Pure Storage Inc (PSTG) 2019 Q4 法說會逐字稿

Good afternoon.

My name is Mike, and I will be your conference operator today.

At this time, I would like to welcome everyone to the Pure Storage Fourth Quarter Fiscal 2019 Earnings Conference Call.

(Operator Instructions) I will now turn the call over to Matt Danziger, Head of Investor Relations.

You may begin your conference.

Thank you, and good afternoon.

Welcome to the Pure Storage Fourth Quarter Fiscal 2019 Earnings Conference Call.

Joining me today are our CEO, Charlie Giancarlo; our CFO, Tim Riitters; our President, David Hatfield; and our VP of Products, Matt Kixmoeller.

Before we begin, I would like to remind you that during this call, management will make forward-looking statements, which are subject to various risks and uncertainties.

These include statements regarding competitive, industry and technology trends; our strategy, positioning, and opportunity; our current and future products; business and operations, including our operating model; growth prospects and revenue and margin guidance for future periods.

Any forward-looking statements that we make are based on assumptions as of today, and we undertake no obligation to update them.

Our actual results may differ materially from the results predicted, and reported results should not be considered as an indication of future performance.

A discussion of risks and uncertainties relating to our business is contained in our filings with the SEC, and we refer you to these public filings.

During this call, we will discuss non-GAAP measures in talking about the company's performance, and reconciliations to the most directly comparable GAAP measures are provided in our earnings press release and slides.

This call is being broadcast live on the Pure Storage Investor Relations website and is being recorded for playback purposes.

An archive of the webcast will be available on the IR website for at least 45 days and is the property of Pure Storage.

With that, I'll turn the call over to our CEO, Charlie Giancarlo.

Thank you, Matt, and good afternoon, everyone.

Thank you for joining us on today's earnings call.

I will begin the call with a summary of our fiscal year results, a look at Q4, and share our excitement for the future; Hat will provide a go-to market update; and finally, Tim will give a detailed review of our financials and updated outlook.

Looking back at FY '19, I believe we made excellent progress on our key company initiatives outlined this time last year.

We invested in key markets and growth areas: the Global 2000, the Cloud 1000 and next-generation analytics and AI.

Our international revenue grew from 26% of sales in FY '18 to 28% of sales in FY '19.

We achieved profitability for the first full year.

We continued our industry-leading delivery of important new products opening up new TAM.

We now provide hybrid cloud offerings, solutions for rapid restore and over 80% of our system capacity now ships as NVMe.

Lastly, with our leading software and increasingly subscription-oriented business model, we continue to deliver industry-leading customer delight, as evidenced by our NPS score of 86.6.

Now our results.

Full year results were strong with revenue of $1.36 billion, growing 33% year-over-year.

Gross margin was 67.6% and operating margin was 3.7%.

Gross margin remained at the high end of our long-term model, and we delivered our first full year of profitability, which we will continue to grow in the years ahead.

Q4 was a good quarter.

Revenue was $422 million, growing 24% year-over-year.

Gross margin remained strong at 67.6% in Q4 and operating margin was 7.4%.

While we were below our guided range on revenue and profits, these gaps were the result of 2 distinct unanticipated items: First, there was a process breakdown at one of our contract manufacturers.

This resulted in a number of orders that were expected to ship in the quarter not shipping until the days following our quarter end.

We have now addressed these process deficiencies.

Second, as you know, we've launched our enterprise storage service, or ES2, offering last year, focused on enabling our customers to have a cloud experience on-premises with a true subscription-based model.

We're pleased to say that in Q4, we exceeded our expectations with ES2.

Increasing subscription services drives better economics and value for Pure in the long term but, of course, creates less revenue in the quarter in which it is booked.

Except for these 2 items, we would have been within our guided range.

As we look ahead to FY '20 and beyond, we're excited with Pure's market-leading momentum.

We will continue to invest in the strategy we outlined last year: driving growth in our key markets; growing the contribution of international revenue; driving scale and efficiency; leading our market in important innovations; and finally, continuing to focus on delighting our customers, partners, employees and long-term investors.

As one proof point for our optimism, I'm pleased to announce that just this week, we signed a deal valued at over $100 million spread out over approximately 2 years.

Our new product and service offerings have positioned Pure strategically in front of 4 major growth trends reshaping our industry and growing in importance for CIOs.

These are hybrid cloud; fast, consolidated data architectures; AI and analytics; and the transition from slow data backup to rapid restore.

Importantly, most of our new offerings in these areas are based primarily on a subscription revenue model.

Let me go into a little more detail on these growth trends.

First, hybrid cloud.

Customers want a hybrid cloud solution that enables them to move mission-critical applications to and from the cloud with ease and agility.

Customers also want to have the same data services and APIs available in the cloud that are common today on-prem.

Our Cloud Data Services running Pure software natively on AWS are designed to enable our customers to do just that.

The response from beta and prospective customers has been very positive.

Second, fast, consolidated compute and data architectures.

As cloud and enterprise customers deploy modern, data-centric applications, they are demanding faster, consolidated data architectures.

They want hyperscale efficiency and performance.

Last week, we announced our DirectFlash Fabric, now available on FlashArray, which promises to revolutionize the way modern computing is deployed.

Once again, Pure is leading the industry, first charting the path to all-flash and now transforming the compute stack, enabling direct access performance with shared storage.

This innovation brings hyperscale efficiency and performance to everyone.

Third, AI and analytics.

We continue to pioneer data solutions for AI across multiple use cases, thus broadening the accessibility of AI solutions in the market.

With fast, consolidated data as the key enabler of AI and analytics, we are well positioned as the first mover in this $6 billion market.

Fourth, data protection or, as we have now redefined it, rapid restore.

