Nutanix, Inc. (NTNX) Earnings Call Transcript & Summary

Great. Good morning. I'm Sanjit Singh. I cover infrastructure software for the Morgan Stanley software team. Super thrilled to have the management team from Nutanix here. We have a CEO, Rajiv Ramaswami; and Chief Financial Officer, Rukmini Sivaraman. Thank you both for joining the Morgan Stanley TMT Conference.

Thank you. Glad to be here.

Awesome. So before we get into it, real quickly, for important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. So let's get into it. I think in this environment where there's a lot of things going on across software, across supply chains, global political activities going on, I think a lot of people getting back to like sort of first principle thinking and like what the companies we're investing in sort of actually do and where the value is being created. And so with that context, as, what would you describe as the core problems that Nutanix solves for customers today that allowed it to become a multibillion-dollar company? And where do you want to take this business going forward?

Yes. Look, I mean, at a fundamental level, we help companies and organizations around the world modernize infrastructure, right, figure out an operating environment on top of which they can run their applications, manage their data very effectively regardless of where all of this is happening. Some of it happens on-prem, edges, public clouds, but we are providing this platform that creates a lot of value in terms of simplifying how companies can run and operate their businesses. Now as we move forward, I mean, if you look at the specific demand drivers under that in terms of value creation, going from legacy infrastructure to modern, cloud-like simplified automated infrastructure, that's the first part, enabling them to run both traditional applications and modern cloud-native applications. That's the second part. Third is there's a whole new generation of AI applications, inferencing and agentic applications that are still on the cusp of being developed in the enterprise, very early days. And those applications, again, are also going to be these hybrid applications that run across different locations. And we want to be a platform that can run those applications very well as well. And you saw some of that in some of the recent partnerships that we've announced.

Yes, which we'll definitely get into. Let's review some of the themes coming out of Q2 results. And on one hand, bookings remained strong. It's a kind of clear message that you saw both in Q1 and in Q2. But Q2 sort of echoed a similar theme on that while the bookings were strong, you had to cut your revenue outlook for the second quarter in a row. The reasons are different. And Rukmini, I was wondering if you could speak to some of the server supply chain issues you're seeing today? And how is that impacting customer procurement behavior?

Yes. Thanks again for hosting, Sanjit. Great to be here. So what we sell, as Rajiv said, is infrastructure software. And so while we're a software company because we're infrastructure software, our software runs on servers and hardware. And what we saw as we got later into our fiscal Q2, which is our January quarter, is that -- and I think this is not news to anybody, the sort of supply situation in the market has gotten more challenging. And what that's meant for our customers is that they have longer lead times in getting access to servers on which they would typically run our software. So we saw -- some of that was a headwind in our January quarter, but was more than offset by bookings sort of better than expectations and some other factors. So we were happy to sort of beat the high end of our range for our January quarter. What it meant to your question, Sanjit, for the second half of the year was that we think this is going to continue. We don't think this supply chain situation will necessarily resolve itself in that time frame. And so while we actually have internally raised our expectations around bookings for the full year because we do think demand is there in the market, revenue timing is going to be later than we had previously expected, which is what we saw and which is why we had to adjust our guide to reflect that.

Can [ I just a ] quick follow-up on that, the confidence in bookings versus the revenue recognition. If there's uncertainty that customers have in terms of procuring the hardware, why wouldn't that also potentially impact bookings? Why don't they say, hey, we'll just plug this off? Or is there sales motions, incentives you're putting in place to incentivize those bookings?

