Datadog, Inc. (DDOG) Earnings Call Transcript & Summary

Awesome. Well, thanks to everyone who is here in person or joining over the webcast. My name is Jacob Roberge. I'm the research analyst at William Blair that covers Datadog. Just before we start things off, I'm required to let you know that for a full list of research disclosures and potential conflicts of interest, you can visit our website at williamblair.com. Okay. Now that we have the formalities out of the way, really excited to have Datadog's CFO, Chief Financial Officer, David Obstler, here with us today.

Thank you for having us.

David, just to kick things off, I guess, for those that might be newer to the Datadog story, would love if you could give us just a high-level overview of the business and some of the problems that Datadog really aims to solve for its customers.

Good. And let me just say upfront that we were -- December year-end, so we reported -- and we don't update intra-quarter so just make sure that everybody understands that from a compliance point of view. So Datadog has a fully integrated observability and security platform that enables DevOps to deploy, monitor, remediate mission-critical applications, mainly at our client-facing in real time. And most of what is the workloads that are being monitored by Datadog are modern development workloads, you would think Kubernetes, containers, microservices. The company started out with one of what we call the 3 post, which was infrastructure and expanded that over the years to include leading APM and log products and most recently has expanded both into DevSecOps as well as CICD or developer tools.

Awesome. Well, yes, so infrastructure monitoring is Datadog's heritage and where it really began. But this past quarter, you actually talked about APM, application performance monitoring and your log management suite surpassing over $1 billion in ARR. What are you doing differently with just only releasing those in the last 5 or 6 years than some of the incumbents in the space to see that type of momentum in the business?

Yes, we're very proud of that. You can tell from the attach rate that it's basically 2x versus infrastructure. At the core, it all goes to product strategy where everything is knitted together from the database to the metrics, et cetera. Most of our clients have to solve problems in real time and remediate and what they value is having everything linked what we call metric traces and logs, so they can see the full picture. And that's been one of the big competitive advantages of Datadog, the fact that it's a single platform. In addition, it's taken some years to build out the functionality of the APM and logs. And I think at this point, we have product parity or product excellence and combine that together has resulted in our clients landing 75% of them with more than 1 product or landing with infrastructure and either logs or APM and that continuing to over 80% when it looks in the customer base itself.

Yes. It's been really impressive just to see that attach rate and how many more customers are adopting those multiple solutions. But AIOps, I guess a lot of people hear about it. It's a buzzword that's thrown around a lot. And so I'm curious now that you've compiled a pretty broad observability suite where we can do infrastructure monitoring, APM, logs, the [indiscernible]. And I'm curious, what is AIOps to you? Like how do you actually help customers perform AIOps within their own organizations?

Well, we've always had also in the platform, varying degrees of AI. Watchdog, for instance, is part of our platform that allows clients to analyze and remediate. And in fact, just this week, we launched a sort of product enhancement that is workflow. And what that is essentially everything we're doing is trying to make our DevOps customers be able to analyze more quickly and automate, get recommendations. And so everything from the machine learning to Watchdog to the workflows that we are announcing are all aimed at improving that efficiency in the platform.

Yes. Helpful. And then -- so there's AIOps, but then the buzzword today that everyone's talking about is generative AI.

I have to read the questions twice to make sure that you were going AIOps first and then AI.

Of course. Yes. So on the generative AI front, I'm curious how you and your team is thinking internally just about the magnitude of data, the magnitude of workloads that generative AI could really create for Datadog to observe, to monitor in the cloud one day?