As announced last week, ObjectEngine, formerly StorReduce, is the biggest architectural change to data protection in more than a decade, delivering on our vision that customers want immediate restoration of services rather than just back up.

ObjectEngine is built with a flash to flash to cloud approach, replacing disk and tape.

Customers can now recover data in minutes rather than hours or days by using ObjectEngine with FlashBlade.

And ObjectEngine uses the cloud as a cost-effective archive destination for data, eliminating tape and enabling additional uses of their data in the cloud.

Considering all of our new offerings introduced in the past year, we believe we have grown our addressable market from $35 billion to more than $50 billion.

As we embark on our 10th year, we're focused on continuing industry-leading growth, expansion of our solutions, and advancing what makes Pure great: unrelenting innovation and a laser focus on customer delight.

With that, I'll turn it over to Hat.

Thanks, Charlie.

Momentum was strong for the fiscal year with 33% year-over-year annual revenue growth at scale, nearly 1,500 new customers, a productive and expanding partner ecosystem and excellent traction across our innovative and differentiated product portfolio.

Let me dive into 3 areas that contributed to our progress in the year and the quarter.

First, we finished the year with 5,850 total customers, continuing our momentum with more than 400 additions in the quarter.

We also made progress in the largest enterprises, governments and cloud-native organizations.

We now have 40% of the Fortune 500, we added multimillion-dollar government sector wins in both the U.S. and EMEA, and our cloud customers continue to contribute more than 30% of our total business.

We're also seeing strong demand within our existing customer base, as evidenced by a record number of million-dollar expansion wins this quarter.

Second, our partner ecosystem continues to expand and deliver real value to our go-to market engine.

During the quarter, we had excellent traction across our national channel partners, technology alliances and global systems integrators.

Our largest national partners are increasingly choosing to lead with Pure.

Most recently, Pure received a number of accolades, including Data Center Partner of the Year from SHI, CDW UK Silver Partner of the Year, CRN's Data Center Vendor of the Year, and Best Vendor Channel Program from SVC Awards.

Additionally, we've made significant progress with a number of strategic alliances.

Our partnership with Splunk to advance big data analytics continues to thrive, and our joint customer base more than doubled in the year.

We signed Rackspace as one of the largest managed service providers and see continued advancement with NVIDIA and Cisco, including the expansion of our converged offerings to include FlashStack for AI and FlashStack as a service.

Finally, GSIs are growing rapidly and becoming an increasing contributor to our go-to market efforts; and Accenture, DXC and HCL are all now working with Pure.

Our leading technology, together with our Evergreen business model, are a perfect fit for GSIs, enabling them to reduce costs while improving their service levels to customers.

As Charlie mentioned, the progress in and our commitment to this segment is underscored by our recent 9-figure, multiyear agreement with one of the largest GSIs.

Third, our relentless focus to drive value to our customers has changed the nature of our conversations we're having with them.

Pure's new innovations in the areas of hybrid cloud, next-generation analytics and AI and rapid restore, together with our unique subscription model, has enabled us to become more relevant in their digital transformation journey.

As Charlie noted, we launched our Cloud Data Services to enable customers to make the most of their data, whether they choose it -- to house it on-prem, in the cloud, or increasingly a mix of both.

Enterprises that benefited from Pure's differentiated approach of automation, simplicity and the Evergreen business model can now seamlessly extend that experience to the cloud.

In fact, our 100% software subscription-based Cloud Block Store offering, which we launched for beta access in Q4, is oversubscribed and early customer feedback has been overwhelmingly positive.

FlashBlade delivered another great quarter as we've achieved consistent repeatability and grew our use cases around next-generation analytics, AI and rapid restore.

Looking forward, we're very excited about our recently announced ObjectEngine solution, a revolutionary subscription-based approach to data protection that we believe will enable Pure to capture a large portion of this roughly $8 billion market.

This cloud-first solution leverages FlashBlade to deliver 10x faster recovery for our customers' most critical data, while the deduplication and native cloud integration of ObjectEngine protects the entirety of our customers' data in the public cloud, all for less money than they're spending on legacy backup appliances.

Lastly, as customers are increasingly looking to consume data solutions as a service, our teams are seeing success in selling our core portfolio as a subscription with our Evergreen Storage Service, or ES2.

We were pleased with our momentum in this area, which included several multimillion-dollar deals globally.

We believe that the continued momentum of ES2, together with our new subscription-based products, uniquely positions Pure to deliver the industry's leading data solutions on-prem and in the cloud in the way that customers want.

In summary, the fundamentals of our business remain strong, and with our customer, partner and employee momentum, we continue to take share in the overall storage market.

Our portfolio has never been stronger and now includes new innovations across major growth trends, including hybrid cloud, next-generation analytics AI and modern data protection.

We had a great year, we're enthusiastic about 2019, and it truly feels like we're only just getting started.

With that, I'll now turn the call over to Tim.

Tim?

Thanks, Hat.

Fiscal 2019 was a strong year for Pure as we continued our march towards $2 billion, growing 33% year-on-year and delivering our first full year of profitability.

Before I dive into the specifics, I'll make my usual note that the gross margin, operating margin, OpEx, net income and free cash flow numbers I will use are non-GAAP unless otherwise noted.

Reconciliations of these non-GAAP metrics to their GAAP comparables as well as our full Q4 results and presentation are available on our website at investor.purestorage.com.

For the quarter, total revenue grew 24% year-on-year to $422 million.

Product revenue grew 19.7% year-on-year to $340 million, and support revenue grew 47% year-on-year to $82 million.

As Charlie mentioned, we had 2 unanticipated items that adversely impacted our Q4 results.

The majority of the impact was due to an unexpected distribution issue with one of our contract manufacturers, which we have addressed.

In combination with this issue, we launched our ES2 subscription offering last year to empower our customers to have more of a cloud-like experience on-prem, and we overperformed this quarter relative to our original expectations as we shift more business to this as-a-service consumption model.