So there are a few things. So there's -- I think on the one hand, in many cases, customers may have gone through all of their internal decision-making processes, procurement processes, and they are ready to make a decision. And so in those cases, they're ready to move forward and in some cases, give sort of us the software order, but also get in line for the hardware, right? Because they know these lead times are long, prices are likely going up on that front. So why not get the price today and get in line for the hardware as well. So that's one reason. I will say there's also the other side of that, which is sometimes people are realizing that, oh, I went through my procurement process for a certain price, but now hardware has gone up even since I did that analysis. So there's puts and takes on that front. But the net result we saw certainly in our fiscal Q2 was that bookings were higher than we had expected. What are we doing? I think that's part of your question, Sanjit. So what we're trying to do with our customers is we've always been about choice -- so we're giving them a choice of server platform, so they can go out there and try to find the one with the best lead time and the best price that they can use to run our software. Over the last year or so, we've also started supporting external storage, which was not the case before, where our architecture was software-defined cloud-defined storage was part of our reason for existence, but now we're supporting external storage. And what that means in many cases is customers can use their existing hardware. So they may not need to actually change it out. So that's another thing we're doing. We also have a solution called Nutanix Cloud Clusters or NC2, where customers can use bare metal -- public cloud bare metal to run our software. That seems to have more availability now. That's another option. So we're trying to make as many options available to customers as possible, while also, again, to your point, being cognizant of where they are in their decision-making. We also have some commercial levers that was part of your question as well, Sanjit, where we say, for example, even if they're only buying the software from us, where we have a mechanism called a future start date where you can make the commitment today, we get a committed order today, but the start date for the licenses out in the future. We try to be judicious about how we use that, but that's another option we have for customers.

Yes, that's very clear. Maybe the last question on this specific point coming out of Q2 is just how do we compare the cycle that we're going through now? I mean you guys have been around for a while have seen multiples of these supply chain issues, probably pandemic was the last one. If we kind of compare to the pandemic, any sense of like timing on how this cycle could last versus some of the past cycles you've been through?

Yes. I think there's a supply side to this and the demand side to this, right? So on the -- starting with the demand side, obviously, what's triggering this is the massive AI build-out that we're seeing from a handful of very large players, including the hyperscalers. And to the extent that continues over multiple years, right, that's going to continue to put pressure on the supply side. Now I think if you're a supplier of whether it's memory or CPU, right, and the condition is slightly different, then you're looking at this and saying, how long will this demand last? And can I -- should I be building incremental capacity to go fulfill this demand? And on the semiconductor side, as you know, right, taking -- building new fabs and it takes time to bring on new capacity, especially with related to more fab capacity. And so that's -- that itself is a multiyear cycle, right? So this time around, I think if this demand continues to last, I think this is not a short-term couple of quarters kind of phenomenon. We do believe this can last much longer than that. On the other hand, if there's a hiccup on the demand side, then I think things could normalize faster, right? So that's kind of what we see. But from what I can see right now, it does look like this is going to continue for a bit.

Yes. That's great. That's a great perspective. The other big news coming out of last week's earnings was the strategic partnership with AMD. And here, we have AMD making an equity investment in the company to the tune of $150 million and an additional $100 million for joint engineering and go-to-market activities for AI inference at the edge for both large enterprises and service providers. So Rajiv, I was wondering if you could give us a back story on how this partnership came together. And from a commercial standpoint, how big of a revenue contributor could this be for the business over time?

Yes. We have been working with AMD for many years now as a partner because we -- our software works on their CPUs -- not the GPUs, but the CPUs for regular compute, right? It's -- our customers have a choice of Intel-based platforms or AMD-based platforms that they can procure through their server vendors. So this is not a new relationship for us. It's an extension of the existing relationship. Both of us see an opportunity for inferencing workloads and agentic workloads in the enterprise and that we are in the early stages of this. I think this market is going to continue to grow. And from our perspective, we want to be able to offer an open set of choices for our customers to be able to pick what they would like to do and build solutions off. And so AMD looked at this, and they looked at Nutanix as a great partner to enable them to take their GPU ecosystem into the enterprise market. And so that is the genesis of the partnership where they are investing in us, first of all, to get this full solution built, which is our infrastructure software stack, combined with their GPU ecosystem, their software and ecosystem to put together a full solution. So that's the first part of the R&D investment and then joint go-to-market efforts to take that complete solution to market. So that is the commercial side of the arrangement. And then, of course, in addition, they also put in an equity investment of $150 million. And from our perspective, that aligns the 2 companies. AMD has invested in Nutanix being successful and it gets the companies to work together better.

