Dynatrace, Inc. (DT) Earnings Call Transcript & Summary

All right. Good morning, everybody. Day 2 Communacopia. We have Dynatrace kicking it off for us this morning. Rick McConnell, CEO, Jim Benson, CFO. Thank you guys for being back at Communacopia conference.

Always happy to be here.

All right. Good. Rick, let's start with you. The observability market feels like it's entering a new phase. The AI is changing both the workloads customers need to manage and what they expect from the platform. How do you see the category evolving? And where can Dynatrace play a broader role over time?

Well, there is no doubt that the observability category is evolving in a substantial way in the age of in an AI-first world, we would submit that die in cast that observability is more critical each day. And I remember -- I remember 6 months ago, 9 months ago, when we're in the midst of SaaS Pocalypse when it was deemed that the vast majority of software could be rewritten, it has evolved since then. And I think we've evolved to the point where there are clearly categories there are winners and maybe some other categories that will struggle more in an LLM driven world. And I would submit to you that observability is one of those categories that is going to win. So we are entering a new era. And in observability, we have all the traditional software observability that we've been driving today. This is the couple of billion dollars of ARR, plus it's growing at 17% as our core business, and that continues to evolve to a world where customers want automation. They want autonomous operations. And the goal in this traditional business is that how you have software that essentially runs itself that corrects itself that auto remediates and that can reduce the human load required to manage it. Similarly, though, you have this new expanded category, which we're thinking of as AI observability. And AI observability is the observability of AI workloads and agentic models. And this is a market that we believe is going to expand to in the range of a $10 billion a by the end of the decade, growing well more than 50% per year. So we've got large traditional $80 billion-plus traditional observability market growing in the mid-teens, supplemented by a brand-new category and AI observability that is even more critical because you need more observability to oversee AI workloads not less. And what you need is context be able to do that and manage what observability systems provide in real time by analyzing billions of interconnected data points. And that is a huge differentiator from LLM models than what they would be doing.

Okay. Great. That's a great level set. Let's dig in a little bit deeper there. Given Dynatrace's footprint across the Fortune 500 you have a pretty unique purview into how large enterprises are adopting AI. It seems like the opportunity is quite significant. You're talking about a $10 billion market, 50% CAGR, but production deployments are still developing, at least from what we can see. So what are the main bottlenecks today and what needs to happen for AI workloads to become a more material driver of consumption for Dynatrace.

It's a great question, Matt. The biggest delta that I see, and I'm -- I always have believed that I'm in a privileged position of being able to meet with CIOs, CXOs, all over the planet of Global 500, Global 2000, even Global 15,000 organizations and here I could tell the inhibitor to broader-based deployment of AI is confidence. And what I mean by that is when you start relying upon LLMs to provide data to end users from your mobile app, from your website, from whatever it might be of the company that would be our customer. The challenge is you better make sure that, that information coming out of that LLM is right because if a bank, for example, has a prompt in a chat bot sitting in their mobile app and somebody says, "Well, where should I invest $10,000? And the answer comes back, "Well, the highest return over the last 90 days was to invest in Bitcoin. So why don't you put all of your money in Bitcoin?" Probably not the right answer depending upon the characteristics of that investor. And so the calibration of the input coming from the LLM who deliver to end users is really critical. And it is that confidence level that I think is required to get it over the next time. And this is precisely what the domain of AI observability is AI, observability is around LLM experimentation, LLM observability. It's around AI development life cycles and workloads. And the result of this or the intent of this is to ensure that you're answering an incremental question from the traditional question of observability. The traditional question that we're trying to answer in observability, this couple of billion dollar business, what I talked about of Dynatrace's, is it running is it running? Are your workloads running the way that you would expect? Are they optimized? Are they delivering? Are they up? What is the availability time? These are the elements of the domains of traditional observability. Once you move into an AI world, you have to ask an additional question. Those AI workloads also have to be running. And so you start with the same question, but you then expand to another question. And the next question is really around, is it accurate? Meaning that are the LLM delivering accurate information that can be relied upon to be delivered to end users so that you, as an organization, have confidence that the end user taking action on that can actually bank discernible progress. And then the third question that is associated with those AI workloads is, are my models working right? Are my agents behaving the way that you would expect them to behave? So the point is, in a traditional observability sense, is it resilient? Is it working? Is a pretty good start when in in AI workloads, you need to supplement with a couple of additional questions related to the accuracy of the data flow? And that's where, a, it's getting really exciting for observability, but b, observability is becoming completely mission-critical in delivering AI workloads with competence and successfully.

