Good afternoon, everyone. Thank you for joining us at the Morgan Stanley TMT Conference. My name is Elizabeth Porter. I'm an analyst on the U.S. Software Equity Research team, and I'm really excited to have with us today Klaviyo CEO, AB. We are taking audience Q&A, so mics will go around at the end. And for important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/researchdisclosures. And with that, thank you, AB, for joining us. Yeah, thanks for having us. Last year was your first year in attendance as a public company. For those that still may be new to the Klaviyo story, I think it'd be great to walk through the founding story, what you set out to do with Klaviyo, and how the founding experience helped frame not just the strategic, but also the financial principles of the business today. Sure. So we founded Klaviyo back in 2012. And my co-founder, Ed, and I, we had built data systems and analytic systems for a lot of the largest consumer companies, a lot of retail and banking in the world before starting Klaviyo. We'd worked with businesses like McDonald's and Starbucks and Bank of America. And what was fascinating to us, and this was sort of the mid-2000s, was there was all this first-party data that these big retail businesses had. But it was hard to pull it all in one place. It was hard to make sense of it. That was the part that we were not only working on. And then it was really hard to action. And if you think I'm a big fan if you think forward of what we think the nature of buying between businesses and their customers is going to look like in the future, it just felt like more of that was going to get automated. It was going to get driven by software. So it was really important for these large retail businesses, consumer businesses, if they were having a hard part of this pipeline of aggregating data, making sense of it, and then doing something with it, boy, somebody better build the technology and infrastructure to do that. Along with that, just we're big, both of us, we're just kind of students of history. And we looked at, gosh, there's this category called CRM, which normally has been kind of focused on B2B businesses. If you have a sales team, an account management team, it's great software for them. But that addresses like a third of the economy. Two-thirds of it is consumer spending. And well, nobody calls me after I buy a pair of New Balance sneakers. So what's the experience for those folks? And gosh, there should be software that governs this. So we founded Klaviyo. And we started out saying, well, look, if we're going to replicate what CRM did where there was a human in the loop, and now we're going to have software to do it, we better give that software some human-like properties. For instance, it better be able to think and act. It better have a brain. Building out the data infrastructure, which we now call the Klaviyo Data Platform, that has the ability to store and ingest data, much like you might think like a data warehouse or data lake, but critically indexes all that data so it's available with a low-latency real-time API for applications that need it. The same way that we expect that, yeah, our brain better be able to do some computation so we can spit it out in the form of words or writing or what have you. We built that. Then as we got going, we got our first couple dozen customers. We asked them, OK, well, we feel like we've built a database, but databases are only really useful once you've got some applications on top. What application category should we work on first? And what our customers unanimously told us was, "Look, if you can help us apply this to marketing, that would be great." And so we started out Klaviyo, what we're known for up until two weeks ago, was really around marketing and messaging. And maybe the last thing I'll say as a kind of core to us as a business is we spent the first three years building all of this just the two of us. I know now it's kind of in vogue to say, "Hey, small teams can build sort of large companies or a lot of stuff." Well, we believed that back in 2012 and still very much do believe that. So it's been very exciting watching a lot of those technology advances. I think with that has instilled Klaviyo some really great properties of we really believe in being very product-led. We work very closely with our customers. We really believe in automating as much as possible inside of a business, and that's resulted, I think, in some really good unit economics and a good overall financial profile. Yeah, and you actually just reported your Q4 results where you were sustaining about 35% revenue growth, which is really rare today for a lot of software companies, particularly just given the macro uncertainty that we've seen weigh on other more SMB-exposed vendors. So can you just help us unpack what's driving the stickiness of the platform for customers and the more defensible budget for Klaviyo? Yeah. Well, first of all, I think we're proud of that, but I think we think we can go even faster than that and look, the reason Klaviyo has become so core to businesses is this idea of if I'm a business and I have all these consumers, and I know the happier they are, the more I understand them, the better experience I provide, the more they're likely to spend. I mean, that is kind of the fundamental that's the fundamentals of any business or enterprise. Klaviyo sits right in the middle of that. We store all that customer data. We pull it together and then critically, we action it first through our marketing applications and we can then measure the actual return. So we have a term we've coined internally, Klaviyo Attributed Value, basically the revenue that we can tie directly back to customers taking actions after seeing marketing or messaging coming from Klaviyo. Over the last 12 years, that's now over $150 billion. Last year it was well over $50 