine-tune the models to be able to do that. So cycle time enables faster time to market, obviously, and ideally more revenue. And then I talked a little bit about the efficiency of our sales folks. What is still pretty emergent for me at least is in the finance and people operations area, I think it's much earlier in terms of the efficiencies that can be achieved there, but things like collection should offer an opportunity even the way that you handle bad debt should evolve. So when you stack that up across product, engineering or product managers now prototyping features before they ask engineering to build them, closer partnership and integration between engineering and product, that's a whole opportunity. Sales and marketing becoming more efficient, customer success being more efficient, finance, people operation. I mean, it's literally the entire enterprise is going to get more efficient. So that's why we said that we believe that there's a very good path to margin expansion over the next several years, even though we have chosen here in 2026 to invest in the business.

So I want to hit on capital allocation, and then we'll talk about the core ad business after that. So just on capital allocation and strategy, you did a pretty sizable buyback in the most recent quarter. I think you took on debt for the first time. Maybe just talk about how you're thinking about capital allocation going forward.

So we did deploy a considerable amount of capital in the first quarter, Hatch was a $270 million purchase, and then we do have $30 million in retention set aside that will pay out over the next 2 to 3 years. So we had $270 million there. We also did choose to repurchase $125 million in stock in the first quarter, which was considerably more than we've purchased in the past. Now we did see that diluted shares outstanding from the first quarter of '25 to the first quarter of '26 went down 12%, which we think is a significant number. At the same time, you can expect a much more moderate pace of share repurchases through the remainder of 2026 compared to the first quarter because we did really obviously buy a significant amount of stock in the first quarter at $125 million. And then, of course, over time, what we do want to do is to be able to look at M&A opportunities. And we have heard from investors that they'd like to hear more about the capital allocation policy, and we're going to share a bit more about that on our Q2 call.

Okay. Perfect. So shifting to the ads business, still over 90% of your revenue today. Just talk about the state of the RR&O and Services segments and the trends that you're seeing across both of those.

Yes. So RR&O has continued to be challenged. I think we all saw the -- both the CPI and the PPI prints of the past week and consumer sentiment is challenged. We see the 10-year at $460 million. So that's also another headwind, plus just a degree of uncertainty about the direction of the economy and price at the pump being something that's very top of mind for folks. So when you look at people making those decisions about eating out where it's more expensive to eat out and they're concerned about having to spend money on other things to transportation or whatnot, we've definitely seen a continued headwind. Historically, for us, our Services business was a very consistent double-digit grower. And what we saw as we came into '25 is that there was a pretty rapid reaction, faster reaction than historically among smaller business owners to tariffs. And that impacted '25. We came into '26 and then the conflict in the Middle East has also created uncertainty for business owners. And in California, at least you have diesel at about $8 and you have gas at $6.50, and that's top of mind for folks. So the dynamics are playing out in that physical local economy. The overall GDP headline number is still obviously quite robust, especially with the CapEx going on in AI. But when you look under hood in that physical local economy, it has become quite challenged. So our goal is to continue to deliver value, whether it's on the restaurant, retail and other side or the Services side. And that way, we are participating as things improve, but it's also driven this focus on revenue diversification so that we're not as exposed to the cycle as we have been.

Someone who just moved to the Bay Area the gas prices are eye-popping relative to. All right. Macro. We got an electric vehicle though. Macro. So like you said, it's been a headwind to RR&O for a while. I think 2 follow-ups on that. So one, what needs to happen for trends to improve? Like what are the leading indicators that we should look for? And then secondly, one question we get on macro is, is the headwind more due to consumer spending less, advertisers lowering their budgets or perhaps both?

