Zscaler, Inc. (ZS) Earnings Call Transcript & Summary

All right. I think we're all back. Good morning. Thank you for joining us on day 2 of the Goldman Sachs Technology and Communacopia Conference. I'm Gabriela Borges. I head up the emerging software vertical here at GS. Delighted to have on stage with me the executive team from Zscaler. We've got Jay Chaudhry, CEO and Co-Founder; and Remo Canessa, CFO. Thank you so much.

Thank you, Gabriela.

I wanted to start with a conversation with an architectural question because I know since the time of the IPO, you've been consistent in saying that Zscaler win deals because of its architecture. And so the question that I would ask is, help explain to us why Zscaler's architecture is the right architecture for the security problems of 2022? As well as when it was originally found in 2007, what makes the architecture durable for today?

Yes. You can always add features and functions on top an architecture. Think of architecture like the foundation of a building. When you set a foundation, your foundation is meant for certain types of buildings. And you can add rooms and features and windows and doors on top of it. Those are like features. In technology, architecture doesn't change very often. 20, 30, 40 years, sometimes an architecture endorse. The traditional network architecture was invented in late '80s, early '90s are all network based on IP addresses and all are 30-plus year old architecture. And still, if I give you an example, in the auto business, all the cars we have today, most of them are the still internal combustion engine. That is the core architecture of the car, what kind of engine is. Now we are finally, after a long time, an electric engine. The 2 cars may look similar from outside. They all have seats and -- all have radios and everything, but if you open the hood, they're very different. Similarly, security for the world of cloud and mobility has required a different architecture. When we said in the office, our data and application set in the data center, you had a network security architecture by building a moat like firewalls and VPN. You're in, you're safe. You're out, you're not safe. The new world where data is everywhere, applications are in the cloud somewhere, users are somewhere, you cannot be building moats around your stuff. You build a switchboard. We are like a switchboard and exchange. You come to us, we say, who are you? Where are you going? We connect to the right party, very different architecture. You can't build it on top of firewalls and VPN. You start from clean slate. Yes, legacy firewall companies are claiming they got Zero Trust, they got everything Zscaler has. It is like a traditional engine car company bolting on an electric engine on it and say, I am an electric car. I got better thing than Tesla does. That's the fundamental difference. So since -- even though you could say, yes, I started this company 14-some years ago. It was designed for the world of cloud and mobility. I got inspiration from Salesforce. I saw how Salesforce was overtaking this big company called Siebel Systems. That was a single-tenant architecture. Salesforce has multi-tenant architecture. That's one big difference. In security, we are a proxy architecture that inspects, stops, checks things. Firewall is a pass-through architecture, which can't do the kind of inspection. I hope it helps.

You mentioned clean slate, and it raises the barriers to entry question, which is, when you think about security companies that today building their architectures, perhaps with a clean slate based on public cloud, how do you think about the barrier to entry for some of the new entrants that are coming into the market?

Yes. There are certain products you can build easily than others that are hard to build. Take, for example, identity. It sits on the side as a database. It takes effort to build it. But we are like sitting in line, the traffic goes through us. If this thing doesn't work, you had no access to anything out there. So being able to sit in line, inspect everything for security and other things without slowing things down, is a monumental task. It takes years and years to figure this, to cook it to make it work. I can tell you, CIOs are realizing it. I was talking to a CIO last night at a dinner. I was figuring out the importance of cloud and application. I said, the Office 365 must be your most critical application. If it goes down, I mean you can't really do anything. He said, no, no, no, Zscaler is my most critical application because you are the switchboard connecting everything to everything. So being highly reliable, resilient that takes tons and tons of experience. You could have a good product, but if you can't run the cloud well, it's no good. I have lots of scars over the past 14 years, making sure that cloud works well, and that's a big competitive advantage for us.

I wanted to also ask the forward-looking question, which is, I know that Zscaler employs a consultative approach with its customers and that you're very closely aligned in understanding your customers' longer-term road maps. What are your customers telling you, and what are your engineers telling you is the #1 technical challenge that needs to be solved in security architectures looking forward?