billion. I mean, it's just an enormous amount of economic impact. And so when every business is looking at budget, I think there's software that's not only, yeah, it's a nice to have, it's productivity, but it's hard to measure the ROI. For us and our customers, I mean, they literally get daily and weekly reports of the revenue that it's putting back in their pocket. And critically, they love it because it's very high-margin revenue. The typical ROI for a Klaviyo customer, it's not 3 or 5X. It could be 70 to 200X. So there's actually a lot of leverage there. Oftentimes, the conversations I have with customers is, how could I spend more? What more could I be doing? But anyways, we can talk about some of the product that we're working on to help do that as well. Great. Yeah. And so certainly, the KAV metric enables this really hard ROI. I do want to take a step back, though, and just get your view on the current macro. The company has referenced seeing some stabilization in the SMB segment. You've had some consistent demand from the low end, and also you're focused more on the high end as well. And I appreciate we're still early in the year, but are there any sort of changes in behavior that you're seeing thus far into 2025? And what's your broader macro outlook? Sure. So I mean, from our vantage point with SMBs, they're very resilient. And they're very focused on not only maintaining their business, but growing it. And what's interesting for us, it plays very much into Klaviyo's thesis. I think they're very focused on taking the existing assets they have, which are those existing customer relationships, and maximizing the value of that. One, because they know that it's more economically valuable. But two, it's something that they have already. It's something that makes them unique. A fact that we share, something we calculate every holiday season, is over the holidays leading from Thanksgiving through Christmas, how much of the revenue that businesses acquire is coming from new customers? They They have to go find net new demand versus comes from folks they already have a pre-existing relationship with. The last few years, that percentage has ticked up and up and up, where I think it was either last year or the year before, actually the majority of revenue from businesses in the busiest time, holiday season, came from existing relationships, and I think a lot of folks look at that as, hey, this asset that I'm building is sort of, it's the real enterprise value of my business is these existing customer relationships, and so they're investing harder in that. Something that you referenced earlier was a key differentiator for Klaviyo being the data-first approach. Can you just unpack the data-first approach and what it enables and how it differentiates Klaviyo from other marketing technology providers? Yeah. Well, just in general, I think we have a lot of software applications out there that I would say are pretty dumb. They don't really know what to do. They're tools. They're things where you need somebody to sit in front of a keyboard or on your phone, and you need to tell it, direct it how to go achieve some goal. A lot of software has grown up that way, whether it's PC era, internet era. I mean, we can change the format and how it's available. The reality is you still have to press the buttons to make it go do something. I think that era is coming to a close. I think we should expect our software to do more of the thinking for us. And so we talk about the shift from sort of dumb applications to smart applications and like, well, what's the difference? How do you get a piece of software to think? And the reality is it all starts with, well, you had better give it a data set that it can pull from, either to collect information, to make better decisions, or that it can iterate and learn from. And at Klaviyo, that's kind of the core of how we thought about building our products. The whole point of our data platform was, well, we're going to store all the data. Sure, we're going to allow you to query it and do compute, do thinking. But we're also going to give applications a feedback loop so that as they run their own experiments, they can learn from what's happening. So just as an example, with our marketing products, when a marketer designs a marketing campaign, they'll often create one or two pieces of creative. Now, with obviously generative AI, you can create sort of an infinite amount of creative. That's great. But how do I figure out how to match that up to different customers? And And based on what properties? Is it demographic data? Is it behavioral data? What we're able to do now with our smart applications, with marketing, we can say, look, here's 100 variations or 1,000 variations. Here's 100 or 1,000 different segmentations of these customers. Figure out exactly which content works for each person. Match it up and do it all automatically. That only works because that back-end data is readily available in real time. It can then literally run a test over the course of minutes or hours to figure out which content resonates with which person, assign those variables, and then go, and then distribute it at scale to millions and millions of end consumers. We're doing this, obviously, with messaging. We think we can do the same thing with customer service. We can do it with a mobile or web experience. I think this is the future of what we should expect from software, that it's doing this kind of computation. It's hooked into this kind of back-end. And so it's actually doing a lot of this work that normally a human would have to do, but instead now it's just done through software. Great. So you've built this data asset. You've been able to build these smart applications on top of it. You've actually started to push that farther, as you mentioned, not just going out within marketing, but also into customer service, which