Everpure, Inc. (P) Earnings Call Transcript & Summary

I think we'll go ahead and get started, everyone. I'm Kevin Kumar, I cover software here at Goldman Sachs. And with me, I have Rob Lee, Chief Technology Officer; and Kevan Krysler, CFO of Pure Storage. Thanks so much for being here. Just to start off, I wanted to read some disclosures here. Statements made in these discussions, which are not statements of historical fact are forward looking based on current expectations. Actual results could differ materially from those projected due to a number of factors, including those referenced in Pure Storage's most recent SEC filings on Form 10-Q, 10-K and 8-K. All right. So maybe to start things off here. Rob, for people who are newer to the story, maybe just give a brief overview of the company and kind of the solutions we offer to your customers.

Yes. No, absolutely, Kevin, and thanks for having us here today. So for those of you who may not be familiar with the Pure story, we started about 14 years ago. And what we saw at the time was an opportunity to fundamentally disrupt the storage industry, the catalyst for that change being largely driven by the advent of the new technology, Flash, at that time. Now we weren't the only ones that saw this potential. But what I'd say is we were the only ones that took a specific set of approaches and bets for the long term to make that a reality. Number one was we recognized that while Flash had a ton of potential for it to succeed in the long term. It was largely going to be driven by the consumer industry, which was going to push the flash technology down a direction of higher volume, lower cost, lower quality parts and so a different approach to software, a different approach to building systems to take advantage of that semiconductor was going to be required. So number one, we took a very different approach than everybody else in the market, largely driven by a software focus. Number 2 was we recognized that by removing a lot of the moving parts in traditional storage systems by, removing spinning discs by building more efficient systems that require less cabling, less fans, we had the ability to build much more reliable systems. And so we recognized that the entire storage buying experience, this idea from a customer's lens of I buy something day 1, and I immediately start planning for what happens when that asset becomes obsolete in 3, 4, 5 years, we could fundamentally change that. And so number 2 was we invested in a set of technologies, which we now call Evergreen or Evergreen architecture that make that promise a reality. And then number 3 was we recognized very much like how a [indiscernible] consumer was driving the semiconductor road maps. Consumer industry was also driving a sea change in management expectations and operational experiences, right? This was about the time that iPhones were taken off. You buy an iPhone, you don't sit there and study a manual. You probably -- I don't even know if they come with manuals. But if they do, you certainly don't read them, and you don't take weeks long courses to get certified in them, we saw an opportunity to bring that experience into the enterprise. And so if you fast forward 14 years, that really leads into what we see as our key unique and sustainable competitive advantages today. That initial software ethos growing into our Direct Flash technology today, what that enables us to go do in terms of bringing flash to all parts of the data center. Number 2 is that core ethos of building longer lived, constantly refreshed, nondisruptively upgraded systems growing into Evergreen in the Evergreen promise and business model we've built around that. And then number three, our focus on management simplicity, not just in isolation, unit by unit, array by array, but then broadening that over the entire portfolio set. So a little long winded, but hopefully, that catches folks up over the kind of the key elements for the Pure story over the last 10 to 14 years.

That's great. And I'll definitely have some follow-ups on those. But maybe, Kevan, just given that you recently reported, can you just maybe give us a brief summary of kind of the results, kind of the key highlights, key points that you want to highlight to investors?

Yes, sure. We just completed our Q2 earnings. And I'd say that the significant highlights for the quarter probably would start with our portfolio of FlashBlade and the performance of FlashBlade across our entire portfolio. So when you think about our FlashBlade portfolio today, we run from the highest performance all the way now to FlashBlade//E, which is price-sensitive, driving the disc takeout initiative and vision that we have. But across that entire portfolio, we just had tremendous performance. And I view that as one of the key highlights for us this last quarter, and that obviously was part of the Gen AI win that we talked about, which was low 8 digits. So that was terrific. Evergreen One, which is our Storage as-a-Service, and that would include our entire portfolio of solutions just continues to outperform. We originally, given the macro backdrop had assumed more strength with Evergreen//One going into the year. That's completely outperformed our expectations. So another significant highlight for us. It does present a near-term headwind for us on the top line, but we'll take that trade any time given the value that we're driving for our customers on Storage as-a-Service. The other area I think was really coming into validating our structural advantages that we're driving with our Purity software, with our Direct Flash modules. You obviously saw that coming through on our product gross margins at our subscription gross margins. I'd consider that a highlight in terms of really validating the differentiation and competitive advantages we have with our Purity software as well as our architecture around our direct flash modules, which then lends itself to the operating margin expansion, which we're able to drive for the year. So I think that's kind of how I would kind of summarize the key highlights for the quarter.