Yes, that's a great, great way to incentivize alignment. So let me just talk about Nutanix and sort of the question that everyone is asking across software sort of defensibility and sort of the age of AI. When this year started, investors came to me he's like, Sanjit, you have a great coverage universe. You don't cover any of the seat-based models, pure infrastructure, data platforms, compute networking, that was the beginning of the year. And now every single company is sort of being -- the question is being re-asked about any company's defensibility, software company's defensibility and the age of AI. And so in terms of some of the flavors that I get asked about, I thought you'd maybe get your perspective. And it sort of goes like this. To what extent can customers use agents in combination with open source software and open source tooling to run and manage their own compute environments more cheaply than using a platform like Nutanix?

The answer to that is very little actually because down in the stack, first of all, there's a lot of work to be done to optimize your software to run on the hardware. And we do that, right? It's very core to how the hardware platforms themselves are built. So we do a lot of that. Second, we do a lot of work to build resiliency for enterprise-grade mission-critical applications. Keep in mind that the kind of applications that you're seeing from a seat-based SaaS application that are very different from what runs on our platform. People are running their businesses on Nutanix software. And these are -- if the system fails, then you're going to lose your business or you're going to have create significant impact. Your manufacturing might stop working, right? Your ATM might stop dispensing cash. So when you have this kind of a situation, right, it's very core infrastructure. And people are very careful, right? They're not going to simply -- they have been, for example, open source capabilities available for many, in fact, our own hypervisor built on open source. Our cloud-native platform was built on open source Kubernetes. But there's a lot of work that we have done on top of all of that to make it enterprise grade, mission-critical. And we don't see that being displaced anytime soon by agents or others, right? So that's not as much of a factor for us. I mean we -- in fact, we look at it from an opposite perspective as AI creating potential tailwinds as more and more customers build these new applications with agentic AI and inferencing. They need a platform to run on, and that's creating additional opportunities for us to capture that business as you saw, for example, from the AMD partnership.

Yes. The point around running your business and executing ATM transactions, you need some pretty precise deterministic outcomes versus just relying on an undeterministic technology that GenAI can be. In terms of the AI opportunity itself, there's a lot of conversations on the risk side of the equation for companies. But from -- as we move more towards an agentic world, what are the opportunities for Nutanix in that construct?

Yes. And I think as people start building these agentic applications, but even before agentic, I would say, simple inferencing applications, these applications are going to be built where the data resides typically. And not -- data is going to be everywhere. Some of it in the public cloud, some of it on-prem, some of it being generated at the edges where data is being generated and you need real-time inferencing. So in all these cases, again, we have a platform that makes it really simple for companies to build and deploy these agentic applications. And our job as an infrastructure software company is to create a turnkey environment on top of which people can build and run these applications. And that's what we are aiming to do. We have a solution in the market with a full stack, which is really our full cloud platform that's running on GPU-based servers instead of CPU-based servers as we've traditionally supported, right? And so that opportunity, I think, is still pretty early. We think companies are still very early in building these applications. Today, what we see is the first set of simple use cases being deployed. These are not multi-agent use cases. These are simple use cases like customer service type use cases, document search and summarization type use cases. We had a bank, for example, that was recording conversations by salespeople and had an LLM go through those conversations, summarize them, but also point out places where there might be potential noncompliance. This is a homegrown application runs very simply. And a lot of these applications, by the way, don't require massive clusters. They run on small clusters, 4-node, 8-node GPU clusters that can be cost effectively run with privacy and security in an enterprise.

That's great. I want to switch the conversation to the sort of unpacking the growth drivers for Nutanix. But just sort of at the high level, the basic growth equation, the existing customers versus the new logos. Looking at like kind of the net retention rate, that has come down steadily over the last several years, recently sustaining in the high single digits. I guess the question here is basically how much opportunity do you see in the existing customer base? And then we'll speak to the great new logo performance you're seeing. But just in terms of the existing customer base, how would you sort of frame out the runway within your existing customers?