So I think if you think back Datadog, so basically, it's called Datadog because at the core, it was found as a data platform. So one of the competitive advantages is that we've been able to aggregate and organize data, which is needed to train models. So essentially, there's a number of ways we think this could impact us. First of all, every time there's been an acceleration of technological change in software development, containers, micro services, et cetera, the speed of development and the complexity has increased, and that's been complementary to Datadog because as you -- if this results in more efficiency and more rapid application development. That will mean more workloads and more modern workloads for Datadog to monitor. So overall, we think this is complementary to our end customers, and therefore, they will need more tools. I'm starting out for facing first, and then we'll go back to Datadog itself. Yes. Another thing that we've seen and we talked about this in our earnings call is, we've always been a very good fit with machine learning and AI and data companies. We highlighted one of those in the earnings call, we didn't give the name, but one of the companies that's providing technology. And to the extent that there's more traction in those type of companies, they are excellent fits with Datadog, and we'll see accelerated growth in those customers. So that's outward-facing. Now inward-facing to Datadog, we have begun to look at ways that we can use the product internally to speed up our software development or produce marketing collateral faster. We're early on, but we're optimistic like others are that there will be use cases inside Datadog that will help us become more efficient, churn out more software, churn out more marketing -- marketing collateral, et cetera. And then in terms of the platform itself, I mentioned in the AIOps answer that we have all been doing this for some time, but there'll be other things we can do now, pilot, et cetera, that will essentially help our clients use our platform to resolve things more quickly. And we're optimistic that there'll be product functionality. We're earlier on. We don't know how this is going to go, but we're optimistic that this could be an opportunity, both inside with product as well as for our customers.

Yes, that's actually 2 things I kind of want to dig into on that answer. So you talked about earlier where, hey, Datadog is really helpful when we get these complex infrastructures with micro service with containers. And so generative AI move into the cloud just creates this messier infrastructure. So that's one side of the equation. But then the second side of the equation is, and I'm curious for you, are you starting to have conversations with customers that are saying, "Hey, we might go to the cloud 2 years or earlier than we originally anticipated because we want to access these generative AI technologies and move more and more workloads to the cloud quicker than they may have previously expected."

I think it's too early. What we said is we don't know how this is going to go. Being honest to treat this is the world -- the software world doesn't know how this is going to be monetized, right? You've seen, of course, you all as investors have seen an investment cycle in the infrastructure with [ Navidea ] and others and that has to be -- you can't do any of it unless you get the infrastructure going. So it's a little too early for us to give examples, but we certainly said in our call and in our subsequent investor meetings, that will let everybody know as we see that in which way it goes.

Okay. Helpful. Well, that will be it for my generative AI questions but that's the top...

You're very efficient with that. Usually occupies 50% market share in most discussions.

There you go. But just taking a step back, I think a common feedback point that we get from a lot of investors is Datadog is a great business. It's operating in a really large market, but there are a lot of other alternatives in there, whether it be log management players or APM players. So I'm curious, how do you compete in that type of market? And how do you make sure you're the long-term winner in the true observability space?

Yes. So it's a big market. It's a market that has a long way to go. Looking at research, Gartner and others, you see 1/4 of the workloads, maybe 1/4 of the applications. So we have a lot of runway. And the main way that Datadog has competed and one is through investment in product. If you look at our R&D as a percentage of revenues in the low 30s, it's an outlier relative to our competitors. We are lucky enough to have designed or the product to be very efficient to be able to be installed by customers in a very frictionless way. That is the core to a very efficient sales and marketing go-to-market that we plow back in a material way in R&D. So I think the first part of it is that we have invested substantially. The next part of it is the clients are speaking that they want -- if they're going to buy a SaaS software solution, a tightly integrated package that economically can be deployed ubiquitously and can allow collaboration. And brilliance or luck, the founders of the company designed the platform that way. So essentially, everything tightly integrated and developing more and more functionality has been one of the major factors in us winning versus competition. I would say, the next thing is that we have expanded from there to instead of price raising, we've been value raising. And that has to do from everything from an APM to digital experience monitoring, the shift left to security, which we'll talk about. So we are constantly over the AIOPs, which constantly trying to stay ahead of the game, look forward to our clients' needs and invest in the platform.

Yes. No, it's really impressive to see the margins that you have while still spending low 30% on R&D. But just thinking about the actual performance. So 2022, really solid year. Over 60% revenue growth, really solid margins. But we did start to see revenue growth decelerate a bit towards the back half of the year and into this year. Just as people -- we impacted the macro and there are things people going through cloud optimizations, everything around that. So I would love to kind of hear how the macro impacted you, especially as it relates to those cloud optimizations?

