Elastic N.V. (ESTC) Earnings Call Transcript & Summary

Good afternoon, everyone. I'm Brad Reback with the Stifel research software team. Thanks for joining us. As we wrap up day 2 here, presenting for Elastic is Janesh Moorjani, CFO; and we also have Anthony Luscri, Head of IR. Gentlemen, thanks very much for joining us.

Yes. Thank you for having us, Brad. Much appreciated.

And just one housekeeping note for everyone listening. If you have any questions, please submit them in the chat. I will moderate anything that comes through. I'd love to make this as interactive as possible. So Janesh, maybe for people newer to the story. Obviously, you guys have been out public for a while. So there may not be a ton and we probably don't need a lot of time spent. But maybe you can just help us understand what your core value proposition is as you sort of think about the installed base and where you're going.

Yes, happy to. So Elastic is a search company, and search is so much more than just text box. It takes all kinds of different forms. And essentially, the idea that you can read large volumes of data, index that and then return near-real-time results on that when searched, it finds application in many different use cases and many different solutions. We focus on 3 solutions, primarily what we call Enterprise Search, which is -- think of it as search on a website or search in an application; as well as Workplace Search, which is the traditional domain of thinking about enterprise content. So that's broadly an Enterprise Search solution. The second big bucket for us is our Observability solution, which combines logs, APM and monitoring information. At the end of the day, when you think about that individual whose job it is to keep the user experience or customer experience as rich as they can, they're constantly searching for information in the underlying machine data that they have. And that's -- in our minds, that's just another form of search, and we've been incredibly successful with Observability as a core solution. And the third one is Security, which, again, if you think about just what security analysts and threat hunters do, they are searching for anomalies and searching for patterns and searching for bad actors and large volumes of data. And so that plays very well to our technology from a search standpoint. And so we focus on those 3 main solutions. The company is just about 9 years old. And from the standpoint of go to market, we have a presence throughout the world. We have a free and open adoption model that serves as a very easy on-ramp for customers to start adopting the technology. And as of Q4, we have a cloud business that's got a run rate of over $200 million, where customers are using our service directly from our hosted Elastic Cloud offering. So that's a quick recap, Brad.

That's great. On Enterprise Search, Janesh, obviously, as you mentioned, you've been at it for 9 years here. There have been search functions on websites for a long time. How should we think about the opportunity in that part of the core of the business?

Yes. So first off, I think it requires us to think differently about search. It's not just the text and the text box. That will always be powerful. There's nothing as liberating as saying, "Okay, I need to find something. So I'm going to go look for authenticated results with speed and relevance." It's just -- it's incredibly powerful in terms of what we can do for our customers. But search takes many other forms as well. If you think about -- many different applications today use map-based search. So when you fire up an application, you're initially geolocated within the 4 corners of your screen. And then there are different things that you can search for around you, whether that's restaurants or whether that's rideshare. There are -- different applications have different purposes. Those are -- all those example -- different examples of search. Or if you are searching within a histogram for data, that's just another form of search. Filters, sliders, those are all just examples of search. And so we think about Site Search and Enterprise Search as -- excuse me, Site Search and App Search as one big bucket. That also follows e-commerce applications. And many of the online e-commerce applications today where you're searching for something on a vendor's website, many of those searches are powered by Elastic. And then the other piece for us within that is Workplace Search, where if you think about traditional enterprise content and searching for enterprise content, the world has shifted now where so much of enterprise data doesn't reside within the 4 corners of the enterprise or within the 4 walls of the enterprise's data center. It sits in different SaaS providers. And to be able to search across all of those SaaS assets with the security and with the privileges, information that attaches to each rule across those, I think that's incredibly powerful for a lot of our customers, and that's where we are seeing a lot of the shift in terms of customers adopting our new Workplace Search solutions.

And competitively, who's the alternative there?