So I would say very loosely, consumer sentiment broadly is an indicator of willingness to spend. So I think that's certainly a component. Inflation does need to moderate as if inflation expectations are high, that has an impact on decision-making. But also, it's very hard for businesses not to pass along cost now. They've been squeezed. So when a small business can't pass along cost, which has been the case, and they're seeing higher input cost, whether it's ingredients, materials and labor, then the room to advertise is compressed. And so that's certainly been an indicator. So at that very top level, I think small business sentiment, important, consumer sentiment, important, inflation, important, but I think those are probably broadly indicative of how you could expect many businesses to perform. So I'm not sure that's particularly unique to Yelp per se. But those are the actual factors that we think are going to indicate whether or not our business performance can improve. All that being said, it's not -- we're not pleased with this performance. It's not as if we're sitting on our hands waiting for those things to improve. We want to continue to improve the product. We came out with, as I mentioned, our Yelp Assistant, which works across all categories. We think there's an opportunity to increase our monetization there as we engage folks in a conversation. So that's one thing that we're certainly doing. Clearly, for advertisers, you want to continue to improve the value that you're delivering. So it's not cost per click they care about, it's cost per lead. So what are all the things that we can do. In that conversational interface, we actually get more information than we do when someone just types in the hey, plumber. So can we match more accurately and deliver a higher quality lead that converts better, that therefore, has a lower cost per lead for those advertisers. That's absolutely something that we're continuing to work on. So we're not sitting on our hands. We're not satisfied at all with the revenue performance, and we're certainly continuing to take action to improve that.

All right. One more on macro, then we'll move on. So at earnings a few weeks ago, you did say the elevated uncertainty weighed on ad spend in March. I think trends maybe started to improve in April. Just could you expand a bit on what you're seeing and what you're assuming in your outlook?

Sure. So one of the dynamics that played out in '25 and repeated in '26 is with the tariffs in ' 25 April, let me back up. Normally, we see the seasonal ramp in advertising spend as you go into the spring and summer, makes sense. We're doing repairs on your house, you're traveling more, you're going out more, weather is better. And so that seasonal ramp was interrupted last year in April, really around tariffs and that seasonal ramp was interrupted this year in March. And so you got a flattish shape for ad spend trajectory in both April of last year and March of this year. And then we did see a bounce back in April of this year after that flattish performance in March, but you're starting off of now a somewhat lower base compared to what the expected trajectory would have been. So we need to reflect that in the guidance that we provided for the year. We need to see how things play out, but that's the shape of the revenue as we see it for 2026.

So a couple on Services, and you've hit on this a little bit, but I think you mentioned strong double-digit growth historically in recent years for that business. Trends have started to moderate. What's been the driver of the slowdown? And maybe more importantly, how do you think about the growth trajectory from here for that business?

Sure. So service is now more than 70% of revenue at Yelp. We definitely have pursued that. For those who have followed Yelp for a period of time, we really differentiated the experience on Yelp between restaurant, retail and other and Services, and that really served us well. We're now in a new era, which is this conversational era. And we continue to see an opportunity to really drive that engagement on the consumer side and deliver value to them. And with that additional information, be able to match those consumers with the exact right service provider. So we're also in a new -- emerging into a new era around Services. It is still a little bit puzzling to us why Services decelerated the way it did after such consistent double-digit growth, which we actually saw through the first quarter of last year. So that hurts a little outside our control. But the thing that is very important, obviously, to service pros is, "hey, I got that lead". And it's a little bit of what motivated us to make this Hatch acquisition because if you can use an AI to answer every message that you get or click that you get or call that you get and really improve the conversion, that's a great opportunity to provide value to them, but it also means that they're getting the return they expect and are willing to spend more on advertising. So these -- the strategy is actually very integrated across both the AI tools that we're building, what we've learned from doing that, applying some of the capabilities that we've built at Yelp around voice, for instance, and serving our advertising customers more effectively. So it's not that the focus on this revenue diversification is something entirely separated. It's actually very integrated as a coherent strategy across all the customers that we're serving.

And then home Services is historically, I think, still your largest category within Services. What type of trends are you seeing within home Services in particular? And any other -- where are the growth opportunities within the Services categories that you're seeing?