Yes. Honestly, the bigger challenge for customers to deploy this is inertia and mindset change then technology. Our technology is proven. It works. There are thousands and thousands of companies doing 250 billion transactions a day. If you have been doing something for 30-plus year the same old way, sometimes when our approach goes, CIO loves it, CISO loves it, Head of Infrastructure loves it. But when you go down, people start getting a little uncomfortable with the change. They like what they know. So often -- and this is not a challenge for us. This was the same challenge when CIOs wanted to move out of data center to public cloud. Data center operations teams felt very insecure and uncomfortable, but I think they all realize that things need to move on. That's why getting a buy-in of the CIO, CISO, CTO is important. Once they buy-in, they drive transformation. Technology actually works very well because we have done it thousands of times with thousands of customers.

It leads into a conversation along the lines of how do you think about the catalyst path to essentially create a trigger point for customers to change the way that they think about architectures. And originally, there was a conversation around Office 365 and the move to the cloud. During COVID, there's been a very large conversation around distributed workforce. What do you think is the catalyst path for customers today? And how do you think about the risk that the best part of the catalyst path is behind us because we've already had Office 365 transformation, we've already had work-from-home transmission?

Yes. Very, very good question. So overall, transformation is a phased journey. CIOs are driving it. Now a lot of this focus on still customer-facing applications, all this stuff is happening. But there are 2 pieces that are the biggest catalysts today. In the past 2 years, fiber has become the #1 priority for CIO as well. When SolarWinds happen, and then Colonial Pipeline happens, and then Microsoft itself had issues itself. So it is #1 priority. 5, 6 years ago, CISOs would complain to me and say, we don't get enough attention. Security is so important. Why isn't my management focused on it. And now they're saying, man, I'm asked to present at every quarter, at every Board meeting. So the budget is not an issue, it's the #1 priority. Right behind that, especially in today's uncertain market, cost management is a big priority for them. There are fewer resources, so the CIOs are asking, say, how can you help me? I still need to carry on with my transformation, but I'm prioritizing transformation projects. But if you can help me eliminate a bunch of my point products, so you reduce cost of those boxes, but also reduce my operational cost because it takes people and cost to run those systems. And you improve my security and user experience. So those 2 things are the biggest drivers for us.

It leads a little bit into a question around the macro and what you're seeing over the medium term. When you think about big digital transformation projects, what are your customers telling you about their willingness to invest even as their visibility into their business may be less than it historically has been? How do you gauge the risk of perhaps some of these big digital transformation initiatives being put on hold or pushed out by 6 months to a year?

So we're only hearing from CIOs that there's no such thing as it's a big transformation project, move on. They are prioritizing transformation projects that they want to do them now. And some, they're going to put on the back burner and pick them up later. So that's happening. Now we go in as a small piece for the big transformation project actually. If you really think about the CIOs transformation project, security is not a very large piece of it overall. When we go in and say, if you look at transformation projects, a lot of those transformation projects are about having applications available to do things because it's applications that drive business agility, being able to have better access, timely access to make decisions alike. And we are enablers to make sure your users can access those applications securely with a great experience without having to worry about the network, which slows things down. Internet is the network in our world. So when they become enabler and we reduce cost, it all comes to as beautifully. If we couldn't really do simplification, then there will be some slowness. But we have done a good job showing -- and to some degree, it's not too hard because companies have got so much this appliance stuff over the years. One CISO describes it, I have appliance overload. One described it having appliance fatigue. So a lot of that stuff has to be kicked out. And what's happening is, it's interesting. When you -- if you sell to the techie level, they don't want to kick out anything. We call them appliance huggers. And CIO, CISO goes in, they are looking at the budget factor. In fact, they actually have to make a conscious effort to say, remove all this stuff, simplify the stuff. So that's where our strategic approach dealing at the C level is becoming very helpful.

On the question on cost reduction or the topic of cost reduction. Give us some examples, order of magnitude on where you're able to realize savings. And it leads actually into a conversation around consolidation, which is, there has been a narrative in security on wanting to consolidate point products for many years now. You already alluded to this. It tends to be a lot of inertia where you have companies that have in common positions, legacy positions. So let's bring it all together. What are some of the ways in which you consolidate specific functionality and order of magnitude? Share with us some of the ROI payback periods, et cetera.