It's a great question. It's -- I can't think of a single company that could beat, first off, in terms of just our overall breadth. But specifically within Enterprise Search, even there, I'd say just given what -- everything we can do in App Search, Site Search and Workplace Search, I don't see a single name pop up. From time to time, there will be point from our competitors, many of which have been around for a very long time and have been around for decades in some cases and really haven't had much of an impact for us. So as I think about that Enterprise Search business, we tend to get used in so many ways, sometimes even without us knowing about it, almost feels like we're the default choice there.

Got it. It sure feels that way.

Yes.

Just wanted to make sure. Now that being said, as we shift to Observability, that seems to be a market that's a lot more crowded but also rapidly emerging and growing extraordinarily quickly. So as you think about your differentiation there, what are some of the key highlights? And why do you win?

Yes. I think for us, again, it sort of goes back to first principles, which is an incredibly clarifying way for us to think about problems. We come at it from a lens of search, and these are very, very large data sets and any user, any practitioner who wants to have incredibly quick results when they look at those underlying data sets. If you think about how all the pillars of Observability came up, they all sort of came up at different points in time, championed historically by different vendors, all of whom were trying to solve point problems for that practitioner. But the idea that you can have all of that data stored in a single data store and you can have frictionless adoption and seamless extension from certain features around logging to other features around APM to other features around monitoring, that's incredibly powerful. So for us, I think there's 2 pieces to it. One is that the data is all in one place. It's a unified -- it's a truly unified stack that is natively built for the purposes of these kinds of analyses rather than trying to cover together a few things from different feature sets. And then you superimpose on that a unified pricing model that is based on resources so you don't have to think about how much do I pay per host and then how much do I pay per user and then how much do I pay per gigabyte of data flowing through the pipe, which is per ingest. You don't think about all of those because you don't need to anymore. So I think it becomes, again, incredibly simplifying and powerful for the customer. So that plays very nicely to our advantage as we work through this in the marketplace.

And on that unified pricing model, it would appear from a simplistic outsider looking in that, that is fairly aggressive versus some of your peers. Does that -- do you feel that gives you longer-term leverage on the pricing front? Or is it really just all about acquiring as many customers as possible right now?

Yes. I think a bit more in terms of does it put more power in the hands of customers or does it end up causing customers to think about -- just get surprised with high bills that they weren't expecting. And because it's resource-based pricing, from our perspective, it allows customers to control their spending a lot better because they are the ones that decide what resources to apply against different problems. They're the ones that decide when they want to ramp up or when they want to ramp down. In many instances, we will come out cheaper than competitors across many of these areas. But to be clear, while that's great and it's something we'll happily take, our goal is not to be a low-cost provider necessarily. In many instances, we focus on delivering pretty strong value to customers. And customers are willing to pay for that value. In some instances, it might be more expensive than the others, depending on architectural choices that customers make, because they see the value and the benefits of what we bring to the table. So that's the way I think about it. It's primarily about customer choice.

Got it. And then maybe wrapping up on the solutions side with Security, obviously, you bought Endgame probably, what, almost 2 years ago at this point and have continued to build out some of your own SIEM functionality. How should we think about the maturity of the product set and the ongoing maturation of it as it relates to winning additional business?

Yes. I think first off, the addition of Endgame to the Elastic stack has gone extraordinarily well. As you mentioned, we bought it almost 2 years ago and now -- it was in October of 2019, I think, that we closed the transaction. And from day 1 when we bought it, we folded endpoints into the Elastic stack and we did away with endpoint pricing. So when a customer buys the Elastic subscriptions, when they buy the enterprise tier, which is where the endpoints are, that technology is, they don't have to pay separately for endpoints. And again, that was a very powerful message to customers that the value is in the underlying data. It's not in the endpoint itself. And so why don't you pay for the value that you're using and deriving from it. And that became an incredibly powerful addition to the portfolio. I'd say from a technical standpoint, the integration has gone exceptionally well. Endgame already used Elastic under the hood, and there was very strong DNA across the companies. So from a technical standpoint, we hit -- historically, we've hit all of the milestones that we had laid out, ahead of schedule in some cases. And almost the whole team is still here and has done really well to build the entire portfolio for endpoint and SIEM. And so if I then think about it in terms of just customer reception and traction that we've gotten, again, that's been exceptional. When we acquired Endgame -- or even before that, a few years ago, if I think back to around the time of the IPO, Security was very small. We didn't even talk about Security externally as a significant solution. And over the last 3 years, including the acquisition of Endgame a couple of years ago, Security is now roughly 20% of the business. So I think it's gone really well both from a technology and a go-to-market perspective.