So in retrospect, during COVID in the middle, we had a bubble. There was a tremendous investment cycle in IT. A lot of companies were growing very, very fast. We also had a capital markets and investor environment that was putting a premium on growth versus the balance between profitability. So in retrospect, we had a spike in investment. And starting in -- we said May of last year, when that broke and you started to have capital markets break the Fed increasing, et cetera. You started to go into a cycle of sort of squeezing that out. And there are some good things that have -- we've seen from this, and there are some more challenging things. On the good side, our gross retention, meaning our customers staying with us has stayed really, really high in the upper 90s. What that indicates to us is we're a must have, once they install us, they're not leaving. Secondly, we've continued to have new logos and new workloads be very steady, which is a good thing, meaning priority workloads. But we've been up against is cost management and optimization. And what that is, is essentially clients are slowing down the growth of their consumption of cost across the board. And it starts with the hyperscalers, but we're attached to that in some other ways. So they started that. And what we said we've been experiencing that. This will be the fourth quarter we've been experiencing that. We -- in the first quarter, we had growth that was in terms of the approximate organic growth that was slightly higher than Q4, but a little lower than Q3. So we're sort of plateauing down there. And we have said we don't know when this is going to end. It's going to be a lot in the economy. On the other hand, we have lapped those comparables of that accelerated or caffeinated growth. So the rate of decel is going down. Now when we will excel again is largely going to be dependent on when all of those customers are finished with their cost management efforts. And there's been a lot of comments on that, and we don't know the answer to that. So we, in our guidance, we just put in, we assume it's going to continue throughout the year. That should set us up very well for resumption of growth, but we don't know when this optimization is going to run its course.

That's helpful. So that's on the cloud optimization front. And where we've obviously seen that from the hyperscalers as well. So there's a lot going on. You've mentioned a lot on the recent calls just that actually Datadog optimization project to have been quite a bit within your customer base. So curious if you could just walk us through what a typical optimization project looks like in terms of duration and what customers are looking at in terms of how much data they're ingesting, where they want to place workloads. I'm curious how that typical project goes and how long it lasts?

So the cloud itself and Datadog is sort of a land and expand. And we have optimizations going on, both internally at Datadog and all of our clients all the time, even in the highest growth. So that sort of continues. And what that means is that since these are mainly new workloads, and important services that the clients are at different times, emphasizing the speed or intensity of putting those applications versus the optimization of those workloads. What's happened in this environment is much more intensity or weighted average of that optimization. And what optimization is, is we have 2 types of products, broadly speaking. We have products that are related to the infrastructure that essentially depend on the number of hosts that the client is buying. And when they go through, and I think that's the part that's most associated with AWS, Azure and Google. When they optimize that, we will be sort of a byproduct of that. In addition, we have a number of data products, logging, parts of APM and testing that are denominated both the amount of data indexed or the amount of tests and clients can at different times, look at that and decide to do it more intensively or not. And so we've gone through both of those cycles in a normal optimization. The fact they matter is, for the most part, our clients, as we mentioned, have 1 product, the unified platform of observability. And they can choose within that what to use and therefore, they're going through and deciding what to do. This also provides a good opportunity for cross-sell because if they're -- if they have a commitment and they're optimizing the infrastructure, they can then use our shift left or they can use our digital experience monitoring or can use security or they can consolidate other solutions on a Datadog and we found the intensity of all that increasing during this period as well.

Yes, that leads me right in my next question. And so I'm curious if we could dig into that a little bit more. So as we head into particularly tougher budget environment where people might look to cut costs, they move costs around. There's also a really nice consolidation opportunity. And so I'm curious what the early results have been on that and what you're seeing in the competitive environment. I think customers really looking to consolidate on broader platforms like Datadog.