Yes. So when we have a dialogue about Zscaler project, the customer often -- now says more often than before, what is my return on investment on this project? What will we get? So for that, we end up doing a business value assessment, where a customer actually is more open to share with us that this is the money I'm spending in this area. Tell me, if you take these things out, what will you save? So ROI ends up being a way of quantifying the investment they make. The savings come from 4 key areas. Number one, there's a lot of point products that are bought. So what are the typical point products we replace. With Zscaler Internet Access, it's a full-blown out on DMZ. A typical large enterprise would have probably 8 to 10 different kind of products and probably, there are 20, 30 boxes sitting in a single data center. They could start from here is a Blue Coat proxy. Here is my firewall. Here is my DLP offering. Here is my antivirus. Here is my sandbox. Here my bandwidth control. So all of that stuff goes away with ZIA. ZPA often think -- people think of it as VPN replacement. Not really, VPN is only a piece of it. We replace the entire inbound DMC, the whole thing from load balancing to DDoS protection to external firewalls, IPs, all this stuff. So elimination of point products is #1 part of the cost savings. Two, operational cost to run and operate those all those devices cost a lot of money. That part goes away. The third piece they end up helping is risk reduction. That gets a little tricky to quantify, but there are many parts of risk reduction. What was the cost of Colonial Pipe being down for a day or 2? What's the cost of your plant going down for 2 hours? That becomes one way to look at it. The second ends up being if my IP is lost and it goes to some competitor stuff, how much loss does that bring to the table? Those -- and that's the fourth end up being called business productivity. In the old world, when you go through a traditional hub and spoke, user experience is very poor. And then the cost of figuring out, the things are slow. It's very hard. With ZIA, ZPA, your response time is sometimes 70%, 80% faster, so better productivity. And also, if things are slow, we have the service called Zscaler Digital Experience. It pinpoints where issues are so you can fix it quickly. So all of this stuff helps you do better business productivity. Four key things for savings or -- and which really results in great ROI.

Jay, you commented on the willingness to invest, continuing to be very strong from your customer base. Remo, I want to ask you the question as well, given your earnings call a couple of days ago. And so one of your comments on the earnings call was on the linearity of the seasonality of billings growth in your next fiscal year and first half being stronger than the second half. So tell us a little bit about what your expectations are for calendar '23, which would be your second half fiscal year? And is there a risk that customers have already budgeted for calendar '22? You have those forecasts. Is there a risk that those budgets ultimately don't get spent for whatever macro reason or customers, specifically?

Yes, that's a great question. So the -- what we saw in Q4, we had an outstanding Q4. It was 57%, 58% billings growth on a year-over-year basis. But what we did see in Q4 was higher scrutiny related to deals. So when we took that into account related to our guidance for fiscal '23, in addition, we took into account is the global macro. What we talked about on the call is for the '23 budgets haven't been set yet for customers. So our billings guidance, what we said is that the billings percent for the first half will be between 42% and 43%. Whereas in the prior year, it was 41.5%. So basically, at the midpoint, 42.5% versus 41.5%. That leads into the uncertainty related to the global macro into the second half. Regarding budgets of companies at year-end, our pipeline is at a record. We didn't see any meaningful change in our close rates. We are in these accounts. So we feel great about our guidance, very good about our guidance. Our guidance is billings growth rate of 30%, 31%. But based on our pipeline, based on the visibility, we feel good.

So there's no -- we're not seeing any weakness in pipeline or stuff. The second half is just factoring in to say, it's farther out. The budget is not there so let's be cautious about it. But the other point I'll mention is, our customers are very happy, okay? Happy customers buy more. Happy customers tell others to buy. It's so fascinating when you have these dialogues. I had a meeting with the CIO of a Fortune 500 company. This is a fascinating conversation. The CISO, in this case, likes appliances and all that stuff hasn't moved to the new world. CIO likes it, and she came to me. She said, I want to meet you, even though security is involved in some of the stuff. I talked to my fellow CIOs, and they all talk about Zscaler. And my security team is still stuck on appliances. I want to personally understand from you, why? I mean how and what should we do? What do you deal with other customers? How do you [indiscernible] Now we had this conversation. In this case, CISO was separately reporting, and she said after a discussion, Zscaler Digital Experience, I want to start with that because the user experience is #1 priority, it has nothing to do with security. Let's start a project. And that POC got started. So it's fascinating to see how CIOs talking about the CIOs, getting great feedback for us are actually pulling us in and talking to us.