Yes. I think you've got to sort of, I think, segment the customers a bit there on that front. So if you look at the top end of the enterprise, the enterprise customers that we have, and we have plenty of them, about 1,000 of the Global 2000 are our customers. Our penetration levels are relatively small in the customers that we have over there, right? So we typically get adopted for 1 or 2 use cases. There's plenty more. So there's a lot of expansion opportunity in that segment. If you go to smaller customers, they tend to be more of, let's take Nutanix and deploy them everywhere. And so less expansion opportunity. Now so expansion comes through in multiple forms, right? One is expansion of different use cases, which is what I talked about, but there's also expansion of more of the same use case. There's expansion by adding more of our portfolio of products, right, more products that we can sell to the same use case. So it's a combination of all those 3 factors that come into play. So I would say at the top end of the pyramid, there's plenty of opportunities there. At the bottom end of the pyramid, these customers are largely already with us. Do you want to add anything to it?

Covered.

So let's talk about the new logo side of the equation, which has been going well, I would say, for several years now. And in particular, last quarter was a great new customer acquisition quarter. A little under half of your growth is coming from new customers. What gives you confidence that this part of the growth equation from new customers will prove durable?

Yes. Look, I think if you look at, for example, the installed [ base ] -- I mean, the addressable number of customers for us in the enterprise is somewhere between 100,000 and 200,000 customers. And we have 30,000 customers today. So there's plenty of customers that we are not in. And so I don't see any reason why we can't continue to capture customers there. And if you look at the kind of things that we're doing also, of course, we've done a lot of work on our sales force to be able to go out there and target these new customers. We have more channel work that we've done over the last few years to bring on board more channel partners and more routes to market through channel partners, through managed service providers. We have more strategic relationships in place now with the likes of Cisco and Dell and even the public cloud providers like AWS. So we have more friends in the market overall, I think, at this point. And certainly, that allows us to go out there and capture the opportunity. There's also trigger points like people wanting to get out of VMware, for example, that also come into play in terms of this being a factor in terms of these new logo additions.

Any -- what's -- in terms of the strong performance that you've seen, not last quarter, but I'd say, over the last couple of years in terms of the new logo side of the equation, what's been driving that? Is it the VMware displacement opportunity coming around? Why has that been so strong? Or is there maybe something on the sales side that you guys been executed?

I mean I think it's a combination of multiple things. All the factors that I talked about really have been at play for the last couple of years. Yes, certainly, there's a VMware effect, no doubt about it. But there's also all the fact that we have more feet in the ground. We have more partners. We have more strategic engagements. All of those, I think, together, I think, are what's bringing out this new logo performance that you've seen.

Awesome. Let's sort of mark-to-market on the VMware displacement opportunity. I mean you guys have been very clear about describing the opportunity here as a multiyear opportunity that sees steady progress with each passing year. I used to come for VMware. Those ELAs are 3 and 5 years. And so that kind of speaks to kind of the timing. From your guys' perspective, what inning do we stand in the VMware replacement opportunity? And what's the profile of VMware customers that will choose to move to Nutanix versus stay with VMware?

Yes. I'd probably say we're in the second innings of that opportunity. We have been migrating a number of customers. I mean, in fact, the vast majority of the new logos that we are adding are VMware customers that are migrating. So that's been a trend for a while. Now that said, like I said, I think the first innings probably is the acquisition closed in 2022, and it's now been 3-plus years now. And people have -- the first round of customers have done migrations. We've done several large migrations. When I say large, I mean like 100,000 CPU core type migrations. These are fairly large companies. We've -- for example, one example is a large insurance company in North America, fully migrated out within less than a year from VMware. Now as you look at the profile, so there's a lot to continue here. I mean now we're in the second innings. Many customers had -- didn't move initially, thinking what would happen, they had time. And then they've just done their first Broadcom renewal and now there's more. They're starting to think about, okay, what do I do next? And what time do I make? So we are engaged with a lot of those customers now. And with respect to the profile of customers, again, the smaller customers, we tend to do outright migrations. At the very top end of the pyramid, if you're like 100,000 cores, we can migrate fully. If you're 1 million cores at the very top end, for example, those customers, I think, will stay with Broadcom for a long duration. They're not going to get out. The opportunity for us in that set of customers is can we get a footprint in some subset of use cases where we can go win. And we have seen many examples of that also.