Certainly. So on the Home Services front, I think you've heard some of the larger retailers mentioned that people have traded down from a remodel, say, to a repair. And some people have traded down from repair to a do-it-yourself because of the pressure they're experiencing. And so the ticket size has decreased. And instead of maybe replacing the roof, you patch the roof and maybe instead of doing a new kitchen, you replace some appliances. And that is going to play out for a period of time yet. Again, as I've already talked about, we're going to do a lot to certainly focus on the things that are within our control, but that does need to be addressed. And I do think that there just needs to be more certainty around the economic outlook for folks. I think we need an environment where people are able to just really focus on delivering against the opportunities in front of them as opposed -- I'm saying for small businesses, as opposed to wondering what might be the case in a month or 2, how do I manage my cost? How do I manage my labor? All of these things, if you think about a small business owner, they're doing everything. And there's only so much time in the day, and there's only so much they can concentrate on. And most importantly, they need to win business, deliver it, build for it, have a happy customer and move on to the next job. And the more distractions there are, the harder it is for that business.

So a couple of questions on product. And then if anyone here has a question, feel free to submit it online or raise your hand, and we'll get to you. But -- so you talked about Yelp Assistant, that's now rolled out across all categories. Maybe talk about what that is for those less familiar? And then what are the early signals you're seeing around consumer adoption?

Sure. So Yelp Assistant is our chat experience on Yelp. It was previously only for Services. Now it's for all categories. I encourage all of you to try it, download the Yelp app. There's an icon for it, a tab. And there are things about it that I really, really like. One is it's very engaging, has a lot of personality and it's very much Yelp. You'll find that the experience is very much what you'd expect from the Yelp brand. And it's insightful. It's taking what you share, and it's really tailoring its commentary to what you're interested in. But even more importantly, it's then surfacing the snippets from reviews that show why it's saying that to you. So the evidence is right there in front of you. And I think this is where everybody is going to have to go because the LLMs have a high degree of error is it's not just that I got a response that seems credible, reasonable, accurate. It's that I can convince you that I have the evidence to support the reason why I'm making this recommendation for you. And because we have this incredibly high quality content, we're able to surface that evidence, and we can do it in a way that's very compact. And so you can scroll through and do a little bit more and research more or you can just see what's the next opportunity and why is it engaging for me plus, of course, there's photos. This surface also presents a lot of opportunity to add additional local information. If you think about things that might influence a decision, what's the weather going to be? Is there an event nearby? Is there a park? Is there something to do with kids afterwards? Or is there outdoor seating that's dog-friendly. Suddenly, you can include a lot more information in this experience that isn't so easy when it's just a list of search results. And finally, because it's conversational, you can ask the follow-up question. "hey, I saw you went out to have Italian food. How is that"? Or "hey, you needed a plumber. Did you hire that plumber? Did you get the leak fixed"? Suddenly, just as we've all experienced with other chatbots, they ask the follow-up question and they're more helpful. They are more engaging, and it's a more useful experience for you. So we're doing all of that within that, and we're pretty pleased with it, the early results and they're very, very early that people are like it. The early NPS is very encouraging, and it's more engaging and people are clicking more. And that means that there's an opportunity also within that to increase monetization without the perception of increased ad load.

Another product I wanted to ask about at earnings, you mentioned Yelp Host. I think you said 1.5 million calls on a run rate basis. How should we think about the TAM and just the monetization opportunity that you're going after with Host?

Yes. We definitely think this opportunity is well over $1 billion. Just to give you a sense, we now have 16 languages supported in the product. We have 4 voices that you can choose from in English, but 6 regional accents that you can also choose from. And some of those accents are, for instance, French. It might be a French accent. And what's super interesting is the person calls up, they realize that the AI is a French accent. Can the AI speak French, and then they start speaking in French and sure enough, it speaks French. And so it's a delightful experience. And it's able to answer the questions accurately. The other thing that's been really fascinating in building this product that I certainly have found extremely interesting is in that first half a second, you have to convince the person that they should engage with the AI voice and not just have to transfer. And what we saw in some of the previous iterations was people would give a monosalabic response, transfer. They like instantly knew, this is a bot. I don't want to talk the bot, transfer me. And we made a few small changes. First of all, we have a very human sounding voice. It has intimation and it has emotional quality to it, has cadence. But a few changes and suddenly, people started speaking of sentences to it. And the moment people move from a monosolabic response, yes, no transfer to a sentence, they use it, and it's useful to them. And then when they come back, they know that they can just talk with it. And that's a pretty incredible moment, but to build that takes a lot of skill. There's a lot of know-how. It's not just stitching together different services. You've got to have the training, the fine-tuning for the models based on the corpus. You have to have good engineering, of course. And most important, because we now have this 1.5 million run rate for calls, lots and lots of reps. You have the opportunity to experiment, refine, change, see what works, see what doesn't work. And so I just see us being able to accelerate the quality of this product over time because we're accelerating the quality of this product over time, it's self-reinforcing.