Remo, I'll ask a follow-up question on visibility, which is, how much visibility does your existing customer base give you about their longer-term road maps? And then it leads nicely into a question about the cross-sell opportunity that you have ahead of you. To what extent are customers already telling you, hey, 2023, I'm going to expand my Zscaler appointment in X, Y, Z ways?

Yes. I mean our visibility is very good. And the reason for it, if you take a look at our ARR, or annual recurring revenue, we've got a good base related to going into fiscal '23. In addition, our engagement with our customers is outstanding. So for our visibility going into fiscal '23 related to our customers and our ARR, it's great visibility. What was the second question, I'm sorry?

It leads into a question on cross-sell.

Cross-sell, yes.

For the customers that are taking ZIA perhaps the plants to expand into ZPA. There's now a longer-term road map as well with ZDX and cloud protection. So a little more on, I guess, 2 parts. One is, how are you incentivizing and enabling your sales force on the newer levers of cross-sell? And for the customers that are engaging with you across the spectrum, how much visibility do they give you into that plant to expand with cloud protection, for example?

Yes. A lot of questions. I'll try to answer. So from a compensation basis, pretty much the same for new and upsell. From a compensation basis, whether it's ZDX or Workload or ZIA or ZPA, pretty much the same. Regarding our visibility that we have with our customers, again, it's outstanding. Our upsell was approximately 60% in fiscal '22. So again, significant upsell. The upsell ability that we have is, we have a 6x opportunity in our ZIA and ZPA base. That does not include ZDX or ZCP. The advantage of Zscaler, if you take a look at what we can sell, we have 4 pillars. When we went public, we had 3 pillars. Primarily one, just ZIA. ZIA basically was the lion's share, over 90% of our business. And we talked about ZPA contributing to that or growing. What we've seen basically, ZPA has expanded to 27% of our new and upsell. This year also, in fiscal '22, our emerging products were 14%. And what we talked about was that we expected our emerging products to be in the low teens, which it came in at that. If you take out that 14%, basically, ZPA was approximately 1/3 of ZIA and ZPA. So from the time we went public and you take a look at basically a ZIA-dominant company, which is protection from a user to the application in the open Internet, to ZPA, which was basically protection from a user internal, we're able to expand it and really deliver. The emerging products is a big portion of our business -- will be a big portion of our business going forward. Again, what you got to realize or recognize with Zscaler is that we're a technology company, and we will continue to innovate. Now those 4 pillars are expanding, and we're using the architecture, ZIA and ZPA, for users to basically applications. We're using the architecture of ZIA, ZPA for Workload to Workload and using ZIA and ZPA for IoT and OT. The breadth of our platform that we have and to go into customers with this type of architecture and this type of security, as we talked about, decreases complexity, lowers costs and increase security, significant increase from an ROI perspective, significant increase from a security perspective with the proxy built architecture. So what we have is -- and the complexity. If you take a look at the number of security professionals in the world, it's not growing to the level required for companies to really protect themselves, which you're gaining basically with Zscaler. You're getting the cloud effect of security, and that's lagged behind. If you take a look at what's happened, you've looked at applications. I mean that ship has sailed. I mean there's a lot of debate on applications going to the cloud and putting a trust in the cloud. Banks are one of the last groups that kind of jumped on board, but now with Office 365, that ship has sailed. The applications are in the cloud. Companies can spin up faster. They can spin down. They can lower cost and lower capital investment. Same thing with security, but I think even more so. Security is complex. Related to just our ZIA and when you're having to use go-to application or Workload to Workload, we have like 23 different type -- 23 applications we're consolidating. We're making it simpler, Zero Trust. You put the traffic to Zscaler and then we take it from there. What does that do? It reduces your capital cost. It decreases your complexity. How do you basically -- if you're a company, how do you protect yourself? Of all these applications related to buy additional product and staying up to that next revision of code, understanding where everything is. You can't do it. I mean, well, it's very, very hard to do. By having a cloud security platform, Zero Trust, what Zscaler's created is that we take care of it for you. You put the traffic to us, and then basically, we secure your company in the most efficient manner. And when you think about applications going to the cloud and users to be mobile, the question comes down to why do you want to bring everything back on to your network? You don't. And that's what a lot of companies are doing, and when you do that, what are you doing? You're basically expanding your network. You're expanding it to the cloud. You're increasing your attack surface. Zscaler does not increase your attack surface. It's Zero Trust. So the answer to your question related to when you have that type of platform and the strength of that platform and you go-to customers, and you talk to those customers who are our Zscaler customers, it is easier to upsell. So as we go forward through the normal maturity of a company or companies, because of our 6,700 customers that we have, 30% of the Global 2000, which is up from 25% of Global 2000 last year, which is a significant number, we have the ability to cross-sell ZDX, Workload Protection, additional applications we have in ZIA and ZPA. It's a powerful, powerful story for customers. And again, existing customers, who are customers of Zscaler appreciate, understand and trust. So the key thing is getting that trust with customers, which we're getting. And when -- 5 years ago when we went public, it was more of an evangelical sale. Zero Trust is here. Everybody understands what Zero Trust is. And when I look at where Zscaler is positioned, I don't see any other company positioned as well as we are.

