My name is Koji Ikeda. I'm a software analyst here at Bank of America. Thanks, everybody, for hanging out for the first day of our BofA tech conference. I am absolutely thrilled to have Janesh Moorjani, CFO and COO of Elastic. Thank you so much for being here. Thanks for having us, Koji. Of course. Great to be here. Of course, of course. So standard question, you know, for those on the webcast, for those in the room that are unfamiliar with Elastic, just very, very high level, what is Elastic? What do you guys do? What is the opportunity? And maybe just a minute on yourself. Yeah, happy to. So Elastic is a search AI company, and, you know, fundamentally, we help people find answers from all of their data, in real time, at scale. That's what we do. You can think of us as a search AI platform. The main product is the Elastic Stack, which is a very popular product among the developer community. The company was started in 2012. One of the co-founders and the initial author of the product was a gentleman called Shay Banon, who is still the Chief Technology Officer of the company. And the company is today about a little over 3,000 people. We just closed out our fiscal 2024 and reported a little over $1.2 billion in revenue. It's been a fantastic ride over the last 10 years. In terms of my own background, I've been at the company for approaching 7 years now, so it's been, it's been quite a while. You know, I've seen the company grow quite nicely and lots of new faces, and lots of new customers that we've added along, along the way as well, and it's been fantastic to be a part of that journey. So you guys just reported your results last week. Yeah. so super excited to have you guys here today. And may, maybe talk about from a very high level, some of the key takeaways. Yeah From the earnings report. I think a little bit more importantly is, how have your discussions with investors been post, and maybe the top one or two questions coming from them, you know, following the earnings? Yeah, I think... So we had a great Q4, and it was a great way to cap off the year. So we printed a fairly strong growth across our total revenue, across our cloud revenue. We beat on all of the metrics that we normally guide to in terms of total revenue, as well as profitability. We had very strong customer growth in the, in the categories that we measure. And so it was a good, clean, healthy print all around, and I think sets us up very nicely for the fiscal year 25 that we just started. Within the overall numbers, a couple of the highlights that we focused on were around the customers that are using us for GenAI workloads, particularly using our vector search functionality and RAG functionality and all of that in the cloud. For a few quarters now, we've been talking about a couple hundred customers incrementally that we've been adding for some time, and we finally crossed a 1,000-customer mark on that front, so that was really encouraging. The other interesting data point was, if I look at some of our larger customers that spend more than $100,000 a year with us in ACV, we reported that more than 145 of those customers are now using us for our vector search and RAG capabilities, and that number is triple what it was a year ago. So again, great momentum on that front. And so it was really encouraging and a great way to cap off the year. In terms of what people are thinking about and some of the questions we've been getting, you know, have been focused a little bit on just trying to understand the momentum that we've got on the GenAI side, as well as just trying to get a read of what the broader market environment looks like and how that might color our outlook in terms of the guidance that we provided. So those have been some of the areas of focus, but you know, happy to talk about those, if you like. Yeah, let's, let's focus on that guidance for a minute. Sure. You guys, Straight to it. Fourth quarter, right? And so it's been an interesting, I would say, first half of this year, all things put together and- Yeah And here you guys have to go out and give guidance for a full year out from now. And so just walk us through kind of the components of the guidance. How are you guys thinking about it? New versus expand- Yeah Demand environment, all that stuff put together, and how does that come up to the number that you gave? Yeah, happy to. So if I maybe just step back and talk a little bit about our approach to guidance as a starting matter. Fundamentally, our guidance philosophy has not changed. The philosophy that we used in fiscal 2024 served us quite nicely through the year. We started the year prudently. We then worked very hard over the course of the year to try and outperform against the outlook that we had provided. And recognizing to your point, Koji, that there's, you know, a lot of things that looking out four quarters out, it's hard to predict, and so it makes sense for us to take a little bit of a prudent approach at the start of the year. And that's the same approach that we've taken. If I think about the underlying components within that guidance, if I think about the enterprise and commercial segments where we have most of our sales energy focus, our direct sales motion is primarily focused on enterprise customers and commercial customers. I think that's been working quite nicely for us over the past several quarters now. It's great to