While this resulted in lower revenue recognized this quarter due to the subscription nature of ES2, the long-term economics of our subscription-based model are compelling and we look forward to continuing to grow this business in the new fiscal year.

Except for these 2 items, we would have been within our guided range on both revenue and profits.

Geographically, 71% of our sales came from the United States and 29% came from our international markets for the quarter.

We grew the overall proportion of our revenue that comes from international markets as we continue to scale and build our brand awareness outside of the United States.

Our gross margins continue to speak to the value and differentiation that our customers see in our solutions.

Margins continued to be strong in Q4.

Total gross margin for the quarter was 67.6%, down 0.5 point sequentially but remaining at the high end of our long-term model of between 63% and 68%.

Product gross margins were 67.8%, down modestly 0.3 point quarter-on-quarter due to a mix shift to our lower margin FlashBlade business.

Both FlashArray and FlashBlade maintained strong margin profiles in the quarter.

Support gross margins were 66.8%, down 1.3 points quarter-on-quarter.

Q4's decrease was driven primarily by investments we are making in our support and professional services team to serve our largest enterprise customers as well as lower onetime support revenue items as compared to prior periods.

Moving to operating margin, we continue to make great progress towards our long-term profitability goals, achieving the key milestone of our first full year of profitability.

In Q4, operating profit was positive $31.1 million or positive 7.4% of revenue and compares to an operating profit of $24.9 million or positive 7.3% in the year-ago quarter.

Q4's profit results were directly impacted by the shipping and subscription items we discussed earlier.

Excluding these items, our profits results would have been within our guided range of between 8% and 12%.

We continue to invest given the positive momentum we are seeing in the business and the opportunity we have in front of us.

Q4 net income for the quarter was positive $37.0 million or positive $0.14 per share.

This compares to the year-ago period of $28.8 million or positive $0.11 per share.

The weighted average shares used for the per share calculations were 264 million in Q4 and 251 million for the year-ago period.

Moving on to the balance sheet and cash flow.

We finished Q4 with cash and investments of $1.2 billion, an increase of $53 million from the previous quarter.

Free cash flow in Q4 was positive $51.4 million, which includes $17 million of impact related to our employee stock purchase plan.

The solid performance at the end of the year enabled Pure to generate positive free cash flow of $64.3 million for the full year.

As we wrap up our review of fiscal 2019, I'd like to highlight something that we haven't talked a lot about to date: the increasing momentum we are seeing in our deferred revenue balance.

Our customer-first strategy has led to the highest customer satisfaction in the industry.

Happy customers stay with Pure and ultimately buy more from Pure, and we are seeing that in our accelerating deferred revenue balance.

We believe a growing deferred revenue balance is an indicator of the success we are having in executing on our strategy of building long-term relationships with our customers.

As a result, deferred revenue grew 43% year-on-year to a record $536 million, an acceleration from 41% growth in Q3 and was a direct result of 3 key trends: number one, significant year-on-year increases in our customer renewals; number two, longer-term initial subscription commitments from customers; and number three, the shift we are seeing to our subscription-based products like ES2.

These are important and positive trends for our business.

Now I will turn to our guidance.

As we look into next year, we are excited about the opportunity ahead for a number of reasons.

First, new products have allowed us to address new markets, including hybrid cloud, data protection and web-scale storage.

Second, we have notably increased the amount of sales capacity we have at the company.

And third, we are delighted about the new partners and routes to market we have developed over the last year.

A proof point of this is the recently signed $100 million-plus deal that both Charlie and Hat talked about.

These trends give us confidence that our strategy is working as we continue to invest.

For the full year of fiscal 2020, we expect revenues in the range of between $1.735 billion and $1.805 billion, representing 30% year-over-year growth at the midpoint; non-GAAP gross margin in the range of between 65% and 68%; and non-GAAP operating margin in the range of between 3% and 7%.

For the first fiscal quarter of fiscal 2020, we expect revenues in the range of between $327 million to $339 million or 30% year-over-year growth at the midpoint; non-GAAP gross margin in the range of between 65% and 68%; and non-GAAP operating margins in the range of between negative 8.5% to negative 4.5%.

Our Q1 guidance is based on the consistent and deliberate strategy we have successfully executed on in each of the last several years, namely to invest early in the year, most notably in our go-to market organization, which positions us well to capture significant revenue growth in our seasonally strong second half.

This strategy has resulted in year-over-year operating leverage in the past, and based on our full year guide, we anticipate year-on-year improvement in our operating margin again.

One final technical modeling note.

As our guidance for Q1 calls for negative net income, we would move back to a basic share count number for EPS calculation purposes.

We anticipate a basic Q1 EPS share count of between 240 million and 250 million shares.

In FY '19, we continued to overachieve along our path to $2 billion and beyond, driving industry-leading gross margin, taking share, driving strong operating leverage and developing long-term, meaningful relationships with our customers.

With that, I will now turn the call back over to Charlie.

Thanks, Tim.

Looking ahead, I'm excited about the tremendous opportunities in front of us.

We have a $50 billion market TAM in key markets that are essential to powering the data of tomorrow's digital businesses: hybrid cloud, AI and modern analytics; scale-out architectures; and rapid recovery of mission-critical applications.

We continue to take market share with our world-class customer service, industry-leading innovation and differentiated business model.

We're winning in the cloud with both a significant cloud customer base and our introduction of new Cloud Data Services.

We're expanding our subscription-centric business model, driving retention and expansion into our growing portfolio while delivering industry-leading customer delight.

And lastly, we're tracking to the financial and business goals we've outlined in the past.

Since our founding almost 10 years ago, Pure has transformed the antiquated storage industry and helped customers derive business value from investing in their data.

