Great. Thanks everyone for joining us. My name is Mike Sicos. I'm the lead analyst here covering Elastic. With me, I'm pleased to say we have a fireside scheduled with Elastic's management, Interim CFO Eric Prengel, as well as, I know Anthony Luscri and Janice Oh from the IR team are here with us as well. We have 40 minutes set aside for this fireside, but I just want to flag to the folks who are sitting in the audience, there should be some sort of Q&A function on your side. Please, if you have questions, your questions are way more interesting than mine, but make sure you're getting those in, and we'll do our best to get to it, while we have Eric here. With that out of the way, Eric and the team, thank you again for the participation. We really do appreciate it here in Needham.
I think super basic, easiest question you'll get all day, but for people who are maybe newer to the name or dusting it off, just a quick refresher. Who is Elastic? What do they do?
Yeah, at a high level, Elastic is a broad search platform, and that search platform can be used for three different functionalities. We've got the core search, where you can use Elastic in your enterprise for search. We can power the search bar that you might have either in enterprise, on a website, etc. Elastic can also be used for security, where we predominantly play in the SIEM space, and then we can also be used in observability, where we play across all the different capabilities in observability. But historically, a lot of our strength is in logging. The one thing I'd add is that Elastic has evolved over the course of the last two years. As generative AI has become a bigger and bigger part of the ecosystem, our core search functionalities are really critical to generative AI.
Well before it was becoming part of the everyday vernacular of software investors, and before ChatGPT became big, we'd been working on vector database capabilities, on vectorization. And so, for us now, generative AI is becoming a much bigger part of what we do, and that really plays across all three of our capabilities. If you think about security, we're infusing AI across everything we do in security with our, you know, a lot of the capabilities that we have, the same in observability, and that's interesting. It makes us more competitive in those spaces. But where it's really most compelling for us is that when you think about the search capabilities, we really are now having a whole expansion to our TAM as people are building their generative AI applications on top of our search capabilities, and that's a really compelling and exciting thing for us.
So we're getting more competitive because of the generative AI that's infused in our platform in both security and observability, but in search, it's really a massive TAM expansion for us as you think about what that creates for us as an opportunity.
Excellent. Excellent. And two things before we go into it as well. Interesting, and I appreciate you calling out vector at the start. Just a quick soundbite from, like, probably a year ago. There was a private database vendor that I was talking to when the market was still going through this debate of whether or not vector needed to be a standalone offering versus maybe more of a feature or capability as part of another broader offering, and the private company, no skin in the game, not necessarily a competitor of Elastic, but volunteered on their own. They're just like, "Hey, everyone's trying to featurize RAG's capability." Elastic's had vector now for a handful of years, so kudos to them for being able to see that. Good on you guys for going to where the puck was going before it was actually there.
Thanks for putting that out there, 'cause it's something that we've been saying, but it's nice to hear it coming from someone else. The other thing I'd say is that a year, a year and a half ago, I'd say that, you know, there was a lot of noise in the market. People didn't know who was, you know, gonna be winning share. And now more and more, I think it's, you're hearing third parties as you talk to some of our customers, as you talk to people in the marketplace. I think that Elastic's name is coming up more and more, and there's, it's clear that there's a lot of demand for what we're doing and that we have a differentiated solution as you think about the vector database functionality and all of the other capabilities that are used to build a generative AI application.
That's great. And I know I definitely wanna hit on the technology while we have you here. I think one thing, again, in the interest of just setting the table for the audience, I know the October quarter earnings feel like forever ago, but there were some interesting things that I think are probably worth highlighting. So maybe for a quick 30-second pitch, like, what were some of the highlights? I know, like, third consecutive quarter where you guys cited that search continues to accelerate now, but what else would you be pointing investors to from that quarter?
Yeah, at a high level, what I'd say is that Q2 was a really strong quarter for us, and it was driven by solid sales execution. We saw a lot of customer commitments come in. We exceeded guidance on revenue and profitability, and Q1, obviously, you are aware, we had some issues with our customer commitments and our go-to-market, and I'm sure we'll talk about that later, so I won't belabor the point, but Q2, we really saw that get back on track, which was great. We saw key wins across all of our solution areas, so where search continued to be the part of our business that grew the fastest, we also saw really strong wins in observability and security, so that was really positive.