As Zero Trust has become more of a paradigm shift, how is the competitive landscape changing in response? And particularly, the conversations that you have with C-level executives that are your customers, are there other platform companies that perhaps are being more aggressive in wanting to be a part of that conversation as well?

Yes. So 3, 4 years ago, we used to evangelize how Zscaler needs to be done. Zero Trust essentially started off saying, you need -- you don't need to put people on the network, you connect the right party to the right party using a proper identity, and we have a switchboard. So now when legacy firewall and network security companies are feeling they're getting disrupted, they all embracing it. So everyone is Zero Trust now. It's creating a lot of confusion out there. It's kind of -- I understand why they're doing it because they need to preserve their business, and -- but it is doing a lot of disservice because it's making some of these companies think that they're doing Zero Trust when they aren't doing Zero Trust. So part of the thing we do actually is now we educate. They all believe that cloud-based security like Zscaler needs to be done and how are we different. And that question goes back to the basic architecture question you asked early on. So we do end up differentiating showing the stuff, but a big part of our success is, actually, we have now lots of customers who are happy talking to each other. And when you start, that reference goes a long way along with the good technology, and that's why we're engaging and winning. That's why the competition on the large enterprises hasn't changed at all. We go in, we win. And legacy large enterprises easily get it. They see the difference and say, I love your firewalls, they're wonderful, but this is not your type of stuff.

What about outside of the large enterprise? Has there been any change in competitive narrative or the way you're educating the channel outside of large enterprise?

Yes. So traditionally, our focus has been from very large down to about 6,000 user-based companies. In the past, about 4 quarters, 5 quarters, we have moved down from 6,000 down to about 2,000. We call it the enterprise level. And there, we do see more players. Now we are just building our presence there. In the past, we didn't really do much there, and we are seeing competition there. And so firewalls, vendors, typical vendors and all do show up. But once we go shop and really engage, almost always, we win. But that's one of the market areas. We're doing well, growing well. We are adding a lot more resources. That channel is playing a bigger role. 3, 4 years ago when I used to talk and used to say, traditional channel was still not ready to sell our kind of stuff. Now they are. We're getting more inbound calls, and channel is playing a bigger role on the lower end of the market, and it's helping us. Our channel come and leads registrations and all and pipeline is going up, and we're very pleased with that.

Remo, how does the unit economics compare when you evaluate your newer cohorts in upside of large enterprise versus your traditional presence in the large enterprise?

Yes. I mean the cohort for the channels, you'd expect them to be better. I mean you -- typically, customers of 100,000 employees are going to get a certain price. When you go down into smaller customers, just a higher price. Also the cost economics, the CAC that's going to be better. Again, it's the leverage of the channel. And as Jay mentioned, channel typically sells boxes. That's the traditional for a channel. And right now, VARs basically are getting inquiries from their customers to sell basically cloud solutions. So it is becoming more mainstream through our channel and select channel. We have a targeted channel that basically we'd expect the channel bars to increase going forward.