Yes. So in terms of cloud native for the whole wall-to-wall and an enterprise for this particular use case, as I said, the platform has been one of the winning propositions. So our clients realize that having everything all they organized for them is advantageous. But there were other products that existed before. We had the products, APM and logs. There are other installed bases. There are other champions, their contracts, et cetera. So what we've been finding over time is, there's been a steady consolidation away from other point solutions or observability towards Datadog. That appears to be accelerating, of which, because they can both increase the functionality of the platform by having it integrated and save money by consolidating, whether that's going to the pace of that, we don't know, but the pace has been positive upwards on that.

Yes. That's helpful. And then switching over to another topic. You've mentioned shift left a few times. So Datadog has been on the security journey over the past few years. Can you walk us through the progress that you've made with that product suite and just some of the recent updates with that business?

Definitely. So first of all, to clear up a little bit of some of the thing that confuses, we're not doing endpoint security. We're not doing network security. We're not doing e-mail. So this is cloud security. This is the security of cloud-based applications, and we've been building out for the last couple of years, a suite of products that includes container security or cloud security, security posture management, security cloud, security cluster management, application security and then a cloud team. And we're in the middle of that. What we said on our call is we have 5,000-plus customers, that's about 20% of our customers. So we've gotten good adoption. It tends to be with our more progressive DevOps who are employing DevSecOps, meaning they're injecting security into the development and deployment of applications. We still have more time to go in building out the suite. It's not fully functioned. So we're happy with the progress, but we have more building to do in order to get the product parity like we've been able to do in logs and APM.

Yes, that brings up a good question, just if you're not going into endpoint security, not going into these different places. So I'm curious, who are you competing with in this cloud security market? And where do the budget dollars come from when your customers adopt Datadog security plan?

So it's largely a greenfield, but I would say that of the existing vendors, one of the largest players, Palo Alto, through their Prisma suite has penetrated this. There are other companies out there, like [indiscernible] et cetera. It's also greenfield. And the users tend to be mainly DevOps, but the budget may vary well and in some cases, come from security. That leads to a go-to-market challenge, which means and we said this all along, that as we're building it out, we're also thinking through how are you going to influence potentially these 2 buyer groups. And it's pretty unlikely that you're going to buy Datadog for security, if you want to have a completely centralized security department because don't forget how Datadogs architected. But if you're adopting modern DevSecOps and you want it out there in the world of the DevOps group, then it could well be that the users are the DevOps world, but you need to go and get out of the security because that's where it's likely to come from. And we said to everybody, we're trying to learn there. We really need to finish the build, but we're willing to invest if that's the case, to have some sort of distribution or influence to the security role.

Yes. That's helpful. You mentioned that you now have 5,000 customers on the security platform. And so I'm curious, now that you have that large of a customer mass, what are you seeing in terms of the data of where you might be really striking a cord and customers really like using Datadog platform and maybe some areas that you really want to -- you don't have right now that you want to dig into deeper over the next few years?

Yes. So good question. So basically, we started out with SIEM and now we're sort of launching the cloud security and AppSec. So essentially, that plus data governance security tend to be the pillars of cloud security. And so it's a lot of blocking and tackling. We just launched the AppSec. So we think that competitive advantages for us could be building the AppSec out more, given that we already have the applications instrumented with APM. Using our data because we have a lot of the data and then having lower barriers to adoption because we also have the DevOps customers. All of that, I think, is -- we think is going to take product parity and we're not quite there yet. That includes things like the database to interpret signals that come from penetration or security threats versus observability. Workflow that has to be included to manage the security. And all of that, those are part of the built-in security.

Yes. That's helpful. So security just launched the last few years, but we touched on your R&D organization earlier. Just over 30% of revenue. I'm curious, in your perspective, outside of the ones that we've talked about, what are some of the newer products that you're pretty excited about right now? Because your product team pushes out quite a few product feature, so I'm curious where you're seeing some early fruit?