We talked about this a little bit on the call, but those new commitments with generative AI almost doubled in dollar volume compared to what we saw in Q1. We actually had three deals that were over $1 million from a, on a commitment basis for generative AI, and that was hugely positive and a really strong trajectory for us, so we were very excited about that. We're really seeing a lot of increasing commitments that are tied to generative AI and that are tied to people wanting to bring our platform on to use generative AI or to use generative AI on our platform that they were already using. While we're seeing a lot of these commitments and we're seeing a lot of trajectory, it still is early days from a revenue perspective, and that's gonna take time to ramp along with these commitments.
Revenue's gonna trail it a little bit. Generative AI is clearly the fastest growing part of our business, and we think that, you know, as it continues to grow, as you see more of those commitments, you'll see revenue in the future years continue to have a bigger and bigger role associated with it. We also saw, in addition to generative AI, we saw customers continue to consolidate onto the Elastic platform on both security and observability, so we saw a nice momentum in those businesses, and we continue to be focused on displacing incumbents and really migrating people over onto our platform. The consumption trends in the second quarter were positive. You know, both enterprise and commercial generally did what we hoped they would do, and we actually saw a little bit stronger than expected consumption from some of our larger customers, which was obviously a huge positive.
So overall, I'd say that Q2 really was a reaffirmation of our strategy, that our go-to-market is functioning the way we'd want it to, and our hope and expectation and what we're seeing in the business is that we're well on our way to getting back to the pace of sales execution that we've demonstrated in the past. So a very positive quarter.
It's great 'cause I know that you guys will also talk about customers, whether it's as a mark on the total customer base or it's a mark on your higher spend cohort, who is actually using Elastic from a GenAI perspective?
Yeah.
I'm curious, when you're having customers sign these commitments, are we at a point now where we can say, like, "Hey, Sicos was gonna come in and sign a commitment for $100." If GenAI's attached to that, that commitment now is maybe above 10 or above 20, like, are we getting to that, or can we pinpoint it to that level of granularity?
Yeah.
Or not necessarily?
No, it's a great question, and it's something that we look at a lot. So as you think about the commitments that we're getting, people, we don't have a separate SKU for generative AI, so when people are buying our products, they're not buying, you know, X amount of search, X amount of, observability. They're buying ECUs, which they can use across our platform, and so we don't have a separate SKU where we can measure the dollars that are coming from GenAI specifically. What's really interesting, though, is that we're seeing so much more momentum, so much more interest around deals that are tied to generative AI, and that's really where we're seeing some traction, and we're seeing a lot of positivity. And to a degree, it's not even that it's a 10% uplift or 20% uplift.
It's that deals that we might not have otherwise seen happen could be happening because of generative AI and because we're getting that engagement with, even if they're not doing you buying us purely for generative AI. There's a level of engagement, a level of interest on upleveling and who we're getting access to in their organization that's helping us move deals forward, in addition to all of the generative AI interest that we're seeing.
That's great to hear. And from a demand perspective, again, I know we're all into the new year here. I think people are trying to get a sense of what you're hearing from customers, how are conversations tracking. I mean, obviously, there's a lot of enthusiasm behind GenAI, but if I compare where we are now versus where we were a year ago, is there a reason to be, maybe more optimistic? Or are you hearing anything on budgets, as an example? Can you comment on that?
Yeah, no, I can. I wouldn't say that there's been a change in the macro as you think about the last couple of months. There's certainly nothing that we're willing to commit to at this date. It feels like it's remained stable. We're seeing customers remain cost-conscious, and they really wanna get value as they're deploying their dollars. To your point, there is a lot of interest in generative AI, and while we're not seeing IT budgets growing as a whole, what we are seeing is that there could be some shifts or some specific carve-outs for AI within existing budgets. So it's not that the budgets are growing, but people are more focused on generative AI in terms of where they're deploying their spend.
And so as we think about the acceleration in search, and we talked about this a little bit, that really is a generative AI-driven thing for us. And so that remains positive. One thing that we've talked about a little bit is the SMB segment, and so far we haven't really seen that come back. That part of our business, which is what we call monthly cloud, and every quarter we report on where that is, it's a percentage of our total business.