see some of the evolution in that and some of the work that Mark Dodds and his leadership team have been driving. I think we're all super excited about that. And so within the enterprise and commercial segments, if I look at it in terms of our self-managed business versus our cloud business, both of those have had good momentum. The self-managed business has grown quite nicely over the past couple of years. And, so we expect that momentum in fiscal 2025 will continue in self-managed and also in the annual cloud selling motion that we have. So we would expect to see good, healthy growth across both of those formats. And then if I think about the SMB segment, which we service primarily through our monthly self-service motion, it's a lower cost of way in which we service that part of the market, and SMB has been macro impacted for some time, and we continue to see that over the course of fiscal 2024 as well, in terms of the underlying growth in that business. And so we've continued to assume that SMB will continue to be challenged in fiscal 2025, and we've assumed that in our expectations of how we believe the monthly cloud business will perform as well. And because of that pressure, that pulls the overall growth rate in cloud down a little bit. So those are the ways we've sort of thought about the guidance and you start with Q1, we'll execute and then update everyone as we go. I want to ask you a question on demand. Yeah. The demand environment, how are you guys thinking about it? What are you guys seeing? Because it's been interesting in software with a lot of different- To say the least. Companies, categories, all, all of it put together. You know, I think it's easy to bucket front office, back office infrastructure, but now I think investors overall are really beginning to dig in. What is it specifically about this part of the aspect of front office or back office, and in your world, infrastructure, right? So, we thought you had a pretty good quarter, right? We had a great quarter, yeah. Great quarter, great quarter. And so what is it about the demand environment that you're seeing out there? How would you categorize it today versus six months ago, versus a year ago? How much have things changed, or maybe they haven't changed at all? Yeah, it's a great question, and maybe I'll, you know, I'll rewind a little bit because you asked about the comparisons versus six months ago, a year ago, and so forth. So, back when we started to first see the impact of the macro environment hit everybody in the industry, in sort of the late calendar 2022 timeframe is when it started to impact us. At that point, we had started to see there was all of this talk in the industry, and we experienced it too, in terms of consumption, optimization, headwinds, and so forth. And, you know, as that got to a more normal level, for the past several quarters, at least the past, you know, four quarters or so, we've been saying that, look, we are seeing broad the demand environment broadly is stable, right? And it's not easy. It's not that customers have started spending like crazy again. In fact, if anything, customers are extremely budget conscious, extremely focused on driving costs out, remaining focused on operational efficiency. Cost consciousness as a theme continues to be the case, and we've seen that for several quarters now. And, and that actually plays to our advantage because it helps customers, and we help customers and lean into that and help them understand the benefits that they can get by consolidating workloads onto the Elastic Platform. And so that actually helps with the platform consolidation theme, that is one that we've been talking about for several quarters now. So nothing really changed on that front. I'd say Q4 was very similar to, to what we had experienced earlier. It was very similar throughout the course of the quarter across February, March, and April. For us, as customers continue to look at ways in which they can maximize the value of the investments that they're making, that plays naturally to our advantage, particularly in some of the use cases around observability and security, as I mentioned, because it drives that platform consolidation theme. I think the other way I think about it is that customers are increasing their spend in Gen AI, and if you look at most CIO surveys in the aggregate, total IT spending is increasing not at a very high rate. It's you know, probably low single digits in many of these surveys. So you think about that shift in the composition of spending. If the shift is going towards Gen AI, it's coming out of somewhere else, and different companies optimize their budgets and manage their budgets differently. But as investments shift towards Gen AI, we stand to benefit from that, and in observability and security, because we are helping customers drive costs out of the system and get great innovation while they are doing that. That's playing to our advantage as well. So we felt pretty good about how Q4 turned out. You mentioned late calendar 2022, and so when I think about software companies as a group and how they reported and how they talked about it, that feels pretty early, right? When you guys are actually calling it out, we are seeing some softness. And now, you've been talking about things stabilizing, even maybe getting a little bit better out there. Stable is good, right? And so it feels