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

Good afternoon, everyone. Thanks for joining our fireside chat. I'm Jason Ader from William Blair, and I'm pleased to introduce Charlie Giancarlo, Chairman and CEO of Pure Storage; and Kevan Krysler, CFO. Before we begin, I'm required to inform you that a complete list of research disclosures or potential conflicts of interest is available on our website at williamblair.com. Also, if you have questions, please put them in the chat, and I will get to as many of them as I can. With that out of the way, Charlie and Kevan, thanks for being with us today.

It's our pleasure. Look forward to speaking with you all.

Great to be here.

Let's start out, I guess, with you, Charlie, on the takeaways from the Q1 report from last week.

You bet, and thank you. We are really pleased. We had a very, very strong Q1. And it was up 12% year-over-year, and that's over a very tough compare last year. We had a very strong Q1 last year because of the beginning of the COVID pandemic, which actually accelerated our sales in Q1 last year. So this year, Q1, which as, I think, most of the listeners know for us is February through April, was still under 100%, what I call, COVID rules, meaning that no real letup from everyone working at home, shelter in place in nearly all parts of the country and actually nearly all parts of the world. And despite that, a really strong Q1 for us. And we're seeing increasing strength. What was nice was that the strength was broad-based and balanced across our entire portfolio across the world, so all the theaters and a lot -- and all of our customer segments. Something else we saw, which was very welcome, was renewed strength in our commercial market. For those of you that have been listening to us quarter-by-quarter, you know that once COVID really took hold in the second quarter -- or second quarter last year, commercial really was much more highly affected than enterprise. And we started to see, if you will, glimmers of it coming back in our Q4, and it really came back quite strongly in our Q1. We also saw great strength from what we call our channel-sourced, that is -- that our channels were sourcing and initiating more and more of the sales with more and more customers. And with that contribution, we had 330 new customer adds in the quarter, which was 15% more than a year ago. And that was true across our product line. FlashBlade, for example, acquired more new customers at a faster rate than a year ago. And overall, over half of our new customer business was partner-sourced. And the last thing that I'll mention is that subscription services were really strong, up 35% year-over-year. As you all know, that's our Pure as-a-Service in our Evergreen subscription service, along with Cloud Block Store and Portworx. So we're continuing to see strong momentum there and strong momentum across our product line and, in particular, FlashArray//C bookings continued at our record pace, with both FlashBlade and Portworx continuing very strong momentum. So overall, a bang-up quarter. We think all the metrics are looking good for the remainder of the year.

All right. I want to get to some of the product-specific questions as we move through the Q&A here. But maybe before we go there, you mentioned the commercial segment showing improvement. What percentage of your business comes from commercial?

It's a bit less than 50% now. We generally don't break it out with specific -- we did mention recently that enterprise is now over 50%. We expect that to continue. But commercial is still a very significant proportion of our business.

And the business has shifted over time really since the IPO towards the enterprise?

Correct.

Can you just talk about why that's happened? That's obviously been intentional, I would think. But what's been behind that?

Yes. In a scaled business such as ours, you would expect enterprise to be a bigger part of the business overall, right? So it's really based on 2 things: one is just our maturity as a company and the breadth of our product line, but the other part of it is purely operational on our part, that is, the investments that we made, both on the product line to be able to support the capabilities that enterprises expect as well as in our go-to-market teams to be able to support those large enterprises with those products and to be able to penetrate against very tough competition that's been in there -- the incumbents have been in there for a long period of time. So really, it's been those 2 investments that have allowed us to penetrate the enterprise. Now we're not -- we expect the commercial will continue to grow. And part of that is really developing our channel and our channel selling capability to take up, if you will, more of the weight of initiating and closing business in the commercial market.

Got you. Good. And let's talk about your subscription offerings. What are the -- what's the composition of your subscription services line? And maybe talk about each of those areas and what you're seeing in terms of adoption.

Yes. Maybe I'll take that, Charlie. We -- again, as we've talked about, Jason, our largest category is going to be Evergreen subscription, which is attached really to every sale of array. And that provides customers the ability to continue to modernize their array throughout the lifetime of that array without requiring any type of disruptive upgrades. That offering continues to be running at scale, has -- I think our customers really appreciate the fact that they're utilizing these arrays for multiple years without having to worry about moving forward with a refresh or any type of disruptive upgrades. From there, we're seeing a significant amount of momentum with our Pure as-a-Service in unified subscription. It's important to -- in terms of highlighting the unified subscription that we are selling both the Pure as-a-Service, which is the on-prem subscription, as well as Cloud Block Store, which enables customers to move workloads either to AWS or to Microsoft Azure without having to have separate agreements or contracts, so the pricing, all the commercial terms, both for the on-prem capabilities as well as the cloud, are all embedded as part of one unified subscription. From there, we have Portworx, which, again, Portworx started as term-based licenses that, over time, we're also looking to increase the consumption-based business model for Portworx. That's much smaller in terms of financial contributions to the top line, but very important, obviously, for us from a strategic perspective.