Yes. I guess on the point on competitive differentiation and how Pure manages Flash natively without going through SSD translation? How important is that? And how is that differentiated from other competitors?

It's a great question. I would say in the early days, it was very important to give us a lot of the disc replacement workloads that we're looking at. And let me maybe peel that back a couple of layers, right? So if we look at what an SSD is, why an SSD came about, it's quite simply a bit of a technology coping mechanism, if you will, right? Flash has been around as a media for 20, 30 years. When flash started showing up in computer applications, all of the software in the world was built for and assuming that it was talking to a disk, right? Flash chips natively behave very differently than a spinning disk. And so a layer of translation is required to read and write data from Flash that is very different than what needs to happen when you're working with disk. What the industry did at that point was they said, "Look, we can't rewrite all of the software in the world, right? We've got to go bring Flash to market. What they did was they built that translation layer into a packaging called an SSD. So an SSD is really kind of a special purpose computer, if you will, running all that complex translation kind of logic. Now what's happened over the years is that as Flash has become denser and larger, as it's become more driven by the consumer market, it becomes harder to work with. That software that's running in SSD is very pressured. It's reached limits of what can run in an SSD. It's bound by DRAM, it's bound by power requirements. It has reliability issues, so on and so forth. Pure took an entirely different approach, right? Because we recognize that, that layer of translation was not necessary. We built our software entirely to speak to Flash natively, removing the need for the inefficiencies that are found within that SSD. So what does that give us? Well, it gives us efficiency advantages when it comes to performance, right? That translation doesn't happen, it's less work, so the data moves in and out of the Flash more quickly. It gives us efficiency advantages when it comes to capacity. We're not repeating work that an SSD would otherwise have to do, that allows us to ship more usable bits, more usable storage to our customers with the same number of Flash chips. It gives us reliability advantages, right? Our DirectFlash modules are hardware modules utilizing the raw NAND, have to serve our application, our software and our software alone. An SSD you buy that our competitors are stuck on has to serve many masters. It has to serve Windows and Mac and VMware and Oracle and that causes reliability issues. And so when we look at historically what that advantage has given us is it's given us performance efficiency benefits. When you look forward at the QLC road maps and you look at the density advantages we can go drive because of that software that's going to allow us to significantly pull away from, frankly, the entire industry that's stuck on SSD technology because they don't have that software.

It seems like the Purity software, to your point, is like the key enabler, everything you really do, right? I mean, Kevan, like do you see -- I don't think that's a separate line item on the financial statements, right? So how do you assess the value that's been driven by that software component?

You're right. It's not showing up on an individual line item on our P&L, but it's really the underpinning of our solutions, frankly, and our competitive advantages. When you think about an operating system, Purity is the operating system driving our entire platform of data storage solutions, which is a differentiator, right, in terms of really focusing on one platform for data storage needs for our customers. And so when you think about, where does it show up in terms of the financial lens. Well, look, I think what we're seeing in terms of gross margins is a significant differentiator. The value that customers see, the business value the customers see with our Purity software, the value is coming through in terms of software like gross margins. Probably a little bit too high for me for my liking on the product gross margins, but I'll take it for now. But on the subscription gross margins as well, we'll see that, especially with Evergreen One, when you think about the engine and the operating system for our Evergreen One offering, which is Storage as a Service completely, that operating system is also being driven by Purity. And then the management system is also very powerful for our customers with Evergreen One with [ Pier 1 ]. So it's really those 2 software engines that drive our entire platform of solutions, which is quite compelling for us. And hopefully, over time, we'll see that through moderate operating margin expansion as well.