When we think about what -- when we think about like the hypervisor opportunity, like one of the ways to get some traction is VMware stopped sales of stand-alone hypervisors. How much opportunity is there to fill this particular need in the market? And what's the impact on growth that the stand-alone sales are having for the business today?

Yes. I mean I think that's a very good point, right? Because until a few years ago, we were largely selling a full stack HCI platform, right, which included only our own storage. We did not work with third-party storage. A lot of VMware has deployed. In fact, I would say 80% of the installed base for VMware is VMware hypervisor working with third-party storage. Now we didn't really have the opportunity to support that until recently because we were competing against those storage providers. But now it's much more synergistic relationship. They see a need for a hypervisor and more than a hypervisor, I would say, a compute platform. And we have evolved as a company to saying we're no longer an HCI company. We are a company that has a full cloud stack, of which storage is an option. You can use our own internal storage or you can use external storage. So from that perspective, that opens the door quite widely for us to go into existing deployments and not have to do wholesale architectural changes, right, but work with exactly the same hardware that our customers already have. They already have their storage, they have a set of servers, and we can just be a plug-and-play software replacement. So that's the opportunity. And we are in the early stages of capturing their opportunity. We support 2 platforms today, Dell PowerFlex and Everpure. And we have more coming online. We have Dell's mainstream platform, PowerStore coming online this summer, and we'll be doing more of that. And so that opens the door, right? It opens the aperture for us to get in the door easier, easier migration without having to go change hardware. So it's a significant TAM expansion opportunity. And the economics on that, by the way, again, there's a whole range, right? There's a floor price for what just a hypervisor is and there's a price for the full stack. And it's kind of somewhere in between, right, on a per node per core opportunity.

And it brings us a whole segue of questions around what you guys can do outside of the VM ecosystem within HCI as well. But one of the things that came up a couple of quarters ago, I think you sort of framed the competitive market between you and the Red Hat, IBM Red Hat and maybe 1 or 2 others, but mostly kind of IBM Red Hat. And the dynamic you described a couple of quarters ago was that Nutanix by far wins most in sort of lift and shift migrations and Red Hat does better for kind of container-based replatforming. When thinking about the opportunity, what percent of the market is electing kind of the migration route versus doing more of a containerization replatforming?

Yes. I think the vast majority of the market, simply, if you're looking to migrate from VMware wants to do it in the easiest way possible. The easiest way means you don't touch the application. You don't impact. You don't have to go bring your developers to work on making something. The IT teams are able to do a simple migration, almost entirely transparent to the application and get the business moved over, right? And that's what we see, and that's what our platform can do. That's why these migrations are easy. They can be automated, they can be done at full volume. We've got to take an existing application and refactor it to containers or replatform it. That requires more work. And of course, there's now talk about AI tools being able to help with that for sure, no doubt about it. But even with that, the application does get impacted, right? And you've got to work on the application, which takes more time, more effort. And the beauty of the Nutanix platform is that we support both, right? If a customer wants to modernize and get into a container platform, our platform supports both virtualization and containers. And also, a lot of these applications that customers run on VMware are mission-critical applications. And so they need a robust platform and a proven platform that they can run it on, which is where we shine.

Let's talk about the opportunity for cloud-native workloads and the opportunity outside of the virtual machine ecosystem. Specifically with regards to NC2 and Nutanix Kubernetes engine, your ability to support and run and manage cloud-native environments. What's the traction today? And what do you see -- what do you think it will take to unlock the opportunity around cloud-native workloads?

Yes. On the Kubernetes front with our platform, it's called the Nutanix Kubernetes platform. It's reasonably early days, but we've been very encouraged by the fact that last year, the first full year we had the solution in the market, we got really good traction. We've been continuing to acquire new customers at a good clip. We continue to grow that this year. Also, what is very exciting is that's also the underlying foundation for all these AI inferencing workloads and agentic workloads, right? Because all of the AI workloads effectively are containerized, right? They're running on top of containerized applications. So the Kubernetes platform that we have is a foundation for all the AI workloads that we're building as well. So there's a lot to come.