Great. And let's talk about Portworx. The biggest acquisition you guys have made. What does it bring to Pure? And thus far, what's the market reception been?

Let me start with the second part and then go to the first. So the market reception has been very strong. We could see that in several different metrics. One is that we're seeing both Pure customers as well as non-Pure customers continue to take advantage of the capability. And we're seeing it is while we offer a premium version of the product, and that the downloads on that have been really tremendous and that, of course, is a very good sign of things to come overall. We haven't given a lot of very -- of numeric metrics on Portworx, but we have indicated that in every quarter that we've had it so far that the product has sold well above its original -- the original plan of Portworx itself. And Q1, under Pure, was another very good quarter. And the last thing I'll mention is that we also announced at our //Accelerate conference just the last couple of weeks that we've now fully integrated Portworx into Pure, and that there are really 2 elements of that. One is Portworx is now completely monitorable and configurable through the Pure1 website, the Pure1 SaaS management platform that we have for all of our products. And secondly, now customers can utilize not only the Portworx software-defined storage to manage data on containers, but completely transparently, they can also use either FlashBlade or FlashArray, and data can be migrated from one to the other transparently to the application. So this allows customers a great growth path, from the initial, let's say, development with just a few developers using a very inexpensive, even a freemium model to test their applications on, all the way through high scale, high performance production levels on FlashArray and FlashBlade.

Great. And have you actually seen from your customers any kind of, let's call it, inflection point in demand when it comes to microservices and Kubernetes storage?

We've seen an inflection point in terms of new developments. I wouldn't say an inflection point yet in terms of storage demand. But the vast majority now of enterprise development of new applications is taking place on containers and Kubernetes. So that's a very big inflection point. Of course, it takes time for developments to become full production, and it takes time for any new production to take on the scale to be truly noticeable, given the installed base of existing applications that customers have. But in terms of -- but I think that, of course, it is, if you will, the leading indicator of what's to come in the future because it's the vast majority of new developments now.

And just for the nontechnical folks in the crowd here, why do you need something like Portworx? Why can't you just use traditional storage?

Yes, it's a great question and a nonobvious one. So I'm glad you asked the question. Containers are very different from both virtual machines and different from what's considered to be standard operating system-type environments in that because containers are based on microservices, and people shouldn't get confused. They could be -- containers could be used for any type of application, including databases. The fact that it's called microservices or the fact that they can be designed as microservices shouldn't be -- you should not allow that to confuse you. The big difference with containers though is that they come and go very, very rapidly. So whereas with a VM or a -- well, in the case of bare metal environment or operating system environment, systems get set up, and they stay that way for months, if not years, okay? In a VM environment, generally maybe days to weeks that an environment will stay consistent and stable for days or weeks and then be able to scale up and down over that period of time. In the case of containers, we're talking seconds, right, seconds to minutes to hours. Containers come and go very, very rapidly. Traditional storage systems are not designed to change their environment, to change what's connected to it, what's not connected to it in that shorter period of time, in seconds to minutes to hours. And so traditional storage systems typically don't have the capabilities of doing that. And so that's why you need high-performance systems, such as we have with FlashBlade or FlashArray, but also why the software-defined storage of Portworx was so special. So the container environment puts very different loads and expectations on storage systems and why not every storage system that was designed for VMs or traditional environments will work in a container environment.

And Portworx will connect into any cloud storage out there?

It will. It's able to be utilized on top of any cloud and any cloud storage. That's correct.

Great. Okay. Just want to move on from a product standpoint moving into FlashBlade. And it seems like that product has just had a really good, I don't know, 12 to 18 months. Maybe talk about the drivers there. Is it AIML, is that the big thing that's created such a strong backdrop for demand for that product?

AI and ML is perhaps the most unique demand for it or the demand that creates the most unique opportunity for FlashBlade because, really, it's almost alone in terms of its high-performance levels of pure data throughput that it can provide to those applications. But it's also been utilized in traditional analytics environment because of the ability to run many different analytics workloads on massive data sets that operate on FlashBlade. So we call it a data hub. It can sit in front of what are called data lakes to allow -- data lakes typically don't allow high-performance users of data. They can store a lot of it, but you don't get a lot of bandwidth out of it. So the FlashBlade can sit in front of that and provide a lot of performance to the applications that need to use the data there. And then finally, I'll mention that a very big driver to date has been the rapid restore capabilities that helps to battle ransomware. That's been a big use case for both FlashBlade and FlashArray//C, depending on the other requirements for the data that the customer might deploy.

One of the questions I think that comes up when you talk about big data and AIML is, there's so much hype and chatter around technologies like Snowflake and some of the AWS, Redshift and things like that. How come companies just don't do this in the cloud?