like you're first to see demand get better. And so why is that? Why is it something specific about what you're selling into, like the end market or the opportunity? And what would, in your view, need to happen for things to get incrementally worse out there from what you're saying? Yeah. So maybe just to step back. In 2022, I don't think we saw it first, and I don't think we are seeing a recovery first. In 2022, software was impacted broadly, and we saw it just like many other players from the industry saw it. And you would remember that time when people started to talk about how they're optimizing consumption and so forth. So for us, what we've just seen is this continuing pattern where people have optimized as much as they can and want to, and they've turned the dials to the left as much as they can and as much as they want to. And it's not like they are turning the dials all the way back to the right again. It's not that they are relaxing the constraints that they had put in place. So when we say it's stable, it sort of reached that level, people are cost-conscious, and they are focused on operational efficiency. Because we help them achieve that efficiency and drive them, that actually acts as a net benefit to us in our business. I don't think we will see those demand signals earlier than other people will. One of the things that we learned over the course of the last two years is obviously the importance of staying very close to our customers, and we made big investments in the customer success function as an example, and that's the team that works with customers to drive consumption, helps them get the best return on the investments that they've already made with Elastic, even if that means helping them understand how they can use their Elastic deployments better and more efficiently. So those are all things that we've done, and I think what we see is general stability at this point in time. Okay. Okay. I wanted to dig in a little bit about these data points that you gave on the last quarter about customers using generative AI. You got 1,000 plus- Yeah. overall paid, right? Maybe, yeah. Then 145 of the 100K+ ACV customers. Are those use cases, you know, trying things out, or are they actual production? Meaning, these companies have chosen Elastic as core technology to drive some sort of digital experience. It's not, we're not trying something new, let's see what happens, and this is a paid customer. This is, "No, we have chosen Elastic. This is how we're going. This is gonna drive, you know, this is a key piece for the next five to 10 years. Yeah, I think so it'll naturally be mixed because customers are at different stages of their journey on GenAI, and some customers are further along than others are. But to your point, these are paying customers, and they are customers that have selected Elastic to essentially be their partner of choice for building GenAI applications. And if you think about the life cycle of what happens, a customer who wants to build a GenAI application will first decide what their technology stack is, and that becomes an important part of the engagement that we work with customers on to essentially be selected, and that's a technical conversation, and it's our presence in the community and our strength with our relationships with developers, and the strength of the technology itself counts for a lot in terms of us winning those situations and essentially being selected for by the customers to build their GenAI applications. Then they build the application, then they have to, you know, go ahead and obviously test it, build integrations, build dashboards, migrate data. Then the deployments happen, then the deployments scale over time, and the usage ramps, and that's when we recognize revenue in a consumption-based model. So the important thing for us is if you think about that entire life cycle, the revenue and the scaling happens towards the end, but you don't get any revenue or scaling if you've not been selected and been part of the stack to begin with. And so when we talk about these 1,000-plus customers, they are paying customers. What was really an encouraging data point for us was that some of our larger customers are now starting to deploy us even more, and that's a really good indication of the overall momentum and adoption. And then, if I think about which of those are, you know, in production, as, as you put it, and, and which are the ones that are larger names out there that, that people would recognize, there's a number of those. If you think about our ElasticON tour stops, and I know you've personally attended, ElasticON, which is our user conference, and we do it in many cities around the world, throughout the year. In almost every ElasticON tour stop, you will see some customer from that, from that region or from that city, that actually talks about how they are using us for GenAI. We've seen examples ranging from Cisco in San Francisco that is using us for a customer support use case, all the way to OCBC Bank, which is one of the largest Asian banks headquartered in Singapore, using us. And lots of different types of customer profiles and use cases in the middle. We've seen small customers, we've seen large customers. There are many that we're not allowed to name because they are large enterprises that wouldn't give us permission to name them, and we referenced some of them on