As we look forward to our next 10 years, we're positioned to deliver the industry's next set of innovations.

I will now open the call for questions.

Operator?

(Operator Instructions) Your first question comes from Aaron Rakers from Wells Fargo.

I guess just first from a housekeeping perspective.

I was wondering if you could help kind of frame the effect between the 2 items that impacted the revenue results this quarter between the partner impact versus the ES2 impact.

And how are you guys kind of modeling out the effect of increasing ES2 as we move forward?

And I have a follow-up.

So Aaron, this is Tim.

We aren't disclosing the specific numbers, but the majority of the impact was from the contract manufacturer issue, and then obviously, the remainder was from the ES2 topic.

As it relates to guidance going forward, we're thrilled about what we're seeing in ES2 and naturally have developed new forecasts based on what we're seeing.

And those are factored into the guidance that we offered here in our prepared remarks.

Perfect.

And then I guess the second quick question would be, talk a little bit about how you'd characterize the current competitive landscape and particularly as we see the industry move kind of more towards NVMe and even NVMe-over-Fabrics?

Have you seen any signs of your ability to kind of separate yourself further from the competitors?

Or just how would you characterize that?

Aaron, this is Hat.

No discernible difference in the competitive landscape.

It's a competitive market, as we know, but our win rates are holding nicely against all of our competitors.

So I think that's the key indication.

You look at what some of our competitors are guiding.

They're guiding 0% growth, we're guiding 30%.

So I think where there are some success where there's single-digit growth in other areas, it's not coming at our expense.

Relative to the moat and our differentiation, I don't think our competitive positioning has ever been stronger from a product perspective.

You look at what we're doing with NVMe-over-Fabric, you look at the other innovations that we launched around Cloud Block Store, around our ObjectEngine, our hybrid cloud story, that's one area that we're getting a ton of traction with.

I just feel like the combination of all of that in the portfolio has never been stronger.

Kix, anything you want to add?

Yes, just regarding the NVMe, I would add on that, look, I think our software-driven strategy for end-to-end NVMe is truly working.

Two things to point to: First off, over 80% of the shipment is leaving the factory with NVMe.

I challenge any competitors anywhere near that.

And then the second thing I'd say, if you look at what we announced with DirectFlash Fabric and our approach of NVMe over RoCE Ethernet, we took a harder path here to go down the path of Ethernet, but we fundamentally believe that that's where the real advantage will lie.

And it's not just about making applications faster, which of course it does, but it's about really redefining the architecture and bringing the hyperscale architecture on-prem.

If you look at the hyperscalers out there, they deliver their premium storage services with shared storage and with NVMe, and we're bringing that same architecture to the masses.

Your next question comes from Katy Huberty from Morgan Stanley.

Now that all the storage companies have reported as of tonight, what you see is a pretty meaningful slowdown across the vendors.

So just curious, your thoughts as to whether there's anything macro that may hit your business.

And it sounds like that's not the case.

Why not press your advantage and actually double down on spend over the next year in order to take meaningful share?

And then I have a follow-up.

Katy, thank you for that.

That's exactly what we feel we're doing.

I mean, if you look at our guide, both in terms of our optimism about the future on the 30% growth that we're expecting, as well as only a modest increase on the bottom line, it's because we've hired a record number of [QBHs] over the last quarter to start Q1 with, and we're continuing to hire at an aggressive rate.

We are still taking market share.

Our opportunity is not limited by macro.

Not that we're seeing any real change in the macro, but that's not our market opportunity.

We can take share in any market.

And maybe, Hat, you spend a couple of minutes on this as well.

Yes, no -- I'm thrilled with the capacity adds that we had in Q4, which is seasonally a difficult quarter to hire, and we did a great job there, continuing to add velocity and capacity on the hiring front into Q1.

So I think that's a great indication of us doubling down.

We look at the opportunity that we have with these new products.

We've doubled the number of products that we put in our sellers' bags in the last year.

And so as those new products get traction and the combination of all those products individually make the portfolio motion really differentiated, we're absolutely investing for the future.

And then just as a follow-up, can you talk about the traction of the AI-ready storage products?

Are you seeing that get pulled into new use cases, new industry verticals?

This is Kix, I'll take this one.

Yes, it was a great quarter for AI solutions on a couple of fronts.

So we continue to see strength with NVIDIA and our solutions there with AIRI.

And we also introduced a new solution with Cisco, FlashStack for AI, and this continues to underscore our strategy of really trying to make AI mainstream and bring it to a mainstream environment that's paying off.

Your next question comes from Pinjalim Bora from JPMorgan.

This is Pinjalim.

The -- Charlie, about the DirectFlash Fabric, I guess congrats on launching that.

I know it's pretty early, but what has been the general feedback from customers?

And more importantly, as the innovation in the space I guess progresses from RDMA to support fiber channel to TCP, do you think you would see storage architectures basically transform more into the data hub architecture that you kind of laid out in the Analyst Day last year?

Yes, thank you for that question.

So we're very excited about the DirectFlash Fabric and, in particular, from the response we've gotten really from our most advanced and largest users.

They are looking at this as a game changer in terms of their ability to deploy very efficient, very effective, highly concentrated and faster operating compute environments in their data centers.

So the most advanced users are doing this and by advanced, I mean the technology leaders.

We feel that it's only a matter of time before it then goes into every customer.

We're convinced that beyond the rack, this is the kind of architecture that next-generation data centers -- the way they're going to reconstructed.

Kix?

The only thing I would add is that we're really seeing interest on both types of architectures.

If you look at traditional, classic applications, this helps deliver hundreds of microseconds of lower latency, which makes databases go faster.

That's a win for everyone.

On the flip side of the coin, it's really interesting for web-scale applications.

If you look at most of them, they're built around DAS architectures today, where individual flash is inside every server.