That has been much flat, more flat, a little bit flattish is how I'd describe it, and we disclose that each quarter, and at a high, it was something like 17% of our total revenue, and it's come down to be more like 13% of our total revenue, where you've seen the subscription revenue that we have outside of the SMB monthly cloud business grow really nicely. That SMB monthly cloud has been a lot flatter in our business.
Just to make sure we're doing our proper diligence here, that flattish-type behavior we're seeing out of SMB, and maybe we can weave this into a broader conversation, but is that in any way related to competition or new entrants or change in win rates? Or is that really just a function of potentially the greater macro sensitivity that those SMB-type organizations have?
Yeah, the way we think about it is, I think, is much more of the macro and the sensitivity that the SMB has. As we talk to some of our peer organizations who have similar go-to-market motions, I don't think that they've, I know that they haven't seen something different than what we're seeing. They tend to say that pay-as-you-go or the SMB portion of their business is a little bit slower to recover than some of the more enterprise side, and I think what we're seeing is really consistent with what others have been seeing, and we've seen it really since some of the changes that happened in the economy a couple of years back now, and it's just been consistent since then. As you think about the competition, we haven't really seen any changes in the competitive environment, you know, across the board.
We think there's an ongoing opportunity for us, and we think that we can continue to gain share. As you think about our win rates, they haven't really changed much. It's been pretty consistent, and if you think about it from a solutions perspective, in security, it feels like SIEM is evolving to this 3.0 version that's gonna be a lot more AI-based, and we think that given all of what we do in AI, that's a place that really favors us and that we have an opportunity. And we've come to market with something called Attack Discovery, which is a product that we have where we leverage LLMs to analyze the alerts and do a lot of triage work. So what a SOC analyst used to have to do and was a very manual process, they can now use large language models and our platform to do.
And we've really gotten tremendous positive response around that, and it, it's something that we think is a big opportunity for us. In observability, it feels like customers are demanding a lot of these similar AI-based innovations, and observability, obviously, there's been a move over multiple quarters and even multiple years away from these siloed tools and more in favor of this end-to-end observability platform with a multi-signal approach, and I think that we're at the forefront of that trend. And then in search, we talked about this quite a bit, but that's where we really see ourselves an established leader in the space. And over the last couple of quarters, there's something that our founder, Shay, said at a meeting that gets quoted internally at Elastic a ton, and he says, "Search has a budget again." And it really is true.
As you think about the opportunity with generative AI and how so many companies are so focused on this, search does have a budget, and search is a market that we're really strong in. And so this interest in generative AI, coupled with our differentiated technology, has been hugely positive for us and has helped us grow our TAM, and we've seen that search business really accelerate nicely.
If I could just continue on that search piece, right? So if you have a customer that comes to you and maybe they wouldn't have come to you otherwise, right? Forget the ASP, potential ASP uplift for a second, but like, are you hearing anecdotally from customers as far as the ROI that they're achieving, or what kind of, are they citing to you specific gains or enhancements that they wouldn't have gotten otherwise? Do you have a better feel or anything to quantify what the customers are receiving from that perspective?
Yeah, in terms of ROI, it's a tough question because a lot of things are still early days.
Mm-hmm.
We have some customers, and what they've said to us is they've kind of seen, you know, a massive decrease in the hours that are needed from some of their, you know, employees in terms of their ability to service, you know, like the call centers, chats, that they've been more efficient. They've been able to do more. We've got one customer, it's a large bank, and they're using us for some of their wealth management capabilities where they're enabling their wealth management employees to engage with customers, to have insight with customers in a different way than they had previously, and there's a ton of value in that. So it's not just that they're spending less, it's also that they're able to do more, and that's creating this tremendous opportunity as you think about what our customers can get, and I think it's early days still.
This whole Generative AI phenomenon is, you know, still in the works on an early days basis, but as we think about it evolving, as we think about the opportunities that some of our customers have talked to us about some use cases that they can see, and even though they're preliminary, some of the use case that they already have in place, there's definitely an ROI that can be really strong with the capabilities.