is a really exciting announcement that you just made recently. So just given your vision of redefining what B2C engagement looks like, how do you see Klaviyo's role evolving from initially being more of this marketing automation to a bigger, more comprehensive CRM solution for B2C businesses? Yeah. Well, this is all born out of when we started with marketing, it was not the only candidate application. Folks said, look, we're using the data we have to improve some part of the customer journey, the customer experience. We said, OK, well, we have to pick a place to start. And I think marketing was a great category, great application, because it was like, yeah, it was sort of at the top of the list. It's revenue driving. There's a lot of things you can do with that medium. But it wasn't the only one. And our point of view was, if a consumer business has their customers, we want to be the software that sits in between all of the experience that that consumer has. And obviously, a critical part that was missing was the reactive side. It's one thing to say, hey, here's some new products that we've launched. Hey, we've got this offer for you. But what happens on the flip side when somebody has a question? And what we're hearing from our customers was, well, what's the Klaviyo of that? What's the sort of data-first, highly personalized, does reinforcement learning version of that? And we said, gosh, we've actually got a bunch of great partners that are working on this. Maybe they can help us build it, but gosh, we better get going on this because this needs to exist. So our point of view is, for all the applications you use to deliver a cohesive customer experience, they should obviously all be backed by one database. And that seems pretty obvious. So that, we believe, should be our Klaviyo Data Platform. But then those applications, look, we need to take more of a leading role in helping build those out and helping people know what those should be so they all really stitch together. So much like when you think in a kind of a B2B context where you've got a sales team, we tend to think of, like, OK, I need sales automation. And there's probably a little bit of marketing that helps sales. There's a set of applications there. We said, well, what's the equivalent for a consumer business? So when I go buy this sweater from L.L.Bean, they have me surrounded. They can craft the experience, and all the software there runs through Klaviyo. So we looked at marketing was one, service was another, and then analytics was an obvious third. How do I understand who my customers are so I can generate more ideas? So that's what we launched a few days ago, is those three applications make up a consumer-oriented CRM. And what we've done for marketing, we launched our first product in customer service and our first product in analytics. And yeah, we're off to a good start. Yeah, and can you just help us understand, as you've been expanding the portfolio, how does the competitive landscape start to differ as you're eating away more and more of this opportunity? Yeah. Well, we've always had it. So look, we have a very high bar that consumers deserve awesome experiences. I think everybody here, as a consumer, can agree with that. And the businesses that want to serve them deserve software that makes it easy to do that. We're frankly disappointed with how much the software is not smart. And that needs to get leveled up because otherwise, I think, frankly, there's a lot of consumer businesses that are just going to fail because the bar that consumers have, frankly, because of a lot of large retailers, just keeps going up and up. So we need to provide the software to go make that possible. So we've taken a dual approach. One is we're a company full of builders. So hey, we can go build this. Part of the reason what's sort of encouraged us to do this is, with marketing, it was actually not our first inclination to go build marketing software ourselves. We actually tried to partner up with a bunch of other marketing softwares, but they just couldn't deliver on it. It's sort of like we gave them this brain, and they didn't know what to do with it. And we said, OK, well, that's not enough. We've got to go solve this. We don't want to make that same mistake. That probably delayed us a good year on the marketing side when we first launched. We don't want to make that same mistake with customer service. So we've gone on a path of building it, but we've also opened it up to a whole bunch of customer service applications and said, if you know what to do with our brain, then we want you to integrate with that. If you can make it so that the thing with customer service that I hate is the number of times that we end up on customer service situations where you're chatting with somebody, you're talking to somebody, and you have to repeat yourself. You have to tell them about the thing that they should obviously know. I was changing a flight that's coming up, and I was like, I had to call somebody. And I said, look, I just have to cancel this leg of it, and I just need to keep the rest. I mean, the number of times I had to go find confirmation codes and which flight and all this stuff. I'm like, yes, I know this already. You should know this too, but don't you literally have my picture sitting in front of you? Obviously, something's broken, right? Because the person there certainly wants to solve for that. So we need to level this up so that we never have to re-explain ourselves. That's our goal with customer service. You want to get something done, you can either do it yourself or when you need to ask for help, just like magic, somebody obviously knows everything about your context. And so we've seen this diversification of different use cases, marketing, service, now analytics. One of the other things that's also been diversifying is your customer base. In