This allows us to bring the efficiency of shared storage to those applications and opens up a whole new TAM for us to go after.

Understood.

One follow-up about the DR strength that we are seeing.

Are we -- is it fair to assume that the guidance basically assumes a higher ratable portion going forward?

And given the DR strength, does it make sense to start looking at some metrics like billings going forward?

Yes, a couple of things.

Certainly, the acceleration that you're seeing in the deferred revenue balance is indication that the percentage of ratable revenue is increasing.

It's a very nice trend.

It's a trend that's been happening now for a couple of quarters, and we thought it was important to call it out.

As we go forward and watch the continued sort of growing of the ES2 business over the next couple of quarters, you definitely will hear us talk some more about additional measures just to give the investing community more insight to how that part of the business is transforming.

So stay tuned as that business continues to grow.

Your next question comes from Wamsi Mohan from Bank of America.

Sorry, switching between a few calls here, so apologies if I missed this.

But can you talk a bit about your view of market growth into this calendar year?

Obviously, you're guiding 30% growth and you're projecting that both for the quarter, for the year.

And a lot of your competitors have seen deceleration and yours seems like it might be more endogenous than exogenous.

So can you just maybe share your market view?

And I have a follow-up.

Wamsi, yes, we've been asking this question of ourselves for quite some time.

We really see the market as being pretty steady, honestly, no dramatic moves up or down.

It's probably best said, it's no clear indication.

If anything, we get that it's going up slightly.

But what gets us excited about our growth opportunity is, as Hat mentioned, we've doubled the number of products in our sellers' bags.

We've got a record number of new sellers and we're hiring very aggressively in that area.

And we have a very large number of existing customers, new customers and partners, indicating to us on a daily basis that they're going to be doubling their -- they're going to be doubling down on Pure this coming year.

So we're getting a lot of very, very good buying signs.

I'd have to say that from an overall market, I think it's steady as she goes.

There are signs in both directions, but for the most part, we see it as slightly up.

Okay, great, appreciate that.

And then just a quick clarification on the shortfall here in the quarter.

Was it 100% from the 2 elements that you cited?

Or was there anything else?

And does the contract manufacturer issue just cause a delay of shipment into this current quarter?

And what was the magnitude of that?

I'll take the first part of that question.

So between those 2 elements, we would have been within our guided range.

And Tim, do you want to take the other part?

Yes, no -- and I guess -- and yes, there was a delay from a contract manufacturer perspective that naturally will pick itself up and be shipped out here in the first part of the quarter.

And what's the magnitude of that?

This is not something we're sort of talking about from a specific dollar amount, but it was the majority of the delta between our results and our guidance.

Your next question comes from Jason Ader from William Blair.

Guys, I just want to look at the FY '20 guidance of 30%, and I guess what really drives your confidence that you can see that type of growth in a market that is changing fast, you have some obviously big competitors that are pretty hungry right now.

I guess, it just doesn't seem conservative.

So maybe help us understand the level of conservatism there.

Yes.

Well, we're using the same methodology that we use every quarter to guide.

And as you know, we've beaten 12 out of 13 quarters.

The things that are giving us confidence are the new products, a record number of [QBHs] and we're starting the year really flush with QBHs and are continuing to hire aggressively.

And again, the new partners that we have, we have new global system integrators, new MSP partners all buying at -- and indicating that they're going to be buying at very, very significant levels.

So we have the same degree of confidence in this guidance as we like to have in every quarter.

And I think the other thing I would add, Jason, is if you look at the markets, the key markets that we're focused in from a storage perspective, whether it's AFA or and some of the things that are now happening in the cloud, those are the growth markets of storage as well.

So we're placing our strategic bets both in innovation and where we're putting our salespeople in those parts of the market that are growing.

And obviously, that bodes well for us, too.

Jason, one last add, I certainly concur with those things.

The $100 million deal over the next couple of years doesn't hurt either.

Got you, okay.

And then one last -- one follow-up just on the new products and especially the cloud services that you've launched.

When do you think that will be meaningful to revenue?

We think that -- now, of course, they're all new products and so I don't want to indicate that they're going to be a very large part of our revenue.

But we're expecting significant revenue from the ObjectEngine product this year.

We're expecting the beginning of revenue traction with the Cloud Data Services set of products.

Your next question comes from Steven Fox from Cross Research.

First, quick clarification.

So with the surprise on the subscription revenues, how did that change your fiscal '20 outlook relative to 90 days ago for subscription revenues?

I know you're not breaking out specifically, but maybe directionally, you can talk about the change.

And then I had a follow-up.

Yes, Steve, in -- within our own sort of internal numbers and then obviously, that flows over to the guide, we've updated -- upped our forecast for ES2 quite simply.

There's no question about it.

We see good momentum in that business and we expect more coming from ES2 now than we did 90 days ago for the full year.

And I'll add the following, that up until Q4, we had some governors on ES2.

Think of it as wanting to make sure we had the product and the service properly managed and set up.

And then beginning in Q4, we kind of took the governors off and it took off.

And as of this year, it's an equal part of our product portfolio, and the sales team is going to sell whatever the customer wants to buy.

So the leash is off.

That's helpful.

And then just quickly on the Cloud Data Services.

When you say early-stage revenues this year, is there any sort of thresholds we should be looking for, either with specific cloud service providers or otherwise?

Well, a couple of things.

One is we're in beta now.

we're in -- or beta 1 now.

We expect to go into beta 2 sometime late next month or early following month, and we expect to go GA sometime, let's say, late summer -- our Q3, sometime in our Q3.

And then revenues, of course, are ratable.

And so from a Pure revenue perspective this year, it's not going to be a tremendously large amount.

But of course, we're expecting a big pickup in the product.

So hopefully, on the -- hopefully, we'll see a good pickup on the booking side from a contract value.