So we talked about Assistant Host. You released 35 new features earlier this year. Anything else on the product side you think it's important to highlight?

We do have a really cool feature that we released called Yelp Menu Vision which is really cool. It's an augmented reality experience where you just point the phone at the menu and it will show you the dishes in the ratings, that's really cool and very helpful because I don't always know what the dish is, and I'd like to see what it looks like, and I want to see what people said about it. And so that's been actually a lot of fun and people are finding that useful.

Interested to try it out. Okay. Two more questions, and I think we'll be out of time. So I think one of the common questions we get from investors is around competition, around consumer traffic in general to Yelp. I know you disclose traffic metrics annually, which you did at 4Q earnings. But what can you tell us at a high level around traffic trends that you're seeing? And what type of impact you're seeing from changes in the search market?

So there's a lot of dynamics going on around Internet traffic right now. One thing that we haven't seen is any meaningful impact from AI overviews. We get that question pretty regularly. Our hypothesis is that these are highly monetizing categories that we serve and Google wants highly monetizing categories to monetize. So that hasn't been something that we've seen, though I believe other content players have seen impact there. In addition, Google seems to also really be prioritizing human-generated content. We're not alone in that. And some of the other platforms have certainly seen real benefit. We've seen benefit from our human-generated content as being a benefit. And then broadly, of course, is just this emerging experience that we're all having chatting with the different chatbots, whether it's an OpenAI, it's now Grok, now Claude, which now seems to be a consumer app and not just an enterprise app. And so I think we're all seeing how that traffic plays out because we have this high authority content. We do see that people want to engage with Yelp content. They want to see the Yelp logo next to Ratings to know what the rating is credible. And we believe the way that this plays out is as you want more information, you're going to click through and come to Yelp, and we like the exposure that we get from that. So traffic is certainly evolving, and we continue to believe that we're really well positioned because of the high-quality content authority and brand that we have.

I have a closing question. Anyone in the audience have a question before I do? All right. Perfect. Easy. Okay. So just ending on a bigger picture question. What are the 1 or 2 things you're most excited about, but investors are not talking about today and you think could really be transformative for the business in the years ahead?

I think we really have covered the topics. I think the part that I'd go back to is this transition to conversation. We've talked about it. We're all experiencing it. I think it's going to be very transformative in the way that we engage with content. What I don't think is going to change there, and we do have this back and forth, and it's certainly going to emerge. But I think in the United States, what we're going to see is specialty verticals where you specialize in a particular type of content like Yelp is going to continue to be very differentiated and enjoy a strong competitive position. And I think there's just this open question that everybody is still wrestling with and it's certainly going to emerge. But do you have terminal value? Is this a sustainable business? Are you going to be around in the years to come? And at least from the way that we think about it and all the research that we do with consumers and what we see through the direct interaction with the app is that people do want that differentiated, vertically specific experience in conversation. And again, this idea that you can add even more information and relevance as part of that and that we can do that as Yelp, that to me is very exciting because it becomes a more comprehensive, answer to questions around local for consumers that we haven't answered in the past, but we believe we can answer in the future. And so there's a really cool road map of things to come around Yelp Assistant in particular, even as we start to build out these AI tools, even as we're present across other platforms and continue, we hope to deliver value to both consumers and to businesses.

Awesome. We'll leave it there. Thank you.

Thanks so much.