Great. Janesh, you talked earlier about unified pricing model. I guess somewhat related, you've also had a recent change in how you -- with your open-source licensing?

Yes.

So I think I know the answer to this but just want to make sure. The change in open-source license, did that impact your pricing at all? And can you remind us the different tiers of how the product is priced? And sort of what's open source and what isn't?

Yes, happy to. So first off, the licensing changes were completely unrelated to the pricing changes. The pricing changes that we made were because we launched a very powerful feature called searchable snapshots, which we put into the enterprise tier. And because that allowed customers to reduce their third-party spending significantly, we took the opportunity to share some amount of that value and increase the price. And so it was only one deal on which we raised the price. And the enterprise tier is a relatively small portion of the overall business. And so it truly had no meaningful impact on the results in Q4, especially when you consider there's ratable revenue recognition in the model. So that piece is sort of isolated and separate from the licensing change. But back to the licensing change, when -- a few years ago, back in 2018, we've already made changes. We started to think about expanding our basic subscription tier. And if you think about what folks -- when you say open source, folks traditionally think about 4 pieces. One is that if the code is open and they can inspect it and play with it and so forth. The second is that it's free, that they don't have to pay for it. The third is that it's governed by a very rich interaction model that they have with the developers that wrote the code and the community around it. And the fourth is that it's covered by an open-source license around distribution rights and so forth. And 3 years ago, we said why don't we give our users the benefit -- all the benefits of that, and we moved them to an Elastic license 3 years ago, where the default distribution became that basic subscription, which is proprietary to Elastic but it is free. The code is open, and they have the same benefits of the interaction model with the developers that we have and the folks that commit code. And so over the last 3 years, we expanded that basic subscription tier quite significantly because it became our default distribution model. So when we made the licensing change here in January, it didn't affect the vast majority of our users. They were all using the default distribution anyway. And it didn't affect any of our customers. So the change was really targeted at making sure that we can competitively ensure that people don't take the efforts that we put into building great software products and then go monetize those without getting back to the community or to us. And so that was the main motivation behind that. And as I think about the change, when we made the change, we already had now a 3-year advantage in terms of putting proprietary feature sets into that basic subscription tier. And then we've got significant investment in terms of engineering. And you see that in terms of feature velocity and so forth that we have. So when we made the change, we were on version 7.10 back in January. We're already on version 7.13 right now. And we've continued to do really well. I think the model has worked really well for us at this point. I think looking ahead, there's obviously the question of what folks like Amazon do because they forked the code. And from our perspective, I think that's playing out exactly as we expected it would be, expected that reaction from them, as you might imagine, and that's exactly what happened. The great thing is they're also renaming the product, their product, and they are calling it OpenSearch so that now, there's no market confusion between which company has which products, and this is just a completely different offering that is the starting point for which is technology that was only in open source. So they're missing all the solution features around -- all the capabilities we have. We have so much that we put out there by way of APM and SIEM and low-level features in the stack over the last few years. They don't have any of that, and you can see that in some of the products and offerings. So from our standpoint, it's actually gone really well. I think this plays out competitively over time because as I said, it doesn't affect our existing installed base at all. And over time, it will just help us attract the right workloads and more customers.