Steven, this is Hat.

The other thing I'd add to that is that does provide a nice halo for our core products.

Customers want to partner with somebody over the long term that are relevant in a multi-cloud or a hybrid cloud scenario.

The fact that we have native software now embedded and delivering the Pure experience, which people have come to love.

That 86.6 Net Promoter Score is legitimate.

So now they can actually have that same experience in the public cloud and manage across it.

And so for us to be able to provide a direction and a strategy and capabilities to help them do that helps us sell products today.

Your next question comes from Rod Hall from Goldman Sachs.

This is Bala Reddy on for Rod.

So FlashBlade saw a strong quarter in your Q3, and you said the mix has increased this quarter.

So I just wanted to check on how the product has performed this quarter.

Any color would be helpful, and I've got a follow-up.

Yes, Bala.

The -- we had a great FlashBlade quarter, the continued momentum in that business.

I think as we, at the beginning of the year, started to see momentum around our rapid restore use case, our next-generation analytics and AI use case and then penetrating into EDA, into the technical computing world, those use cases have now crystallized as being very repeatable and scalable.

And so as we get more and more of our sellers comfortable with them, our participation rates go up.

And they know that it's a differentiator for them in their accounts.

So continued momentum that we see in that business and we've got the repeatable use cases.

When you add the ObjectEngine, kind of hybrid cloud solution for kind of redefining data protection, we're even that much more enthusiastic because it provides a really specific use case to extend the amount of data that we can store for our customers.

The rapid restore FlashBlade use case takes the most critical data and improves that from a 10x perspective on-prem, but with the ObjectEngine capability natively implemented into AWS, you can now take the rest of the data and store that or archive that in the public cloud.

So it just expands our TAM and our conversation.

Understood.

So just wondering if the anticipation of a further drop in NAND prices is causing any kind of spending pause in the market.

Do you see any of this in the market?

Or just because lots of your competitors have missed in the quarter, just trying to understand what caused this sudden weakness.

Yes, there's a very simple answer to that.

No, we're not seeing any change whatsoever from an overall pricing standpoint.

Part of the benefit that we have, and this is really clearly shown in our gross margin, right, is that what customers value in us is the software differentiation that we have in our product, 15 points of gross margin higher than the rest of the market.

So that indicates that it's really about more than just cheap NAND.

It really is about the services that we're able to offer, the density that we're able to provide, the ability for them to -- for customers to be able to get access to all of their data at fantastic rates.

We -- but we do see future decreases in NAND as helping us to start to penetrate the second tier, and we're -- that is also -- we think that's going to happen this year, and that's also part of what makes us very optimistic about the year ahead.

Your next question comes from Alex Kurtz from KeyBanc.

Can you hear me okay?

Oh yes, Alex.

Charlie, just a clarification and a question.

Does NAND price changes really have any effect on how customers think about signing POs at the end of a deal?

I mean, just for multiple years of looking at this, Hat and the whole team, I know there's volatility in NAND, but do customers really care that much about it when it comes time to like sign the PO?

Alex, this is Hat.

No.

I mean [realistically], no.

Okay.

We can move on to my question.

The SaaS vertical and the contribution to the growth assumptions for this year -- and I apologize, I just jumped on, if you already answered this, but it seems to be growing very fast.

What are the kind of assumptions in the annual number on the SaaS contribution versus the rest of the business?

Thanks.

So it continues to be north of 30% of our overall bookings, and we added 400 customers and delivered the business at 33% year-over-year growth.

And so SaaS continues to be a really key driver of that.

We're excited about the expanded solutions that we can provide in that segment, obviously playing in the block and the FlashBlade for next-generation analytics.

But we're also now kind of going after their DAS environments and our capabilities there.

When you get built into the pods of these fast-growing SaaS companies, you're built in to the architecture and the fact that we're delivering better service levels, a dramatically higher density and lower cost, enable them to deliver a better service online.

So SaaS will continue to be a key part of it.

And Alex, this is Kix.

One other thing to add on there.

I'd remind you that when we talk about our cloud business, it's the combination of both SaaS and MSP as well as IAS.

And SaaS is a great business for us, but we've seen real strength in IAS and MSP as well.

We just announced a joint offering with Rackspace as an example, where we're bringing an SSD-powered, ultra-high performance block service to their cloud and their customers.

And so we continue to see that business grow just as nicely.

Your next question comes from Karl Ackerman from Cowen and Company.

So going back to your guide, if I were to poke holes in the quarter, I mean, it would seem that the $20 million of revenue slipped from Q4 into Q1, but that would suggest overall revenues will have decelerated to, let's call it 22% year-over-year growth, certainly below your full year outlook.

Now it seems your ES2 service offering is doing quite well and adding ballast to your outlook.

But having said that, I guess at what point should we begin to consider you and your peers lowering prices in order to drive demand for the associated higher-margin, add-on software and services of all-flash arrays?

And I guess what else gives you perhaps greater confidence in accelerating product revenue growth from here?

And I have a follow-up, please.

Yes, no, thank you for that.

I'll start just talking about the general reduction over time with NAND and what it does for our guide and so forth.

We saw -- I mean, there was an overall reduction over this past year, as there is almost every year, in overall pricing to the tune of 10% to 15% or so.

And -- but despite that, we had the growth that we have.

We expect a roughly -- that's about the annual reduction rate in this market, and our guide addresses that.

Understood.

I guess going back to that then, Tim or Charlie.

Looking at your outlook for gross margins, why should we expect your margins to decelerate in fiscal 2020 when NAND ASPs are set to decline roughly 20% this quarter and the next?

I would imagine that's a material tailwind for your margins in calendar '19.

Yes.

Karl, this is Tim.

We've always talked about the 63% to 68% range is really the sweet spot that we like to be in.