the earnings call, including large banks, the large electronics manufacturers. So, we see lots of positive traction. Yeah. But it's still very early days because if you think about the revenue model, as I said, the revenue is gonna come at the back end as the deployments happen and as the deployments scale. So from all of those customers, we did not see a meaningful revenue contribution in fiscal 2024, nor are we anticipating a meaningful revenue contribution in fiscal 2025. There'll be some growth, it'll just be natural growth, but we're not modeling any big inflection in revenue in fiscal 2025. We're very convinced of the long-term opportunity here, and we're investing towards that at this point. I'm gonna dig in a little bit more here. Please. Yeah. I really like the 145 customer metric, the 100K plus. You know, these are big companies, big budgets, making some very strategic decisions, and one of the aspects I really like about Elastic is you're very ubiquitous within a lot of enterprises. I mean, people know Elastic. Whether it's paid version, free version, you guys are littered all over digital experience, ecosystems out there. And so on one side, it's, hey, it's very easy to say: I want to try something new with Elastic, and let's just move in that direction. On the other side, and I think more importantly, is with these bigger customers, they're making very strategic decisions, RFPs, trying other technologies out there. And so the question really becomes: How many of the deals out there are you seeing you're winning in some sort of RFP? Meaning, this business has been a 10-plus year whatever customer with Elastic, maybe they're trying some other technologies out there, but at the end of the day, they're like: No, we're gonna stick with you guys. Yeah. So most of the wins will end up being technical wins, right? So if you think about RFPs or competitive bake-offs... This is not about, you know, somebody writing a formal RFP document and then saying, like, "Let me, which of these boxes does Elastic check?" And going through some sort of, I'll use the word bureaucratic procurement process. These wins happen with the technology and winning, with developers. In every, in every large enterprise or smaller or small or medium-sized business, as customers are making these choices on which technology partners to go with, they're actually picking companies based on the technical strengths that they offer, as well as their familiarity with the technology, how well the company plays in the ecosystem- How well they integrate with others. From the technology standpoint, it's not just about one set of features. You know, we think about vector databases as fundamentally a feature of the overall stack. So it's not just about one feature, but it's about the overall stack and how well does that work with traditional lexical search approaches. Can you drive costs out by running hybrid search? Can you personalize the search experience? Does it respect my security and privacy and compliance postures that I've had in the enterprise for a long time? Are my developers familiar with the technology? So, there's so many technical questions that need to get addressed, as we think about a big part of why we get selected, because we've been in the enterprise. We've built up over years and years a lot of the capabilities that enterprise buyers are looking for, enterprise technical buyers are looking for in the search platform. To your point, we're ubiquitous, almost when it comes to search, and one of the most, you know, popular products ever from the standpoint of being open source or free and open. And so there is a lot of familiarity out there, and people just, you know, they find the product just incredible. So when you have vector search functionality that is baked in, and the performance of vector search is incredible when you compare it in any of the third-party benchmark tests and so forth, that's a big part of why we get picked. I want to ask you one more question on AI, then just- Easy. Open up the Q&A to the floor to see if there's any questions from the investors in the audience. So just last question for me, AI. Just thinking about all these customers that you sell into, and, and let's just focus on the biggest ones out there. Sure. Have they gotten really comfortable with generative AI, meaning they, they know what to do, and now it's time to execute, or, or are they still really just trying to figure it all out still? What are you seeing out there? Yeah, I, I think maybe just to start with, as I think about the broader GenAI market opportunity- Yeah. ... I think the overall opportunity for GenAI is gonna be bigger than most people expect. It's also gonna take longer than most people expect, and it will actually manifest itself in ways in which a lot of people don't expect yet. So I think as an industry, as a whole, there's a lot still to come in GenAI. So when you have large enterprises that are starting to use GenAI, I don't think anybody has fully realized what the true long-term potential of GenAI can be just yet, and therefore, there's a lot more to come on that front. But within that journey, I think there's a lot of people that are a little bit further advanced in terms of how they are using GenAI in production use cases. There are some people that are still in the earlier stages of using GenAI, and I think it's just, you