Awesome. I want to spend a couple of minutes talking with Rukmini around profitability, capital allocation. One of the questions that we're asking a lot of management teams just given the downturn in the market, particularly for software companies is what is Nutanix message when it comes to share repurchases, the level of share dilution investors should anticipate going forward and the importance of getting to GAAP profitability?

Yes. So first, we are GAAP profitable. We have been for a bit. So really happy to share that. I also think that we look at SBC as any other expense and have done so for a long time now and have done a lot of work to manage that as any other expense line is managed. And so our SBC as a percent of revenue is sort of low double digits right now. It was, of course, much higher a few years ago, and we'll continue to do that. And I would say, I think as with most companies, for us, it's a balance between doing right to get more leverage in that line, but also make sure we're able to hire the right talent that we need in a market that's changing every day. I think to your point, Sanjit. So it's something we'll continue to monitor very closely. On -- in terms of our overall capital allocation philosophy, look, I think we've said that capital return to shareholders in the form of share buybacks continues to be something that we anticipate doing. We did a -- our first ever accelerated share repurchase in December for $300 million. And folks should view that as a testament of where we think the upside is in the company and where we think our opportunity is going forward, and that continues to be the case today given what the markets have done. So yes, absolutely. I think share buybacks will continue to be a part of our capital allocation, capital return approach. We're also doing an Investor Day in early April and expect to sort of share more of our overall philosophy around capital allocation, but share return -- capital return is absolutely a part of that.

And just as a follow-up, the first 2 quarters of the year, you guys have come into a period of uncertainty. And while the revenue outlook may have come down, the bookings have been strong, but also you guys have held the line on operating margins, right? And you guys reiterated your operating margin guidance for fiscal year '26. What are you guys doing to hold the line on margins as top line visibility has been a little bit more uncertain through the first half of the year?

Yes. There, too, I think it's -- there's 2 sides to that, Sanjit. One is, look, we've talked a lot here about the growth opportunity, how we think AI is actually a tailwind for us in terms of landing more workload. And so for all of those reasons, because we see a big growth opportunity ahead of us, we want to make sure we're investing in the business enough to capture that opportunity. Now at the same time, we do believe there's -- from an operating margin standpoint that there's more leverage we can drive in the model. We don't think we're at what we would consider steady-state operating margin for our business. So we'll continue to drive those improvements over time. So I would say -- and I think the third piece of that, of course, is efficiencies and how can we have our people continue to be more productive and more efficient in what they are doing. And obviously, AI is a big part of that as well internally. So I think those are the 3 pillars. And one is making sure that growth is the #1 priority. So we're investing in all the right places for that. Secondly, just driving leverage in the model, and we can talk more about that in terms of we think that, for example, there's more productivity from our sellers, but also more broadly in the organization. And the third piece is efficiencies and making sure that we're adopting AI wherever we can as a company and driving efficiencies through that.

That's fantastic. I want to spend the last couple of minutes going back to the point you made, Rajiv, around is the partnership strategy. So between your OEM partnerships with Cisco, Dell and others, your co-selling relationship with Pure Storage, what percentage of the business come from these channels today? And where do you ultimately want to get to over the next couple of years?

Yes. I mean, today, it is a minority portion of our business, right? These partnerships have been ramping. Pure is just literally now -- I mean, a couple of months in the market. Cisco has been with us now for a couple of years. Dell, it's interesting. The dynamic there is we compete in some areas, we partner in other areas. And the more we can actually partner, the more synergistic it becomes as a selling motion. So with all of these, I think, look, I mean, we've got to -- it's always both routes to market, right, which is -- we have our sellers who are out there creating demand. And we want as many people and as many friends in the market who are helping us, right, in terms of being able to go out there and represent us and be able to go out there and push our products. And we're certainly seeing a lot more of it now than we had a few years ago. And the business, I think, I would say the sell-through business, that's what we call it, through these OEM partners has been steadily growing every year. Now it's still a small portion of our total business. And I don't have a particular model in mind to say this much of the business has to come from partners versus this. But I do would like to see that proportionately grow, right, for us and get to a larger number than where it is right now. And I would like to see that grow faster than continue to organically invest in driving -- hiring more and more salespeople, right? We want to get more leverage, use it as a way to get more leveraged in the market.