And so we as see NAND prices change, reminding you what Charlie said, it's all about the software.

But it opens up additional sort of opportunities for us.

And so we'll put those margin points to work.

And it's the playbook we've run now for quarter after quarter after quarter, and we like where we're at.

Your next question comes from Erik Suppiger from JMP Securities.

This is Michael Berg on for Erik Suppiger.

I have another one just on the same issue around ES2 and the guide.

It sounds like it is coming off -- the manufacturing revenues coming off in Q1.

Before, your guide seems to be fine.

Is there another issue in Q1?

Because if you -- if we assume the revenue does happen in Q4, the Q1 guide is a little light.

Is there some -- another factor to think about?

Or how can we think about the guide?

Yes, this is Tim again.

I think there are some puts and takes there.

I think the -- first of all, is we're very proud of 30% in Q1 just like we're proud of 30% for the full year at scale.

So make no bones about that.

But I think as we saw the overperformance in ES2, we wanted to factor more of that in the business.

And I think you're seeing part of that memorialized in the guide as well.

ES2, long term, great for us but as you know, ratable business has a little bit lower -- or slower sort of revenue build than immediate recognition.

Okay.

And then a quick follow-up to that.

I mean, would it make sense to start viewing billings as a reasonable metric?

I mean, now or at some point in '19 or calendar '19?

Billings is one of a series of metrics that we're looking at to take a look at as this business builds and reaches a sort of point of critical mass.

And certainly, you'll hear us talk more about that over the course of the year.

Your next question comes from George Iwanyc from Oppenheimer.

In light of the strong sales expansion you had last year, can you give us a sense of your expectations for adding new hires this year and how quickly last year's people are scaling and reaching full productivity?

Yes.

So 2 things there.

Q4, as I think we mentioned earlier in the call, was a very strong quarter for us.

We'll have another strong quarter in Q1.

The beginning of the year has always been an investment period of time for us.

It's been that way since I've been here, and it's no different here in this year.

So I again look to see that, and you saw that in the guide.

In terms of productivity ramping, we're seeing productivity ramp very nicely.

It's one of the key measures that we look at to give us confidence to keep making those investments, and we like what we see there right now.

Yes, I just would add that we look at this as a capacity and an advanced challenge.

And so by putting more feet on the street, provided they maintain that productivity, which we're seeing happen nicely, it's all about getting into more fights.

And I said that before, but we really like the progress that we've made on our strategy throughout the year, getting into the right fights.

And so Rackspace was a very intentional investment that we've made over the last year.

So -- in the MSP space, it's one of the largest MSPs out there, and you heard us talk about DXC and Accenture and HCL.

These are the kinds of organizations that have hundreds of petabytes of data under their own management and also on behalf of their other customers from a joint go-to market perspective.

So it takes a while to invest in those, but we see a really nice lift in our ability to compete in more at bats with them.

All right.

And Dave, that $100 million contract with the system integrator.

Over the 2 years, is this something that you expect to just gain momentum from quarter to quarter?

Is that a fairly linear contribution?

Yes.

So we haven't broken that out in terms of how it will contribute.

We think it underscores a great commitment both by the GSI and by us on really getting into a fight together and working together over multiple years.

I do see upside with these relationships across the board, but like I said, they take time to build.

And so I think we've got a good, moderate view of where we are today.

It's built into our forecast, but I do see some upside here in the years to come.

Your next question comes from Eric Martinuzzi from Lake Street.

I also wanted to focus on the GSI win that you guys had.

There is probably not a more sophisticated buyer out there because they're betting their own business on your business.

To what extent can you enlighten us on the competitive landscape?

I'm sure they're -- they have install base customers with your competitors.

And then why do you feel you were the winners?

Yes, great question and you're right.

They're very smart.

They obviously work on behalf of some of the most complex customer environments in the world, and they have some of the smartest architects out there.

So -- and they do have existing relationships with other competitors.

They don't move into these relationships lightly.

I'll say that our core value proposition of dramatically lowering cost while enabling next-generation applications and use cases and better service levels is a sweet spot for GSIs.

If you think about the amount of petabytes per FTE our customers can manage on our equipment versus the legacy equipment, it can be 5 or 10x.

And so that drives directly to their bottom line, but much more importantly, it becomes an enablement platform for all the digital transformation work and all of the multi-cloud architecture work and all of the next-generation analytics use cases that they want to serve their customers with.

And so as I think about it, it's not only an enablement platform, but it's also a funding source for them to add more value -- value-add services off the stack and change that 80-20 mix of run 80% of IT budget and to the change budget to 20%, change that mix because CIOs don't want to give those dollars back, and systems indicators have the capability of delivering solutions, driving digital transformation and change.

And so we're that funding source and enablement platform.

And do you feel like you're on equal footing out of the gate?

And was it kind of they came to you and it was yours to lose with this particular GSI?

No, it's a big investment.

We need to be able to work with them over years and gain their confidence.

And so there's a lot of vetting from a technical perspective, a lot of vetting from a business perspective.

But look, I think we're at a size and scale now where we're a very safe bet.

Our innovation road map is completely differentiated, and we don't have to hold on to the legacy business practices and the revenue of what the other guys do and that -- the GSI is burdened with that.

So we can kind of remove the shackles a bit and actually drive them to deliver better services, but shift their revenue mix in the higher value, differentiated offerings that they want to spend their time on as well.

So it took a while to get in, and we're still kind of plugging away, but very confident that once we get it going into their go-to-market engines and our own, there's a great amount of velocity that can occur downstream.

Your next question comes from Andrew Nowinski from Piper Jaffray.

Just another follow-up question on the GSI contract.

Is that a minimum commitment to purchase over the next 2 years?

Or is that kind of a -- more of a [sentence] specific deal?

I guess what I'm trying to get at is whether that's just a starting point or could there be upside to that $100 million over the next few years?