And for those customers as well, is there anything notable, as far as what kind of companies are coming to the platform? Like, in any way, is the customer profile changing as you see? Again, it's like self-selecting, right? They're telling you they're coming for GenAI, but like, based on that, are they more tech-focused or are they more manufacturing background? Is there anything vertical specific or size employee base, or it's really just across the gamut?
It's a great question. It feels like just geographically, and, and I know this isn't the question you asked, but I'll give it to you anyway, and maybe it'll be interesting, and then I'll keep going.
Please.
It does feel like America is ahead of the Americas are ahead of the broader geographies in terms of their adoption of generative AI. They're a couple quarters ahead, and that's not atypical for the Americas to be a little bit ahead of some of the other regions when you think about technology adoption. It does feel like they're a little bit ahead, and even the public sector in the U.S. is probably a little bit behind the private sector in terms of adoption. Interestingly, we are seeing large financial institutions spend a lot of time on this, and different institutions are adopting things at different rates, similar to the way you'd think about the cloud, and so we'll see how that all plays out.
But it doesn't feel yet that there's one industry that we've seen really spring ahead of others in, you know, some of the customer-facing, you know, digital markets. You know, if you think about a consumer-facing, you know, company, they might be doing some things to help with the search on their website where you'd be going from traditional search to semantic search, so they're making things a little easier in that respect. But we're also seeing financial institutions think of a lot of different use cases across the board, you know, whether it be some of the things we talked about earlier with that wealth management use case to, you know, fraud detection or what have you.
It really does feel across the board, Mike, and maybe as we get further along, you know, a couple quarters from now, I'll say, "Oh, this industry is using it more than this other industry," but right now it feels pretty broad-based.
Understood. Okay. And I, again, you guys have a couple different solutions here and a couple different vectors, so I don't wanna over-index the search, even though it is an exciting part of the story for sure. If we just look at security and observability for a second, right? There have been some announcements from other vendors on the competitive front, and would be interested in hearing how has that competitive landscape changed for Elastic more recently?
You know, I'd say broadly there hasn't been a huge change to the competitive landscape. The consolidation trends that are in play today have been going on for a while, and to a degree it started with when the economy really shifted. I think that with some of the consolidation, we've been more focused on it, and we've, you know, we're continuing to pull on that thread, but it's been here for a while. As you think about our strength, it's being able to ingest unstructured data from any source, you know, in observability. It's logging and security. It's SIEM.
But, you know, what we're seeing, we talked about this a little bit before, is that AI is really transforming both of those spaces, and I think that if you think about our business, it's three solutions built on top of a core platform, and our core platform is what we, where a lot of the engineering work happens, and so a lot of our core platform is being infused with this generative AI DNA, and with our company being very focused on generative AI in terms of the platform and the search opportunity, that plays into how our products are evolving in both security and observability.
And our platform getting that AI, those AI capabilities in it really lends itself to the strength that we're seeing in those other portions of the business, where we think that even though the TAM might not necessarily be growing for those two businesses, we think that we do have an opportunity to win more in security and to win more in observability than we might have before. And we talked a little bit about Attack Discovery, where you can automate those investigations and triage, and that really came about because of all the time that we're spending in generative AI and all the capabilities we have there.
We've been investing in features like that, in features like, you know, our AI assistants that are in observability and security, the ES|QL, express migrations, a lot of things that we're doing that where we're really leveraging AI to make it easier for customers to adopt Elastic and to get more value out of us, and that really makes us more competitive in those two spaces.
Awesome. And if I just think out loud, and again, you guys have had this competitive dynamic for some time, but again, just given some of the changes out there, the characters that we're up against, is there a reason to think that the pipeline building that's going on for security and observability in any way, like, I would imagine that it's accelerating in some way based on that competitive dynamic, but is that fair? Is that still too early? Need more evidence building?