the beginning, it was a lot of retail, e-commerce, pulling from that Shopify partnership. You've also made partnerships in Toast or Zenoti or WooCommerce more recently. So how has the mix of the customer base started to shift beyond retail, e-commerce over the last couple of years? And what are some of the other verticals where you clearly are seeing this same data challenge and reflect a large opportunity to penetrate? Yeah. I mean, I think every sector of the consumer economy. I mean, we literally make a map of if you'll take the different components of inflation and things like this. We literally take the same kind of industry codes and say, OK, which of these industries would not benefit from a better consumer experience that's more personalized? We're measuring outcomes. It's more automated. Literally every single one of them. So we have to do some prioritization. We started in retail, frankly, because we, one, built some great partnerships with businesses, companies like Shopify, but also because we kind of polled around a bunch of our early customers, probably half of them went to retail. So we said, OK, well, there's clearly something here. What's interesting is a good chunk of the other ones. They all had some sort of service-oriented business. So what we nominally define as hospitality, but actually includes a lot of things, includes not just the places you might stay when you're on vacation, but also the places that you might go eat or even travel, like anything that really somebody's providing an experience. What's interesting is, rather than it being all about a transaction, in that case, it's usually about some moments. It's that concert you're going to. It's that yoga class you signed up for. It's that flight you're going to take, and I think in all of those situations, the model is actually very similar. Folks have preferences. They have tastes. You learn about those over time by collecting data, and then we can do the same thing with marketing, messaging, and service there, and so you mentioned some of the partnerships we've built there, Toast. We're also working with OpenTable, just to give you an example, within food and restaurant. And in those cases, I talked to a lot of small businesses and say, yeah, I have a lot of recurring customers. Actually, if you think about food service, people come back a lot. And yet a lot of those folks say, yeah, I might know them on site, but I don't know them digitally. And so there's a whole movement within the restaurant industry to say, hey, how can we provide a better experience where we know who our customers are? That's a highly recurring business. I mean, in some sense, retail is one where it's almost like it's a good fit for Klaviyo because you build up these sort of, what's the digital version of somebody and how can I use that to make their experience better? But you actually have a lot more data in some of these other industries. And so I think there's actually an even bigger opportunity. Folks are even more looking for a consumer CRM like what we're building. Yeah. And at the IPO, you guys clearly laid out four growth pillars for the business. And one of those was the expansion at market. And so I want to better understand, how does Klaviyo's value proposition look different to more complex, larger customers relative to the long tail of SMBs that you serve? And what is the most significant competitive advantage you have as you start to move more at market? Yeah. Well, it frankly gets better. So I mentioned that before we started Klaviyo, we'd worked with a handful of really iconic businesses. And what everybody was convinced was, hey, this data is really valuable. We just need to find out, make it easier for us to use. We started out with SMBs because we had realized, gosh, it takes a long time to get an enterprise. If we were going to bootstrap our business, boy, we better start with smaller businesses. But we had built Klaviyo to scale. And what's starting to really show now is, as we work with larger and larger brands, the amount of data they have and the diversity of customers and use cases really, I mean, the true capabilities of having all this data in one place and the flexibility of our marketing platform really start to show off. So just as an example, I was talking with a customer of ours that's an enterprise cosmetics company. They're launching five new brands this year that are adjacent, and they want to do all of this cross-selling and cross-promoting. With Klaviyo, they've been able to build models of who's likely to be a customer of these new brands. They've been able to build multichannel campaigns about affinities towards this kind of person that prefers a text message or an email. They're able to measure the ROI of that, and they're able to do it scalably across five different businesses. I think it's really starting to show off what Klaviyo can do really when you want to do it, not just for a smaller customer set, but a large customer set, and then the last thing is, I think folks are really realizing the enterprise, gosh, it's not just about marketing. The fact I have all this data in one place, we see a lot more folks building custom applications. So that same business has taken the data that's inside of Klaviyo. They've used our APIs. They've plumbed it into their web stack, their mobile stack. And now they're using the same source of truth to power all of these experiences. So they're super consistent. I mean, that's the kind of stuff that SMBs want to do if they obviously don't have the time or the budget. In enterprise, they say, well, boy, that's a must-have. It's this dream of an individualized customer experience for every single person that is completely consistent no matter where you go digitally or sometimes even visiting a physical store. And one of those other drivers is also expanding monetization. And you have