So how would you frame sort of like the maturity of the product capabilities to meet that moment relative to sort of the enterprises in terms of their own progress and readiness?

Well, we at Dynatrace had been investing internally in this AI observability category over the last 18 months, and we're delivering capabilities in AI development life cycles and experimentation and so forth. But we see this as such an extraordinary category for which the time is right now, that a few weeks back, we decided to acquire a company in the name of Arise based here in San Francisco. Dead smack in the middle of AI central, so to speak. And as Arise is the category leader in AI experimentation and AI observability. And so the result of it is that we're betting with our pocket book, so to speak, that this category expansion is mission critical to our ability to deliver broad-based observability because I don't think it's going to end up bifurcating. We're not seeing these 2 disparate markets. Rather, it is a converged market where our companies, our customers want to deliver traditional end AI observability. They want to do it on the same platform in a bimodal way, and they want to be able to leverage the assets that they've deployed with Dynatrace to be able to do so. So they really want an end-to-end platform to be able to deliver these capabilities, which is why we'll take a rise marry it together with the Dynatrace platform. And then you go all the way from developers to production. You go all the way from preproduction to more sophisticated workflows. And in doing so, you can also go from developers to IT ops. So you span the gamut of personas as well.

Okay. Great. I want to move into rise a little bit later in the discussion. But first, Jim, I want to pull you in. You've given us some disclosure that customers monitoring AI workloads are already consuming the platform at a faster rate. What's different about how those customers use Dynatrace? And what does the early behavior suggest about the longer-term expansion opportunity?

Yes. And just to level set for everyone that the statistic was that we went from our fourth quarter to our most recent first quarter, from a little over 800 customers, where we were monitoring some LLM workload to 1,000 in 1 quarter. And then we went from $500 million to $850 million customers that are leveraging Dynatrace for some of our agentic capabilities. So kind of 2 unique cohorts. And so to your point around enterprises, which certainly are the bread and butter of our business, enterprises are experimenting. There is work being done. You can kind of see that. I would say that what we're seeing is this is in maybe some of the more progressive enterprise companies. And what we found is the characteristics of these cohorts as they consume the platform at a much more significant rate pace. And so their consumption growth is 50% higher than customers that are not leveraging us for an LLM workload or for agentic capability. And so we view it a little bit as this is a precursor of what we think is going to happen when others start to do it and it will build. And you say, why is that the case? Well, one, I think, Matt, what ended up happening is they start leveraging more of the Dynatrace capability. So those customers tend to use Dynatrace for more of our capabilities. And so think of it the breadth of the platform that they're consuming is not maybe just using us for application performance monitoring our infrastructure. They're usually using us for application performance monitoring, infrastructure monitoring, digital experience, logs and in some cases, security. And so they use more capabilities of the platform. And I think it's also true that the LLM workloads tend to be chattier. And so the kind of the characteristics of those workloads is you consume at a higher rate. And so again, going back to our -- I'm sure we'll cover it later, our Dynatrace platform subscription model, which at its core is consumption-based. If you consume at a greater rate and pace, which AI workloads, you will, you will burn through your commitment earlier. If you burn-through your commitment earlier, you'll do an earlier expansion. And so the economics work for Dynatrace. And I'd say where we are, I would admit we're still in early innings, but I'd say we are already seeing the benefit of consumption of the platform.

So it sounds like you actually are starting to see red curve on the side.

Yes.

Okay. Let's double-click on the DPS opportunity. This is the first year the major DPS are moving through annual resets and contractual renewals together. What are you learning about how consumption maps to committed spend as customers move through these milestones? And how is that relationship developed as you expected?