Given the pace of improvements in Pure's flash offerings, I know the management team has talked about the disk market potentially going away in the next 5-ish years. So just maybe add some context there in terms of what needs to happen from kind of a product perspective to kind of reach that end point.

I think from a product perspective, certainly, there's areas where we'll continue to add features, we'll continue to expand our integrations. But the big rocks, right? We've got them, right? I think with expanding the DirectFlash technology to what we've brought out with the e-family so far with FlashBlade//E later this year with FlashArray//E, debuting the denser Flash modules, but also targeting the smaller capacities, I think broadly expands the range of workload sets we can go and drive disk replacements for. And then certainly, the road map beyond that for further density improvements only makes that value proposition even more compelling. I do think you have definitely a couple of barriers you got to overcome though, right? So number one is there is still parts of the customer base and market that are religiously tied to the solutions they know. And so we've got to go and drive the education of, hey, you always thought about this as a disk footprint. But it's not that you actually like disk. There's no attribute of disk you actually want. It was just the cost-effective solution. What if we can give you all the benefits of Flash that fits the power, fits the energy for the same cost of the disk base system, why are we even having this conversation, we've got to go beat that drum. And I think the other headwind to be aware of is, I mentioned before, the entire industry historically has been on this buy an asset day one, depreciate it over 4 or 5 years kind of treadmill and so we've got to go work with customers as they have systems coming up for refresh and renewal -- to pound home that message, but also get ahead of it, right? We have started to see opportunities and customers that have seen the significant savings that they're getting from operational power -- space, so on and so forth and been able to use that to pull forward their planned refresh cycles with the savings that they're getting on Pure solutions.

I did want to talk about Gen AI. Obviously, a hot topic right now. Just maybe what are the opportunities that it's creating? I know last earnings call, you talked about training large language models and inference. So just maybe help us understand kind of the opportunity there.

Yes, absolutely. So I think that the recent wave and renewed focus on AI and Gen AI is only constructive for the industry as a whole. It's -- analytics and AI has always been a strong segment for us and continues to grow well. That said, if we unpack it one level further, I think there's actually a couple of different opportunity sets within that. You certainly got the AI training infrastructure itself. The environment that folks who are building the models, training these models are working on and building out, but I think the broader opportunity set is also looking across the rest of the enterprise at folks that, hey, maybe I'm not going to go build out thousands of GPU servers myself and hire hundreds of machine learning scientists, but I want the benefits of that technology. I want to go deploy it. I want to connect it to my enterprise data sets. How do I go and do that? So I think you've got these 2 separate opportunities. And so if we start with the infrastructure set, that's a space that we have been leading since 2017. We first started working with the Facebook AI Research team at that point, which later grew into the [indiscernible]. And that segment continues to be strong for us. We did call out the 8-figure win this quarter as a further validation point that, that translates into Gen AI environment. But I think the broader opportunity set actually has the potential to be quite a bit larger. As the enterprise steps back and says, "Hey, I want to go deploy these models. I want to go deploy this technology, it means a couple of things. One, every bit of data I'm collecting and storing has value. It can't just be shoveled off in the corner and kept cold. And so that's very well aligned with the e-opportunity that we have. And then number two, I have a need for connecting all of the disparate parts of what used to be siloed applications together to do that. And then three, we're also starting to see signs that as these deployments are taking root, they're increasingly being built on modern software stacks, cloud native architectures, and that's where PortWorx is seeing some early traction in these types of opportunities.

Yes. So you mentioned FlashBlade//E. I know Q2 was the first quarter, really strong performance, so maybe just talk about what does that really open up for the company in terms of kind of market opportunity, maybe start there.