You know, we've continued to build pipeline. We're super focused on it. It's an important thing for us. I'd say that, you know, it's probably a little. It's still early days to say that there's been a massive change. I think that we've built pipeline nicely. We're happy with what we've been doing with pipeline in the last, you know, since Q1. We obviously had some issues with execution. We've put a lot more emphasis and rigor into how we think about building pipeline. We have a weekly call that we go through where we work through the pipeline, not only for Q3, obviously, which we're deep into, but for Q4 and even into Q1.
and there's a focus on that that's even more intense than what we had prior to Q1, and I think that that's helped us not only build pipeline, but make sure that we're building the right type of pipeline and that the pipeline that we're building is moving through, is progressing through the pipeline in the right way, and I'd say to bring that back to your question about, you know, security and observability. I'd say that we're seeing it build in those spaces as well as others, as well as search, obviously, and that we talk a lot about search. There's a lot of excitement for search, but observability and security are key parts of our business, and we continue to be focused on building pipeline there.
Our goal is that in security and observability, we continue to grow the business really nicely, and we view ourselves as a market share taker in those businesses.
Got it. And I wanna shift to the tech for a second, just because it really does feel like over the last year, the number of product announcements or enhancements, seemingly just, it just feels like it's accelerated, right? And so, as we exited calendar 2024, go back to November, December timeframe, there were a couple of different announcements that I wanted to hit on. The first is, the company had announced Elasticsearch LogsDB index mode. That went GA in mid-December. And just as a reminder for folks, what are the anticipated benefits for customers from this LogsDB index mode?
Before I get to that, I just wanna say something else. You said, you know, we've been making a bunch of product announcements. An investor once said to me, he said, "Eric, we probably see more press releases from you around product announcements than we do from most of your peer companies." And, you know, I'd say yes, we've been very focused on technology and on innovation, and hopefully that shows up in terms of some of the things that we're adding to our platform, some of the things that we're doing to differentiate ourselves. So, and he was like, "No, it's a good thing, definitely." So just wanted to call that out since you mentioned it. It pulled on that string in my mind. So back to the LogsDB point.
What we've done with LogsDB is we've really used it to optimize storage and query performance for log data. As you think about log data, it's really the foundation to visibility into the IT environment. And these growing log volumes are generated, you know, by infrastructure and applications, and that's creating more costs and forcing customers to limit how much they collect and retain. And they want to retain more. They want to collect more, but at some point they have to make a choice between what is the value to retaining these incremental logs for this additional time and what is the value of collecting these additional logs in our system. And they have to make a value choice. Is it worth the dollars to do that?
With LogsDB, we intend to address this challenge by reducing the costs to retain their logs in their system, so obviously we wanna be the leading data store for logging data, and I think with LogsDB we make that even more clear as we're gonna reduce the storage footprint significantly to enable customers to store and retain just much more data and enable them to get more meaningful insights by having that data in their system for a longer amount of time, and as you think about it, the full capability of LogsDB is something that we're gonna hold off for the enterprise tier, and what we've seen from customers is that, you know, as we provide them with features like this, it's gonna help them upgrade to enterprise because they're gonna get more value out of it.
And then once they've upgraded to enterprise, they're gonna be able to put more data into our platform and get more value out of our platform. So it really plays to the benefit of our customers in a lot of ways.
Terrific. And similarly, the Elastic Rerank model, that was announced in December, similar, dynamic, but can you just walk us through what's the expected benefit then from this new reranking model?
As you think about the reranking model, there are a lot of things that the reranking model does in conjunction with RAG and a lot of the other things we do for generative AI. And so we've brought a lot of different capabilities to market around that. We've got RAG. We're working on ways that we can have inference services to bring more generative AI capabilities directly into our platform to sort of enable large language model usage directly in our platform. We've got ELSER.
The way we think about it with the reranking model, with all these different functionalities, what we're trying to do is give our customers really more capabilities that they can use to build out their generative AI capabilities and just put more tools into the Elastic toolbelt that'll help them get value out of our platform and use in different ways and use us more efficiently and have their models really work better for them because of the technology that we're putting in their hands.
Nice. And I apologize for the background noise. The radiator in the New York City office has gone a little crazy today between that and the camera issues you were having earlier. We make quite the couple right now.
You know what's funny, Mike? I got into the office early today, and there was literally somebody with a chainsaw cutting something out of our roof. I was in pretty early, so it wasn't during the regular workday, but it wasn't a chainsaw. It was some sort of a mechanical chainsaw-like piece of equipment, but they were cutting some wiring in our roof. And I was like, "Oh God, I hope this doesn't go on all day." And it didn't, thankfully. It was done pretty early, but I luckily Zoom doesn't pick up too much of that sound.