Yes. I think we've said in the past, and you're right that this is the first year that it all converges, where all of your cohort classes come up for their annual reset or renewal. And then that will be the new normal every year, Matt. So this is the first year, you'll see that. And I'd say the general characteristics that we've seen, most of our DPS contracts are 3 years, not all, but most. I think the general characteristics that we saw is customers that are maybe earlier in their life cycle, maybe year 1, they may have a predisposition if they're consuming at a pretty significant rate to go on to that. They had just on a renewal maybe a year ago, and it isn't worth necessarily going through the bother of an expansion. And then in year 2, year 3 cohorts, more inclined to do an expansion. So I'd say that behavior remains. And what I would tell you is that we're going to -- most of the renewals are in the back half of the year. So 70% of what you outlined is renewals or annual resets that are going to happen in our third and fourth quarter. So only 30% are happening now. So to the characteristics that we're seeing now are consistent with what we expected but you still have a very -- your volume of them is significantly less. And so for us, the big push is continue to drive consumption because, again, that's the core of the DPS contract, and we've publicly talked about that our consumption growth rates are well over 20%. Our ARR growth rate is 16% to 17%, and -- and so if you continue to grow consumption in the 20s, you will see a convergence of ARR, where you will start to see ARR inflect upwards more towards the rate of consumption and I think the back half of the year, if that continues, we expect that we'll see an increase in the expansion rates and the NRR will then inflect.

I guess just to put a finer point on that, like the 30% of renewals that came due in the first half of the year and what you've already seen inside the business today, is that commensurate with the 20% plus consumption rate that you guys have been talking about for the last several quarters?

It is. It's it's just a small percentage of your customers that, as you know, we had a huge quarter in the first quarter with new logos, and I'm sure we'll cover that later. But reps are incented on maximizing their quota. And so in some quarters, you're going to be more new logo weighted. Some quarters, you're going to be more expansion weighted. I think just given the nature of our renewals, I would say we're going to be more expansion weighted in the back half of the year, Matt, probably a little bit more new logo weighted in the first half. But that just coincides with what the renewals are when renewals come up for their exploration.

Okay. Great. Rick, let's move to another growth pillar for Dynatrace tool consolidation, right? That's been a major growth driver but consolidation can mean very different things across customers. So what are enterprises actually replacing today? How broad are the initial deployments and what tends to prevent a customer from consolidating more of their environment on to Dynatrace?

Well, if the goal, which is where I started, is autonomous operations that ultimately you need to move out of this manual oversight of your software because there's just too much software to oversee and to keep running. And then even once you find out what's wrong, it takes you too long to fix it. And with agents now developing code, it's expanding at a rate that no organization can keep up with, you've got to find a way to automate that. The challenge with today's environments that are fragmented at many customers that we see day in and day out, are that they are using one vendor for application performance monitoring. They're using another one for infrastructure monitoring, another one for log management, another one from [indiscernible] experience and synthetics and applications and the list goes on and on. And the problem is that uses different data stores, different collection methods, the data is fractured and you end up having to do manual tagging, manual oversight, manual collection and a simulation of that data to make any sense of it. And that doesn't make the problem go away. That makes the problem worse because now you've got all of these data flows and you really can't have an automated system that's overseeing it. So the trend, and this is sort of irrespective matter of Dynatrace even the trend in the industry that we would say is toward end-to-end consolidation of observability tools. And the reason is because then you get data collected in one environment, you can oversee it with an overall analytics engine that for us is Dynatrace intelligence that we can then provide answers and not just guesses and the answers can be acted upon by an agent system and trust it. So that's what's driving it. And I would say that end-to-end observability is getting driven in 3 levels. Number one is the data level logs, traces, metrics, behavioral analytics, business events, all in one integrated data lake house, Dynatrace is unique in delivering that as part of our GRAIL solution. An integration at the domain layer, which are application performance, log management, infrastructure monitoring, all as I mentioned, and then an integration at the persona layer. You want platform engineering, SRE development teams, IT ops and so forth, all looking at the same data. end-to-end observability is really about aligning and integrating it all three of those layers, and then in delivering that a unified solution that can result in much, much more automated response through this analytics of data that you just couldn't do manually.

Okay. Yes, that's very helpful. I guess maybe to drill into the consolidation point, I think it makes sense in context of what we're seeing in your new logo ACVs and new logo ARR...