Yes, absolutely. So we believe E -- well, with FlashBlade//E, it's really the first step in road map that's going to play out over the kind of next couple of quarters and years of breaking open the disk-based storage market, which today still remains the larger portion of dollar spent and the vast majority of capacity out there in the industry. . If you look at the total dollar spend, probably 40% -- 45% sits on flash, the balance on disk by capacity, that's even more on the disk side. We believe we can go after most of that opportunity today. And certainly, that increases as we continue to execute on our drive density road map. So I think what you're seeing with the first quarter of sales, and certainly, we've commented also on the strong pipeline build, is the demand that's out there for the benefits of Flash to go address these workloads. That's with FlashArray//E coming out. That's really going to complete the e-family, so we can go address the entire set of capacity points. We are today seeing ourselves getting brought into much larger capacity opportunities than even I had personally expected, which is great to see. So from a stand-alone market segment point of view, a very strong opportunity set. The other thing I would say is that E built in a very important piece of the puzzle for us as we go into enterprises and want to compete for a much larger share of their footprint. Every enterprise that we sell, our high-performance Tier 1 systems into has just as many, if not more, bids sitting on disk. Well, now we can go in with a unified platform, consistent software and hardware management and go compete for the entire footprint, which makes us a much more strategic vendor, if you will.

Maybe you can touch on the system comparison, if you would, between E from a system perspective versus disk, I think that would be really helpful to provide some context.

Absolutely. And not to ask the question for you, but one question we do get sometimes is, hey, so I hear what you're saying about E, but I'm also seeing that disk prices are still well below Flash, help bridge that for me. And the bridge is actually very important, which is that our customers, at the end of the day are not buying media versus media. They're buying usable system versus usable system, right? Meaning, I'm a customer, I've got a petabyte of storage. If I put it on a disk-based system, what's my total cost for that, versus a Pure Flash-based system, what's my cost for that. And if you start to unpack it, that's where our advantages, our structural advantages that I highlighted before, working directly with Flash, the software that enables us to do that actually drives a very significant delta in those comparisons. Number one, because of our software, we get much more efficiency out of the Flash than anybody else can with SSDs or disks, meaning we can just ship less Flash and provide more usable storage out of the bits that we ship. Number 2 is our software now allows us to build larger and denser modules. We can put those modules behind effectively the same amount of compute, the same amount of nonmedia costs, can be amortized over a much larger amount of capacity. And so if you think about the cost in the system, you've got the media, disk or Flash, then you've got everything around it, the CPUs, the DRAM, the chassis, the cables, none of that is coming down in cost. If you look at a disk-based system, you might be surprised to learn that in some of these systems, you're going to pay just as much for the sheet metal and the enclosures and all of the nonmedia as you are for the disks themselves. While we can go drive that down considerably, that's what gives us that structural advantage.

I did want to touch on Evergreen One. Kevan, you mentioned that really strong performance again, I think you said doubling year-over-year last quarter. Can you just maybe a question for both of you, but talk about kind of what does this mean for the customer journey? how important is this removal of that friction of upgrade cycles? And what are the kind of financial implications in terms of the model?

Yes. I mean, look, with Evergreen One, if you take all the goodness that Pure structurally has provided customers via CapEx sale, you get that all with the Storage as-a-Service in an Evergreen One model and more. So customers no longer need to worry about -- all they need to understand is what do I need in performance for service and what do I need in capacity. We'll take care of everything else for them from a service standpoint. So simplicity, speed, utilization, you get higher utilization as a customer, the flexibility that offering provides. It's a traditional service offering. And when you think about how powerful service offerings are, it really comes down to simplicity, reliability, dependability in terms of what that offering is enabling for customers. And so as customers are growing their footprint, they can grow under one platform. Our data storage offerings now really cover the majority of what an enterprise needs in terms of all the way from the highest performance to capacity, price sensitive workloads, and we can cover that entirely. And it's made much easier through an Evergreen One storage as-a-Service offering. What that lens for customers is really when they think about the outlay of capital, it reduces their outlay of capital costs. While at the same time being very competitive in terms of utilization costs as they are consuming the data storage services. But from a lifetime value from our lens, it's much more compelling in terms of what that looks like.