Excellent. And just the last product announcement that I wanted to hit on specifically from this, like, December flurry of press releases we got. But, on the security front, you guys had announced these expanded Cloud Detection and Response capabilities, all coming from the single SIEM. And so what is it these new capabilities are delivering? And then is there a way to think of, is this picking up spend that had previously been going to another competitor's tool? How do we start to compartmentalize where that incremental dollar is coming from?
Yeah. The way that I think about it, as we build out all these capabilities, you know, Attack Discovery, the CDR capabilities, what we're really trying to do is have more of an end-to-end solution. We've had a really strong platform around SIEM for a long time, and we're clearly one of the leading vendors in that space. And we've got a big
All right. Awesome. Sorry about that, everyone. We're just making sure that we got some good audio going for you for the conversation. But Eric, if you could just bounce back into that question again. We were talking about the CDR capabilities.
Yeah. So as I was saying around that, you know, as we think about CDR, as we think about Attack Discovery, as we think about all of the capabilities that we've brought to the table, the goal for us there is really to have this end-to-end security solution or broader security solution that with all the strength that we have in SIEM to enable us to bring more to the table and to do more the same way that we talked about this on the generative AI capabilities. We're trying to enable our customers to get more value out of our platform. That is the same goal as what we're doing with security. And Mike, am I, are you still able to hear me or am I breaking up again?
I can. No, no, no. You're perfect now.
Okay. Let me know if I'm breaking up. Sorry about that.
Good. Trust me, we will. All right. Just going back to the AI question, I guess when you're talking with customers, and again, they're hopefully communicating to you those the ROI and those productivity gains that they're talking to. But curious to hear, are those AI applications that people are currently deploying, or are you seeing an increasing volume of those moving to production? And then the follow-up is, are most of those applications today still internal facing, or are we starting to see a greater volume of those AI-based applications become more external facing?
Yeah. I, I think we're still in the relatively early stages of seeing external facing applications being built and deployed. We mentioned it previously, but it feels like these AI applications have largely been internal facing. There are some that are external facing that are starting to be deployed, but it's generally early days. You know, what we're seeing in the area of Gen AI is that we're still in the first phase, and the most natural thing is for our customers to go from sort of traditional search to semantic search as part of that step one. And then, so we're seeing a lot of semantic search gaining traction.
And the way I think about that is if you had a website and it used to be, you know, you had to click certain things, "I'm looking for this." Now you'd be able to say things more along the lines. You'd be able to look for some specific, like, "Hey, I want a pink rabbit with a blue this," you know, like a little bit more context that gives you the search. You're able to say something and get a search response for that. And after that semantic search is when customers move into more of these chatbot-like features, and that's really where it's more of a RAG application. And a lot of those tend to still be internal facing as people are concerned around the risks with these large language models and the potential hallucinations and compliance issues associated with them.
So right now I'd say it's still more on the internal facing side, but we're seeing that the potential for that to migrate.
Great. And I do wanna be true to my word. Again, if folks have questions, feel free to lob them in. I know that we do have one here. Just gonna read it off to you now. So the question goes, how do margins vary across search, security, and observability? What EBITDA margins can you ultimately achieve?
Yeah. As we think about search, security, and observability, I don't think that we necessarily would break out a separate margin for them. We sell it all as one platform. There's not a separate SKU across those businesses. And so people are buying ECUs and using those ECUs against whatever solutions they might have in their ecosystem. There are three main use cases that they might use them for, being search, security, and observability, but there's not a different SKU underlying those use cases that's creating variance in the margin. So the way I think about it is there's not necessarily a different margin across those three businesses. You asked the question about EBITDA.
I won't go into EBITDA specifically and some sort of a target margin type thing, but as we think about just the direction of our business over the last five years, we've seen our business go from a negative operating margin to the midpoint of our guidance this year is, you know, in that low to mid teens, and so we've clearly made a lot of progress on the margin of our business. We continue to make progress on that margin while we grow the business. So if you think about how we're oriented, our business is focused on growing and growing profitably, and that's the direction that I think you'll see our business continue to be going in.