Got it. I got a question from an investor, and it's something we have heard over the years, maybe less so recently. But customer feedback has been such that the Elastic product historically had been harder to use for non-deep developer, IT people versus some of your peers. What have you done over the last few years to help close what I'll call some of that usability gap?

Yes. So a couple of things. One is there was probably some fun in that and some truth mixed into it as well because if you think about what our philosophy is around product releases and the way we traditionally launch products, it's a good thing to contrast it with what some other companies do, right? So typically, when some companies will announce a product, they will build a product and try and make it as feature-complete as they can and then put it out there for their entire customer base to use. And our approach as being different. And SIEM is probably a great example of that. When we decided to develop SIEM features and capabilities, very quickly, we will develop those capabilities in the span of some months and then put them out there. And those will be foundational capabilities that will be needed for that particular solution, again, just continuing down the same example, foundational capabilities for SIEM. And then over time, we improve on that with additional features, additional capabilities, ease of use, things that would appeal to a broader community of users. And our community continuously informs our own road map. APM is another great example of that. When we acquired the foundation of our APM capabilities, we put that in the open space for everyone to adopt. We very quickly put -- built and put APM capabilities out there, and then we iterated on that over several releases. And if you were to go back and look at all of our release announcements, you'll see that with every iteration, APM got better. So I think some of that feedback is just a reflection of our release cadence and release methodology, which, at the end of the day, has served us really well. It's an incredibly powerful business model to have that bottom-up adoption. And some of it, I think, is just dependent on the duration that these solutions have been out there. So you wouldn't, for example, hear that feedback on anyone using us for Enterprise Search. You probably wouldn't hear that anymore for logging that -- you probably used to hear that for logging 4 or 5 years ago, but now you don't hear that for logging anymore. And look, I'm a nontechnical person myself, but just for -- to understand what an actual user goes through, I tried spinning up a trial on our own Elastic Cloud and bringing in some data as a potential -- for a potential logging use case, and even I was able to do it. So that shows you that the bar isn't very high.

Well -- and I know we spoke about that previously after the earnings call, but just pulling on that thread as it relates to SaaS. So I'm assuming that your SaaS product helps people stand up the solution much more easily and quickly. And so maybe some comment on that. And then beyond that, as you think about your go-to-market vis-à-vis self-managed and/or SaaS, do you incent the sales force to push one product over the other? Or is it entirely customer-driven?

Yes. It's generally customer-driven. I mean we prefer to lead with SaaS. We invest more in terms of marketing. We think SaaS is better for customers. We think SaaS is better for us. It makes it much easier for customers not just in terms of the initial deployment but really the ongoing management. If you think about the value that a SaaS provider brings relative to an on-premise product -- for any product out there. It's not just about our products or our industry, there's -- obviously, a customer doesn't have to host the software themselves. They don't have to pay for the software itself. And then they also don't have to manage it. So that -- the management value is pretty significant. And so we offer all of those. And so it becomes a lot easier for customers to grow in SaaS, which is why we think it's better for customers and it is stickier and better for us. That's our belief. So we often lead with that. But at the end of the day, we don't provide the customer with special incentives. We don't even provide our salespeople with special incentives to go sell SaaS. Naturally speaking, a SaaS transaction will be bigger than a self-managed transaction because it includes the value of the management and the infrastructure and so forth. And so we feel that that's good enough incentive for our folks as well. And then customers will just be on the journey wherever they are. They tend to deploy us where their apps and infrastructure reside, and very often, that can be on-prem. Sometimes, it can be a self-managed purchase that they then go deploy in a public cloud themselves somewhere, maybe where we don't have a physical presence yet, for instance, but we don't influence customer decision on that eventually. And I'm sorry. What was the first part of that question, Brad, again?

Around usability, which I think you answered with sort of the management side and all the value a customer gets.

That's right.

So maybe just wrapping up on SaaS and then moving into the financials or segueing from there. As the SaaS business gets bigger and bigger as a percent of total revenue, how should we think about the potential impact to gross margins from the solution?