It's an extension of already existing relationships.

It is a binding, minimum commitment over the 2-year period, but there's upside on top of it.

And I think the other important point that Hat made there is that this customer that just made this relationship with us has already been a customer and have been a happy customer.

Okay, very good.

And then with regard to the ObjectEngine, I think you mentioned that there'll be significant revenue this year from that.

So just to clarify, is that an organically developed solution by Pure?

Will there be some revenue sharing with any other technology partners that might be part of it?

This is Kix.

I'll take that one.

That's the result of our acquisition with StorReduce married with all the work we've done around FlashBlade already with rapid recovery.

And so part of why we're so excited about this market is that we've already seen a great start to our rapid recovery business with FlashBlade on its own.

And then adding the ObjectEngine solution in just helps us address a wider range of workloads.

I'll just add on to that, that again, we're offering ObjectEngine 2 ways.

One is with a capital purchase and the other -- and a subscription service, and the other way is all subscription.

So again, a lot of that revenue is going to be ratable.

Your next question comes from David Ryzhik from Susquehanna Financial.

Just going back to StorReduce.

To what extent can hyperscale players be customers of FlashBlade with StorReduce IP?

I think you've talked about a 5x to 10x cost improvement using StorReduce on FlashBlade.

It just sounds like that's something they'd be interested in specifically given all the investments they're making around AI and machine learning.

And I have a follow-up.

One of the great things about ObjectEngine is that it was built natively in the cloud to begin with.

And so as we acquired StorReduce, they already had a great working relationship with Amazon.

And so we see purely deployment in the cloud as an absolute use case that we're going to invest in, and that's for cloud-native customers that want to leverage it there.

And there's use cases for customers on-prem who want to back up on-prem to flash and then use the cloud as their long-term repository for backup data.

So absolutely multiple ways to take advantage of it and we expect a lot of volume in terms of using it in the cloud itself.

Got it.

And I just wanted to step back and talk about the future of all-flash arrays.

To what extent do you see storage-class memory playing a role in all-flash arrays?

You have differentiated IP around optimizing flash.

Can you do the same optimizing XPoint?

And can we see tiering of XPoint as a cash would cheapen the flash as a storage?

Well, I just wanted to get your views a few years out on how this evolves because there's certainly -- we keep hearing it more and more.

I think this is an area where if you look at our history of optimizing our software for solid-state memory, we're just unmatched in the market.

And so if you look at our competitors who just take kind of package solutions, SSDs and plunk them in their existing platforms, we go far beyond interface or software directly with solid-state mediums.

And we've now had kind of multiple generations of adopting our software to flash over the years.

And we think that just puts us in a great position to look at the next generation of these technologies, and there's both storage-class memory coming out at the high end of the market, and then at the low end of the market, there's QLC.

And we're excited about taking advantage of both of them in our products as it makes sense.

Yes.

And we expect to be doing that soon.

So just to put a wrapper on that one, we're really expecting that the new technologies that we've been coming out, our unifying cloud and on-prem, we haven't really talked about that, but Cloud Data Services, what's fantastic about that is it's exactly the same in the cloud as it is on-prem.

And what that means is it makes it that much easier for our customers to be able to place their applications where they want them, whether it's on-prem or in the cloud, and to be able to choose between the 2. We're also consolidating data.

Customers today are putting data on -- not only are they copying it multiple times for different applications, but every time they copy it, they're putting in a different array that is tuned to that application.

We're creating an environment where they can access the same data with multiple applications in the same array without having to repurchase the same storage 2 or 3x.

That data now is hot.

We believe in no [cold] data, meaning that data be available to any application at the performance that data -- that application requires.

And increasingly, as you can see with our new introductions of Cloud Data Services, of ObjectEngine and even ES2, we're increasing the amount of software and subscription-based services that we have in our overall environment.

So I just wanted to put a wrapper around that.

Your next question comes from Simon Leopold from Raymond James.

And this will be the last question, Simon.

Great.

Hopefully, you can hear me okay.

I'm on a cell phone overseas.

Nonetheless, I wanted to touch on how the competitive environment may be changing this year.

So clearly, mathematically, you're doing quite well and you had a big time-to-market advantage with NVMe, but the expectation is competitors will be launching products using this new protocol later this year.

How do you think it plays out when you have competition with the implementation and you're not alone in the market?

So I would just say the competitive environment -- this is a competitive environment.

We've known that, but we've done very well in it and our -- the technology that we have is so different, and you marry that with the business model of Evergreen in an increasing subscription suite of products, and our customer satisfaction, the moat that we created, we had it with our core products and NVMe is at the root of that.

We have 80% of our core FlashArray products rough and tough already on NVMe.

I mean, ask any other competitor what percentage of their state is there.

So we think we've got a lead.

We think that DirectFlash Fabric extends that.

And then when you combine that with these new subscription services and new markets that we're pursuing, I feel like our competitive advantage actually just continues to grow.

Yes.

And I'd say if you just kind of top level up to the overall market, we moved into the #6 position overall in the overall enterprise storage market, and yet all-flash is still just 1/3 of the market.

And so there's an enormous amount of the rest of the market for us to convert to flash and to take over.

And that's -- the NVMe position of our value profit, that's also just everything that makes Pure great and delivers the 86.6 Net Promotor Score.

I will now turn the call back over to Charlie Giancarlo for closing remarks.

Thank you.

Pure had a great FY '19 and we're very excited about the future, as you can hear.

We're proud of our progress and the opportunities ahead this year and actually in the years to come.

I want to thank the entire Pure team and our global partners for their tireless effort and dedication this past year.

I also want to thank our customers for their business and confidence in Pure and the solutions that we provide.

And thank you all on this call for your time today.

This concludes today's conference call.

You may now disconnect.