been accelerating the pace of additional products, one of the key ones being SMS. And as of last year, about 18% of your customers were taking the SMS product. That's up about from 16% in the year ago period. When we think about kind of where customers are finding the most value with Klaviyo and SMS, where would that be? And then from the spending side, what type of uplift do you see when customers also are attaching SMS? And where can the mix trend over time? Yeah. So anytime you build a platform, a data platform, any kind of platform, you have to see there's a lot of applications that can go on top of it. And one of the things you have to ask is, like, hey, is it sort of like is there one sort of killer application, or is there actually a set of them? What's great about when we think about the customer experience space is, oh, there's a bunch of applications that matter. So within marketing, we have different products by marketing channel, so email, text messaging, et cetera. And then there's obviously, we think about beyond marketing, there's customer service and analytics. All of those are growth vectors for us. The first thing that we added was the additional marketing channels, so text messaging, chief among those. When we got into the text messaging space, there were a couple of players there. But the idea of this personalization measurement was still pretty nascent. So we've seen good growth and adoption of that. Not every retailer wants to use text messaging to reach their customers. Some people feel like it's kind of an odd experience. One of the things we had to do is actually educate folks that, hey, there's a lot of value here. Even I was probably a little bit of a skeptic to start off, like, well, do I really want to get text? I mean, texting is pretty personal. Well, the reality was there's some great use cases. And actually, the bar for personalization goes way up. The experience has to be even better. So we see good adoption there. I think that can go much higher. I think the opportunity to grow that. I think it could be north of 50%, I mean, certainly amongst our larger customers, and then I think that lays the trail for what we think looks like good adoption for our other product categories. If you think about every retail business, once you get to above any kind of reasonable size, everybody's doing customer service. Our analytics products, we found there's not a single person I talk to that says, "I want to understand where customers are, and I just want the ideas handed to me so I can action them," so we think the attach rates for all of these products should be very, very high. And because nobody's really tried to build the application set for consumer businesses that makes up the complete stack, we think there's a big opportunity to say, OK, great, you should use our applications for that. And if not us, then you should use one that folks that we partner up with that are built on top of our data platform has built. And so one of the things that makes Klaviyo very unique has been your efficiency in the go-to-market process. You've really leaned into an ecosystem-led approach. And kind of the most well-known partnership here is certainly Shopify. So I wanted to kind of dig into where you've looked at the Shopify relationship as a model for adding additional partnerships. How has the top of funnel evolved as you've been adding on more of these partners? Yeah. So in general, our belief on what's the best way to get to customers, well, the best way to do it is let your existing customers and your partners do the talking for you. Nothing is going to help your reputation versus having other people vouch. So we talk about having peers, people's friends. When they talk to them, are they talking about Klaviyo? Are they recommending it? We talk about our marketing agency partners, service partners, SIs. Are they saying, yeah, we deliver service. We love Klaviyo. That's what we default to because it's easy to run. It's very composable. And then obviously, there's our platform partners, other technology companies that are defining their categories. And hey, we work well together. So obviously, we love working with Shopify. It's been a partnership we've had for, I mean, gosh, almost the entire history of Klaviyo. And what made it great is what it was founded on, which was, do we have shared goals with our mutual customer? Every business Shopify and Klaviyo serves wants to grow. They want to be more successful. And we started with, hey, how are we doing that through the products that we're collectively building together? I think that was the foundation of, let's build something that's accretive to customers. Customers then voted with their feet and said, yeah, yeah, this Klaviyo Shopify stack is a really good one. And then we layered on top of that, frankly, more of the go-to-market, the co-marketing, co-selling kind of activities. And that, we think, is the right way to do partnerships. Sometimes I think people work them backwards where they, let's figure out how are we going to divide up the pie? And let's start selling it. And let's figure out on the back end, what's the actual customer experience? The reality is that's not good for the customer. It can waste a lot of time where two puzzle pieces don't really fit. I think we did it the right way. And you mentioned some of the other relationships we built. We announced the partnership with WooCommerce as kind of another commerce platform that we're excited to work with. You mentioned Zenoti, Olo, Toast, those other verticals. We think the work that we've done with Shopify, there's a lot more to do there. But I think we can take that, especially to other use cases. And I think this idea of, let's build a tight product integration that's beneficial to a customer, and then let's figure out how we evangelize that world, that's a great way to do customer