With respect to the Dell partnership specifically, the PowerFlex solution is now GA. PowerStore, I think, is coming -- PowerStore is coming up in the summer of 2026. What customer segments do each of these solutions open up for Nutanix? And how big can Dell be for the company over time?

Yes. So great question there. PowerFlex is very much deployed at the top end, I would say, several hundred kind of accounts with large footprints. And we've captured several wins there already, and we'll get some more over time there. But it's not a mass market platform. PowerStore is their mass market platform, right? It's sort of with the equivalent of their midrange. It replaces all their midrange storage, and it's sort of a big chunk of their market, right? So as we -- and a lot of their selling motion is tied to PowerStore and selling PowerStore. And so the more we can be aligned with PowerStore, I think the better we are going to be with Dell, right? And they're going to be more motivated to sell our platform with PowerStore. So I expect that to grow. And I think that will be a key unlock as we move forward with this summer, I think, once we get that to market.

And can we compare and contrast the partnership with Pure Storage versus Dell? What is the strategic rationale of partnering with Pure Storage? And how does that compare to your other OEM relationships [indiscernible] Cisco?

So again, I think the context for Pure and every other storage provider is by having our product work with theirs, we are able to get into our customers without any hardware replacement, business as usual, just replace software, right? That's the value proposition. And from their perspective, again, they don't like VMware either, right? I mean they're trying to go out there and push the whole stack and displace Pure Storage and other storage providers. So -- so that's the synergy that we see with Pure and with Dell. Now I think with Pure, we have a co-sell, right, which is we don't -- they're not reselling our product. With Dell, we co-sell, but they also resell. So -- and they have a huge sales force in the market. They have a sales force. They have a broader sales force in the market. So they do both resell and co-sell.

Awesome. Maybe in our last minute or so, just talk about what are the right metrics to assess growth. So as I took coverage of the company, Rukmini, I came from a software background. And so I and my head things in ARR, net new ARR constructs. And when I hear the discussion on Nutanix, a lot of it goes to obviously like revenue, but then we go down the rabbit hole of all the various permutations of RPO and cRPO. And so from like your guys' perspective, maybe give us a sense of where investors should look at to sort of assess growth. Maybe it's not one metric by itself, maybe it's a combination. But when it comes to understanding the growth dynamics of the business, how should -- which metric or set of metrics investors should sort of look at?

Yes. So I would say, I think I agree with you that ARR, net new ARR is a good metric in sort of steady state. And what I mean by that is when there's not some of these timing differences that are arising today that we talked about in our last earnings call and here, Sanjit. So absent that, I would say ARR, I would agree with you that ARR, net new ARR is a good way to assess the health of the business because it's independent of duration, et cetera. Given all of those, though, I think we have tried to give you all a little more information to understand the underlying health of the business because there is a timing difference right now between bookings and ARR or revenue, for example, in that we expect -- and we said this as well in our earnings call, right, the TCV bookings growth, we expect to be higher than our revenue growth this year, for example. And so we felt it was important to convey that so folks can see the underlying strength in the business even while some of these timing differences work themselves out over time. So I would say in the long term, ARR, net new ARR revenue, of course, everybody needs to model the P&L. And so that's a given anyway. And then I would say in the near term, folks should also look at something like an RPO, which incorporates not just deferred and so on, but also looks at the bookings performance, which has been quite strong actually for us that we've tried to highlight.

Well, I certainly appreciate the extra color that you guys have given as we try and understand some of the near-term kind of headwinds going on, on the timing differences. So I appreciate that, and I appreciate the conversation -- both of you giving us an update on the Nutanix story. Thank you so much.

Thank you.

Thank you for having us, Sanjit. Thank you.