And just to add on one bit to that from the customer's lens, I think flexibility that the Evergreen One unified subscription offers is key. I think we're seeing customers really recognize that. Whether it's flexibility to grow at different rates, whether it's flexibility to utilize different tiers of performance over time or whether it's flexibility to take some of their workloads that might start on prem or move into the cloud solutions, the ability to go do that all within and under the same subscription umbrella without another purchasing event, I think that's driving a lot of, again, awareness within the customer base that, hey, Evergreen One is very compelling from an optionality, flexibility point of view.

Great. Maybe to round off the product discussion, just maybe a little bit on Fusion and Portworx, where do they fit in the overall portfolio? And how do you see maybe that contribution kind of trending over time?

Yes, absolutely. So let me hit those a little bit separately, right? So if we look at Portworx, coming up on 3 years since the acquisition, and we're seeing a steady build of demand and drive for cloud native applications. I would say, over the last couple of years, as the industry has been a little bit more constrained environment. perhaps a bit more affected by that environment. But as we gave some comments in the earnings call with the push for new applications connected to AI, we're starting to see a lot of that come back. And so longer term, definite momentum there is, look, the modern application set and developer are overwhelmingly choosing to build on cloud native architectures. And Portworx remains the most complete offering there. When we look at Fusion, I think this is -- we're still in the earlier innings of that journey, but it's a very, very compelling value prop for customers, right? Charlie -- well, the whole team has talked about the value we provide in a consistent software hardware management platform across different parts of the customer's portfolio or estate. What Fusion then will allow customers to do is to build on top of that and provide their internal customers a private cloud-like experience. Said more simply, if you're a storage management team within a large organization, the simplicity that your internal customers are looking to AWS or Azure to get, you can now provide with Fusion on top of pure technology.

Any questions from the audience? So I also wanted to ask about overall demand environment spending. Kevan, maybe just an update there. What did you see last quarter? Any changes, just a little bit of context around the macro?

Yes. Look, the macro backdrop itself has remained relatively consistent with what we've seen exiting Q4. It hasn't worsened anything from a more meaningful perspective or improve, but I think the positives we're seeing within this macro backdrop is -- customers are still prioritizing key strategic projects. Our solutions are a part of that. We saw that with our record performance on FlashBlade across our portfolio as well as our Evergreen One Storage as-a-Service offering. And so that's very meaningful to see that we're still being prioritized with these key projects. And so when we think about and frankly, we do see decisions that companies are making within their budgets to defer potentially some of their investments of -- we've talked -- it seems some sweating of assets with projects that are of lesser priority from a customer lens. That will come back. And so we're well positioned to take advantage of that. I think potentially, we've seen some of that come through on our Evergreen One offering. And so I think that's a plus when we think about the coming out of this macro environment that we're currently navigating. But in the meantime, look, I think we're quite pleased with what we've seen in the first half. Our guide for Q3 is reflecting double-digit growth getting back to double-digit growth, assuming a macro environment, again, that's not meaningfully improved from our perspective. And I think that's only a testament to our -- the strength of our portfolio and data storage platform.

So obviously, we've talked about the product capabilities, how they've evolved. There's a lot more products now. Just on go-to-market. You've also pushed more into enterprise -- so maybe frame what are the key priorities? How do you see the go-to-market evolving over the next couple of years?

Yes. Maybe I'll start and Kevan can add in. I think a couple of things. Really, since Charlie has come aboard, right, we've increased our focus, both from a product portfolio point of view, but also a go-to-market point of view in addressing the enterprise. We serve a lot of enterprises today, but we still are minority part of their wallet share. And so we're now in a position with a much broader portfolio set addressing the entire range of performance all the way down to the most cost-sensitive systems. We're in a great position to go and increase that wallet share. And so a lot more focus on our direct touch sellers in terms of driving those opportunities, having those more strategic conversations, engaging the C-Suite more meaningfully while at the same time, enabling our channel partners to be more independent and go and drive more of that commercial business independently.