acquisition. Great. I'm going to ask another question, and then we'll open it up to the audience if there's any questions, but obviously, I have to get some generative AI questions in, and so it's really around, where does generative AI fit in the product roadmap? I think you've done a really great job of integrating things like with product recommendations, predicting next best actions, but what are some of the capabilities with AI that you're really prioritizing into 2025? Yeah. So one of the things we believe is that the future of these smart applications is they need the ability to run themselves. If you think about, let's take marketing as an example, the idea of a marketer at a small business, even at a large company, as having mined through, looked through the entire data set that they have, having networked with all of their peers in industry to figure out all the best, latest trends and what's working, the idea that they can come up with the exact right copy or content or graphics, I mean, it's just, of course, that's not going to happen. So look, we have artificial intelligence and data-backed algorithms to help us solve this. The most natural application of generative AI to Klaviyo was literally in the content production process. You know, we had folks that say, like, gosh, I don't know how to write great copy. I'm writing a text message or I'm writing an email subject line. I don't know what works. Well, here's good news. We have hundreds of thousands of Klaviyo customers who have sent collectively millions of different marketing campaigns. We've trained models that will just say, hey, here's the kinds of content that works best for your audience. That's an immediate leg up. You give us kind of the seed, the concept. We'll help you make it better. And obviously, you can overrule along the way. I think that's a basic kind of step one. What we're actually even more excited for is that helps sort of automate more of the creation process. But we now are starting to say, hey, can we play with agents and other AI algorithms to literally define your marketing strategy for you? So you can imagine prompting an AI and giving it all the context that's related to a specific Klaviyo customer, maybe some general trends that we've seen, and say, develop a marketing strategy for this business, and then go implement it using Klaviyo APIs. That's possible. And for a lot of small businesses, they'd say, well, that's exactly what I want. And can I have creative control? Can I sort of thumbs up, thumbs down, ask it to make some edits? Yes, absolutely. So this idea of marketing and frankly, CRM as autonomous in the future, I think is going to happen very quickly. so we have a lot of stuff planned for this year that basically puts us on this path of how do we make playing the game of Klaviyo, doing marketing, something that requires less decision-making on behalf of our users. Awesome. Do we have any questions in the audience? Are you the only one that is developing this consumer CRM platform in the way you are? What competition do you face? Yeah, it's a great question. So is anybody else doing a consumer or B2C CRM? And I would argue it's like, look, I would say yes. And I'll tell you why. First is 99% of when you see CRM, I look and say, like, OK, is it predominantly for sales teams? So that's a different category altogether. And then for the folks that are doing anything that's around the B2C customer relationships or experience, none of them have invested the way we have in the underlying brain, the data platform that powers all of it. And our point of view is you cannot just build the applications. They're not going to be smart otherwise. You have to invest in the thing that's going to make them intelligent so they can run themselves. And that is, by definition for us, a consumer that has to be software-driven. Great. Do we have any more? One more over here. The mic's coming around. On your example on OpenTable, where if OpenTable knows more about the customer, what's an example of what would be different if they know the customer outside of OpenTable versus I show up, they give me my table, I'm happy, I go, I review the restaurant? Oh, sure. Well, so one of the things OpenTable does, other than just helping make reservations, is help give a little bit of some data back to that business. So our integration with OpenTable allows the data that business collects on who's coming to dine, when they visit. And sometimes they can combine that with restaurant, effectively point of sale data on what dishes diners actually consume. They can use all that data to build a profile of when somebody likes to visit a restaurant and sort of intuit some of the occasions. So an example of something that I'm actually very passionate about that we're working on a model for is the canonical problem of you're going to go have dinner with a friend of yours or say a significant other or family member. How do you decide where to go and then what that meal should be? We think in the future, restaurants can do a better job of basically predicting. Let's say for my co-founder and I, we have a tradition of every year around Klaviyo's anniversary, we go grab dinner together, and the reality is we always do it at one of a handful of restaurants because those are the restaurants we used to frequent when we were just starting out. There's no reason for that restaurant to not be able to know, hey, Andrew and Ed always come back here. We should remind them a few weeks in advance so we're not late booking a reservation that this is something that they should go do. That sounds fairly basic, but for a lot of restaurants, that's really hard to do, and we can make that something that's out of the box. Awesome. So we are actually up on time. But AB, thank you so much for joining us today. And we're really looking forward to everything in 2025. All right. Thanks.