know, overall still very, very early, in terms of the maturation of the overall market opportunity. So we think this is, this is gonna play out over the next several years, and we feel very well positioned in that. Got it. Thank you. Any questions from the audience? Please raise your hand. No, no questions. No questions. Everyone's convinced. I want to ask you a question about vector databases- Please. and vector search. Calendar 2023, it was vector databases, what's gonna happen to Elastic? And then that narrative kind of slowed down a little bit, but it feels like it's picking back up a little bit again. I definitely get the question more today than I did six months ago. I mean, similar to what I was doing last year. And so it really comes down to what I think is important here is performance of vector databases. I think that... And you mentioned that you think of it as a feature. It does feel like if you're gonna play in this opportunity, you're gonna have to have vector anyway, of some shape or form. And so now it becomes: how does Elastic think about performance and scalability of vector? Maybe help shape what is it specifically about the technology that enables that, and how do you protect that for the next, gosh, five to 10 years? Yeah. Yeah. You know, it's a great question, and I think over the years, we've had different kinds of FUD thrown at us and at different points in time related to different parts of the technology. And, you know, and this is, well, no different. But if, if you look at fundamentally the data, and you look at any of the third-party tests around performance, and you will find that our vector database performance is, is among the best out there, in terms of just the ability to scale to, you know, massive levels of scale, if you're thinking about whether you're thinking about number of dimensions, whether you're thinking about vectors. For example, one of the names that we mentioned on the most recent earnings call was a company called Roboflow. And if you look at their website, you'll get a sense of the scale at which Roboflow operates. And so we, we've not had any constraints or issues as we think about engaging with customers. In fact, in many instances, customers have told us that you know, they've, as they've looked at Elastic and considered standalone vector database companies out there or other technologies out there, that fundamentally, you know, vector database functionality is excellent compared to the to anyone else's. And then you superimpose on that all the other needs that they have for the enterprise that we were just talking about, and Elastic was a clear choice for them. So we feel very good from a technology perspective around where we stand. In terms of the long-term differentiation, I think one of the biggest elements of the long-term differentiation for us just is around our overall back end. If you think how Elastic Stack was built, we did not start with any presupposition of what kind of data would be stored in Elasticsearch, which means that it's built with incredible flexibility and incredible scale in mind, and you can bring any kind of data, structured or unstructured. If you're trying to correlate different things that you never envisioned you would bring into a data store, you can bring those in. And that actually plays so well from the standpoint of everything that you need to do for semantic search. And that's a big part of the reason why I think we'll continue to be differentiated very nicely in the long run. Switching topics a little bit- Yeah. Moving to Elastic Cloud, one of the, you know, very hotly debated parts of the Elastic business, for sure. And so- It's all clear. So, I wanted to ask you a question of how these contracts kind of work. Yeah, especially in infrastructure, software, and consumption, as we look at all these different types of businesses, I think one of the aspects that people are really trying to figure out is: how exactly is that user spending with Elastic? How does that contract work? And you've said in the past, we've talked about it before, you guys have a very good understanding and visibility with these commitment contracts. And so maybe walk us through a typical contract. Size really doesn't matter, but how does that work with the commitment? You know, how long does it take for revenue to show up? I mean, is it a big project? Is it not a big project? What does that look like? Yeah, and maybe to start with, you know, think about the selling motions that we have in enterprise and commercial, relative to, the monthly cloud motion, which is predominantly SMB. So we've said for some time that we would much rather have customers that are higher-quality customers with a greater propensity to grow, and that's where we expend our, customer acquisition, energy and focus on acquiring customers that will become bigger customers over time for us. And, so we've tended not to focus on the total customer count as much as we focus on some of the larger customer metrics. And you've seen that 100K+ ACV customer count tick up quite nicely as well. It's been relatively, stable, I would say. And even if you drop the threshold, we also talk about statistics in customers that are more than $10,000 in ACV, which if you think about it, is less than $1,000 a month, and even above that threshold has been quite stable. So I think over the years, we've done a