Yes. So first off, we're fully expecting that SaaS will continue to grow faster than the business overall, which we think is the right answer for us as a business. In terms of the gross margin impact on that, obviously, SaaS is accretive to gross profit dollars. So that continues to scale. But in terms of margin, I do expect that there will be a little bit of a headwind as we continue to grow in SaaS and as we build out scale because initially, for a lot of our initial deployments, we are going to be operating at full capacity or utilization. And then over time, as that improves as well as we have more value-added features that we charge for appropriately, I think that helps the SaaS gross margin and overall company gross margin in the long term as well. So long term, I think of ourselves as a traditional enterprise software, gross margin kind of profile. And I think in the near term, we will probably see some modest headwinds associated with SaaS as the mix increases.

Got it. And then maybe turning to the model, Janesh, in the guidance you laid out post the most recent earnings call. Obviously, you guys moved very efficiently to reset the model when COVID hit and come forth with what, in hindsight, was a very conservative forecast, which made all the sense in the world given all the uncertainty. As you think about the fiscal '22 and fiscal '23 plan and sort of goal that you laid out here a few weeks ago, should we believe that there's the same level of conservatism baked in but just better overall visibility because the world is getting better?

Yes. I think visibility and conservatism are highly correlated, right? If we have better visibility, we don't need to be as conservative. And to your point, Brad, a year ago was a very different time and place. We had to be ultra-conservative back then because it just felt like the whole world was falling apart with talks of 30% and 40% GDP reductions and so forth. And we've got a lot better visibility this time around. So as I mentioned on the call itself, I've not been as conservative because I don't need to be because we've got a lot better visibility. That said, there's obviously a fair amount that we need to continue to execute. So our core approach is unchanged. We guide based on what we know. But we just have a lot better visibility and clarity now than we did a year ago at this point.

And when you see what's entering the pipeline today versus even 6 months ago, are the transactions bigger? Are there just more of them? Is it a combination of both?

Yes. I think all of the above, right? Through the course of the pandemic, the pipeline continues to remain robust. So we talked about top-of-funnel activity continuing to remain robust and so forth. The thing we were concerned about over the course of fiscal '21 is, will all of that translate into actual customer spending? And does that translate into conversion rates? And actually, in hindsight, we were quite pleased that it did because it turns out that customers spent through the pandemic as opposed to cutting back in a meaningful way. So things went to higher-level sort of approval and customers had to adapt their processes and so forth, but I think we worked through that. And at least people have now figured out how to operate in a pandemic, whether it's customers, figuring out how to buy and/or our salespeople figuring out how to sell. So I think it feels a lot more streamlined from that standpoint. So that's what's giving us the confidence to continue to invest aggressively in the business to just -- because the opportunity set is so rich. And so that's why we're investing.

And then maybe just wrapping up on that investing point, we only have 1.5 minutes left. But as you think -- from my perspective, there is some level of permanent COVID savings that will -- that sort of move forward because some aspects of how we operate will change. Do you envision that most of that gets reinvested back into the business because of the opportunity?

Yes. I think if you look at the gross margin improvement that we had in fiscal '21, it was pretty significant. And only a portion of that was COVID-related. A lot of that was just leverage as we scale the business. And part of it was at the start of the year, we went slower on investment. So I do think that some of that COVID-related savings comes back into the model. I do think some of it eventually becomes structurally permanent. I don't think the world goes back to exactly how it was before. What we modeled in for fiscal '22 is that we will start to see some amount of travel and events and so forth come back in the second half of the year. And exactly how much that will be and how that will play out, I think it's a bit of a TBD from our perspective. But overall, we've tried to factor all of those things into the guidance that we provided. So we're pretty confident on the outlook for the year.

Excellent. Well, Janesh, we're just about out of time. I really appreciate the opportunity to talk today.

Thank you very much for hosting us again, Brad. Much appreciated.

Thanks. We'll see you soon.

Take care.