Great. And I know we have just a couple more minutes here. Wanna make sure that I'm tackling the go-to-market as a final topic. But, earlier this year, Elastic had called out a shortfall in customer commitments. Maybe you can kind of walk us through the dynamics that caused that shortfall in the first place. And then obviously we have the October quarter under our belt now. Does Elastic feel like it's rectified maybe some of those earlier concerns? Where are we in course correcting?
Yeah. It's a great question. So in Q1, we had issues with our go-to-market. What that was really driven by was sales territory realignment and some changes in our account coverage model. It took time for the account coverage to get settled with some of that, you know, disruption. And in Q2, we were really pleased with our sales execution. We were really glad to see the team bounce back and that the additional measures that we took following Q1 were having exactly the effect that we'd hoped for. Sales reps have had time with their new accounts to engage, and we're in Q1 at the beginning, they might not have been figuring out who to call and, you know, for different things. Now they know who to call, and they can really focus on progressing deals forward. So our win rates remain strong.
In Q2, we saw really strong pipeline creation. We saw pipeline progression more at the pace that we've seen historically, and, you know, we're very happy with that, and we're tracking these deals, and they're moving forward, and that's incredibly positive. Both Ash and I, over the last, since the start of Q2, have spent a lot of time engaged with the field team, in addition to our CRO, Mark Dodds, who's just done a fantastic job, you know, transitioning that business, and so, you know, as we think of the time that we spend every week, we have a forecast call that takes up a couple hours on Monday morning, and then we go through pipeline as well. So there's a ton of time that we spend on it, and we're really pleased with the progress we've made so far on the execution.
And as we think about the goal, it's really to continue to build on that positive momentum that we saw in Q2. And so, you know, we think we're happy with our strategy. We're happy with our competitive product positioning and our ability to really get that sales execution back on track.
Great. And I guess the follow-up here, I think you guys have alluded to it, but just to make sure that investors are aware of that, that initial shortfall that occurred in Q1, I believe has been discussed as being something that will kind of carry with Elastic or play out through the financials over the course of the year. Just as a reminder for folks, why is that? And does it become more magnified as we move through the year, or does it become more of a dissipating effect, as we move through the year?
Yeah. So the way I think about it, there's obviously two types of business that we have. We've got the self-managed business, and we've got the cloud business. The self-managed business, we get a commitment, and then you recognize that revenue ratably over the course of the year. And so any shortfall that you have in Q1 would be felt in revenue, not in Q1 predominantly, but predominantly in subsequent quarters where you would've normally been recognizing the revenue from that commitment. As you think about the cloud business, it's a little bit different because you get a commitment upfront, and instead of just recognizing that commitment ratably, you recognize the commitment, the revenue associated with that commitment based on what customer consumption is.
And so as you think about the way that this typically ramps, you'd have a customer who'd make a commitment in a quarter, and in the subsequent quarter, they would start to ramp. And in the first quarter after commitment is the slowest amount of the ramp and the lowest amount of revenue that you get from them. And then in the quarter after that, and then the quarter after that is where you'd really see the revenue continue to ramp. And so if you think about it, this is something that would create a headwind not only to the quarter immediately following the commitment not coming through, but even more so to the quarters that come after that, be it two or three quarters after in terms of that commitment shortfall. And this isn't something, a dynamic that's new.
This is something that we've been aware of, and that's why we adjusted our guidance down on that Q1 call. And, you know, in Q2, we saw healthy consumption trends, particularly with some of our larger customers, and that was positive. But this is obviously gonna be something that, you know, we're watching out for with the lower commitments and just how we think the rest of the year is gonna go. But to be very clear, there hasn't been any sort of degradation in our business. We're just assuming that because of the Q1 lack of commitments, that's gonna impact consumption over the rest of the year.
Understood. Thank you for carving that out for everyone, and I know with that, we're at time. So thank you to Eric for joining, Anthony and Janice, you as well for the participation, and thank you to the audience. Eric, thank you so much.
Thank you, Mike. And I apologize for the technical issues. I'm glad you were here to get through it with me. It was a pleasure.
We'll do it again soon, but hopefully, we'll get through the connectivity in a different way, but thank you so much.
Thank you. Look forward to it. All right. Talk to you later. Bye.