good job in the past couple of years of making sure that we are focusing our energy where it matters and not worried about the long tail and aggregate customer count. And within that, when a salesperson engages with a customer and secures a commitment from them, our contracts are typically one-year contracts. Many of them are three-year contracts as well, so you can have some multi-year business in there. And, even for the multi-year contracts, we are invoicing the customer only annually, so it's not like we, we invoice the customer everything. Generally, we'll just invoice annually at the start of every year. And then the customer starts their consumption, and if it's a newer workload, if they need to plan their migrations and so forth, if they need to bring data in from multiple sources, gradually, that consumption can take some time to ramp up, but then eventually it ramps to the level that the customer was expecting. I think one of the other distinctions in terms of our business models relative to some of the other folks out there, is that we are not out there trying to convince a customer to make a large commitment and then search for what workloads to bring in. Usually, the context of the conversation is very workload-specific. It's, it's almost project-based. So we're engaging with them on specific projects and specific workloads, and as they make the decision to bring those workloads onto Elastic, they are sizing what that might look like. And so we have a reasonably good view of how much that-- or how big that project will be, and the customer has a reasonably good view of how much they need to spend on Elastic and what the contract size needs to be. The other piece is that we generally don't penalize customers for going over, right? Which is to say that if they are consuming faster than what they expected and they've essentially depleted the credits that they purchased, they can always buy more credits, and usually, the higher the volume of credits that you buy, the better the pricing that you get. So they aren't penalized for consuming more in that sense. And as a result of that, customers generally tend not to overcommit. And so that's also worked out quite nicely for us because when we get the commitments, they provide pretty good visibility, and customers generally have not overcommitted for the most part, as I said, and it's worked out quite nicely for us. Got it. We got a couple of minutes left. I do wanna sneak in two questions with you. Just- Fire away. From a customer perspective, how big can a customer theoretically get with you? Just thinking about generative AI in the future. Could there be a time in the future where we're hearing about eight-figure deals, nine-figure deals with Elastic? I mean, is that possible with, you know, everything that's going on? We have some eight-figure deals even today on a TCV basis. Okay. Last question for you, Janesh. Thanks so much for doing this. Yeah. You, you are the CFO and Chief Operating Officer, so lots of responsibilities for you. And I think about Elastic and very attractively positioned in the crazy opportunity here with generative AI, but you guys are also profitable, too. And so how do you think about balancing, maximizing your potential within this opportunity, but also generating free cash flow? Yeah, at the end of the day, Koji, we still view ourselves as a growth company. If I think about the markets in which we operate, they are massive. The GenAI opportunity in front of us is, you know, very significant opportunity. And the product set that we've got, if I think about product market fit, it's just incredible. The serverless offering that we are launching in GA later this calendar year also will meet customer needs really, really well. We are confident of that based on some of the early results that we've seen. So we feel very good about the market opportunity. We feel very good about our product set. We feel very good about our sales execution over the course of the past several quarters, as you've seen us consistently deliver good results every quarter. From my perspective, I think investing towards capturing that future growth is the right answer. That's why we've said that fiscal 2025 will be a year of investment a little bit. We're still going to expand op margins, but we are reinvesting some portion of that natural operating leverage that is inherent in the business model back into the business, and then we manage that as we go. And if we're able to drive growth and realize some overperformance, then we'll keep fueling the engine. And if we find that we're not, then obviously we need to balance growth against profitability, which is something that we've done before. And we feel very good about our ability to manage that trade-off if we need to. Fundamentally, you know, I don't want the reason that we don't grow to be lack of investment. Yeah. And particularly in a business where we can course-correct pretty quickly and modulate the level of investment as needed pretty quickly, I think the right thing for us to do is to invest towards capturing that growth, and then realizing the future growth potential that we so clearly see in front of us. So we're super excited about that. Got it. Thank you. We are all out of time, Janesh. This was fun. Yeah. Thanks for doing this. Thank you for having us again. Super appreciate it. Thank you. Yeah. Of course.