So we talked about cloud security, so I'll leave that aside. But other things that are exciting is we've launched and are still sort of building it out cloud cost management. We have access to a lot of the data on the consumption of the cloud by our clients and particularly in this environment. Having that transparency to manage cloud cost is a good fit. That's got good early traction. It's still early. I talked -- we talked about the shift left in the CICD or moving the use of our product early in the development cycle. ITSM, trying to enable our platform to handle workflows more automatically, we talked about that. We think is a big opportunity to provide more functionality in the platform. So I think those are some of the biggest potential opportunities for TAM expansion and increase value to customers.

Yes. So that's the R&D side of the equation. But on the go-to-market side of the equation, you have really strong unit economics, really efficient go-to-market motion, what allows you to kind of run that type of model with how many different products that you're selling today?

Well, that all goes back to the way the product is created so that our clients can download the product and use it with no intervention. That are limited. That means they can also explore -- they can explore their own data in the application very quickly. They can also use other products very quickly. And that frictionless bottoms-up type of adoption is quite efficient. I mean you don't have to sell every time, clients are selling for you and adopting. That's sort of at the core of why it's efficient.

Yes. And that -- we've seen that buying pattern be really effective, especially with SMB mid-market customers. But you're starting to move your platform further and further upmarket. I'm curious if you think the go-to-market motion has to evolve at all as you move further upmarket, so that might be maybe GSI partnerships thinking about that angle. I'm curious how you're thinking about evolving the go-to-market motion as you move further upmarket?

Good question. So we've been doing a number of things. We have invested substantially in enterprise sales force. We've invested substantially in enterprise sales engineering group. We've invested more in the enterprise account management group. We've invested in technical account management, which is assigned specifically to large customers and helps them to both implement and use. So those are internally all things that enterprise -- large enterprise customers want. Outside, you're right, it's GSIs and channels. GSIs are quite important. We have been investing in it that involves partnerships, commitments, technical resources to architect, et cetera. And we have a number of situations that have been produced by the large GSIs and are continuing to sort of expand that on a global basis. That's an area of significant investment from us, and you can see more of that in the future.

Yes. No. In the -- I'm curious if we take a step back and think about the macro again and how it relates to customer conversations that you're having, in the observability market, where data is this big -- is a big growth driver. It's a really nice opportunity over time. But I'm curious in your conversations with customers, how are you thinking about addressing that in an uncertain macro time where a customer might think of Datadog is potentially a data tax and how they're thinking about things. So I'm curious how you're thinking about those customer conversation in an uncertain macro?

Yes. I mean, all along, we've been working a lot in the platform to make our handling of the data. And others are too more efficient. So that we don't have to charge them. Then essentially, what we've been doing is we've always priced not on an ingestion. Some have priced mainly around the ingestion and storing of the data. We price mainly on the action, the indexing, et cetera. So what we've worked to try to do is to try to illuminate that to client so they can see where they're spending the money. And then through this technical account management, and others, we try to work with them to try to spend their dollars in the best possible way, which means if they're over logging, we essentially tell them that. We don't charge them for that over logging period up to a certain point. We point the investment maybe in a different way. So we're trying to be very responsive to clients to keep them as long-term customers and not have them use the platform most efficiently.

Yes. I know we're coming up on time here. So I'll just ask you 1 last question as we wrap up. What makes you most excited about Datadog and staying with the company and really driving things forward? And yes, what makes you excited about Datadog?

That's a really good question. So this is my fifth software company, fourth public. It's been -- I'm going on 5 years. So it's been a long time. So essentially, one of the things I've learned over the years is that you can manage, but you need to have product market fit and a large TAM and product leadership in order to become a really important technology vendor. And that's what excited me about joining Datadog. It excites me that we continue to invest so well in the products so that we can add more value. And one of my jobs is to make sure we're as efficient as possible so we can continue to invest in the product. In addition, my partners and I know this might sound [indiscernible], et cetera, but they're really good and really smart. And the founders of this company -- the company and the people that have built it are unbelievable visionaries and executors. And that's hard to find in that combination. And I think a lot of credit to them for what they've done and in keeping and building the culture of Datadog. So that's what keeps me very excited all the time, the opportunity.

Well, we're definitely excited about the story as well. So thanks again, David. I appreciate you having me.

Thank you.