I mean, look, the opportunity for us when we think about the largest growth opportunities, enterprise definitely is one of those pillars for long sustainable growth that we're looking to go achieve. And as Rob said, it's really being able to now come to these enterprise customers with the entire portfolio that's very compelling for us. When you think about pure and pure success, we started and grew up as a commercial first and now we're transitioning to enterprise, and you see that traction that we're getting with the Fortune 500 and the growth within the Fortune 500. But the opportunity is significant for us within the enterprise because of the portfolio we're driving and these differentiated outcomes that we're driving that now reach the C-suite in terms of the business value, whether that's security, business continuity, power savings, what we're going to do in terms of cost savings for customers across the entire spectrum for our portfolio. And it's not just power, which is compelling. It's not just data center space, which is compelling, but labor savings, which is significant. The power of Evergreen and the savings around Evergreen and nondisruptive upgrades is compelling, break fixes. Rob talked about the fact that with our Purity software and our DirectFlash modules, you just -- we don't have as many parts and nonmedia parts. Well, what does that mean? Well, they break down less and we continue to optimize Flash even after it's been sold to the customer, which you can't do in an environment with SSDs or disk and so all these advantages that we're driving for our customers now are reaching that C-suite and are really resonating with the C-suite of our enterprise customers. So we're really looking forward to that opportunity that's ahead for us. The one other thing I would highlight specific to the enterprise group is the flexibility of our business model. We offer an incredible amount of flexibility through Evergreen One, which is a pure service offering. We have a tier where we Evergreen Flex, which a company can purchase the hardware at white box prices and then pay for the consumption. And so that flexibility that we're continuing to be creative and managing, including potentially focusing on Purity and allowing Purity to be more emphasized within our larger enterprise customers.

Great. I wanted to ask about the hyperscalers. They have their own kind of software, I think, more on the harvest side. How difficult would it be for them to kind of invest and kind of go the route of Pure and develop DirectFlash? And I guess, do you view them more as a competitive threat or more as an opportunity?

That's a great question. Definitely more of an opportunity. I think when you look at the hyperscalers, it's pretty clear. They see the same road maps we do that, hey, the runway they have left on disk doesn't look great. They know they've got to switch to Flash. They know that SSDs are really not going to be the path for all the reasons I laid out. And so then you come to the question of, okay, well, build versus buy versus partner and a couple of things. One is the -- to truly deliver the outcomes that we're delivering with our Purity software, DirectFlash -- this has been the pinnacle of over 10 years of R&D investment and IP we've built up since day 1. And the second thing is it's not just one hill you've got to go climb, right? We work with multiple NAND vendors. Any of these large players are going to work with multiple vendors. Every chip from the same vendor, every generation behaves differently. Each vendor behaves differently. We've done the work. We've laid the groundwork to go and incorporate the differences between all of the Flash chips out there and incorporate them into our software very quickly. That's all work that would have to go be done. And so against that backdrop, at the same time that the industry as a whole is pressured on spending, everybody is focused in potentially other areas, we just see this as a tremendous opportunity because we've got the technology now. It's ready to go. We see that as definitely an opportunity for us.

That's great. Maybe to close things off. A question, I guess, for both of you, but what do you feel is the most maybe underappreciated element of the story just to give some context to newer investors here.

Well, I'll start 2 things. I think one, we've talked a lot about today, which is the software differentiation in Purity relative to SSDs and kind of how we're set apart there. And I think the second is the power of Evergreen across all of the subscription models.

Yes. I would probably echo that the advantages, the compelling advantages that differentiate us with Purity. I don't think folks really are appreciating how significant that is and what that allows us to do, which includes being very creative with our business models. And then going back to the hyperscaler Titan opportunity, our ability to be very flexible with our business model, meaning even enabling hyperscalers to completely control the supply chain, we can do that with our business model, and it's all being driven by our Purity software. And I think that's very compelling, and I still think that's an underappreciated view from investors currently.

Rob and Kevan, thanks so much.

Appreciate it.

Appreciate it.