I'll ask a follow-up on implications for the longer-term model. And some of the metrics you've provided, rule of 80 on revenue growth plus free cash flow margin, how do you think about the durability of that combination of revenue growth and free cash flow margin? And outside of the channel mix and some of the other positive mix shift benefits in your model, what are some of the other margin levers that investors should be aware of?

Yes. I mean from a rule of 80, I mean it's outstanding. I mean the benefit that Zscaler has, you take a look at our model related to ratable with high gross margin. Our free cash flow margin this last year was, I believe, 21%. And the prior year was 21%. So significant free cash flow leverage that we have. And our operating profitability was in the 10%, 11% type range. So on a go-forward basis, we just have a strong leverage model. Again, when you're having high growth -- high gross margins, really, for me, and as I've talked about in the past is that really what we're trying to do is build up market share. We want to be the dominant player in this space. As long as you've got basically the type of unit economics that we have, whereas in years 2 and 3, the contribution margin is over 60%. And if we can continue to grow with this type of operating profitability, we increased our guidance by 150 basis points year-over-year. And I talked about on the call that I expect free cash flow to be over 20%. We're in a very unique position. And what I talked about also the model that we have, never seen it before. When you've got a -- you have a model like we do and I talked about it when we went through the public offering, you've got so many levers to pull. I mean, if things slow down, we can push out brand profitability. If things are going strong, keep on investing. The key thing I talked about also is that we're an R&D company also. We're not going to stop innovating. You take a look at R&D as a percentage of revenue, it's 15%. [indiscernible] pretty low. Well, factor in a couple of things. I mean we grew over 60% last year. So there's a lot of significant investment we're doing. That's one factor. The other factor is, we leverage India. Majority of our engineers are in India in 4 different locations. And the strength of the engineers in India, I would say, is pretty close to 1:1 with the U.S. engineers. So we have significant leverage in our model related to how the company is distributed. You take a look at our headcount, our headcount is approximately 5,000 employees that we had. Probably 600 or 700 of those are employees are in Santa Clara or San Jose. The rest are dispersed throughout the world. Related to where we're going to get the leverage in the model, it's going to come from sales and marketing. Gross margins have been in that 80% range where we want to stay in that 78% to 82% range. We guided to approximately 80% for short term, midterm. Could we increase margins? We can. But why? The reason is, basically, we take advantage of public cloud, which gives us the ability to innovate and develop at a faster pace than trying to optimize our margins. As long as we're making these gross margins, get the product out to our customers, help our customers. So the overall model, like I said, I would expect R&D in the 15% range. You can see G&A kind of slow down a little bit, still low in the 7% range. So you're going to get the -- basically the leverage in sales and marketing. Long-term model, 20%, 22%. We didn't say when. We didn't say when for a particular reason. We want the flexibility to invest. We want the flexibility to really try to do what's right for our customers and do what's right for investors.

I'll ask one follow-up on sales and marketing, and then I do want to open it up to the audience. Specifically on leverage in the sales and marketing line, remind us what you've said about pace of hiring, mix of sales capacity between ramped and ramping sales reps for fiscal '23?

Pace of hiring is going to remain strong. The mix between ramp versus not ramped, basically 50-50. Sales productivity we're expecting to be flattish to down the reasons because of the pace of the hiring. In addition, we're going to go a little bit downstream more. A little more emphasis this year downstream. So when you go downstream, the reps that you're going to pick up aren't going to have the quotas of the reps of the large enterprise or the majors. So that will have a contributing factor also. But pace of hiring, I don't want to say the exact pace of hiring. All I want to say is, we're in the growth mode, and we're going to continue to invest, and that's the reason for the sales productivity being flattish to down.

Questions from the audience? Please?

So just on, again, many IT companies leverage labor in India. Talk about sort of you're now thinking are 1:1. At what point you guys see that cost base in [indiscernible] declining over time as the standard of living in India increases a bit more in 2030? That's obviously going to impact a lot of technology companies here that leverage that. And I think it's the thing that is not talked about.

From my perspective, there is wage inflation in India, clearly, but there's wage inflation throughout the world also. We take a look at primarily in India in 4 locations, and you take a look at each of the locations. Each has their own basically cost. For example, Bangalore is the highest cost of the regions we have. We're in other regions where basically the cost is not as high. Where is it going to go? I can't answer that. I mean, is it going to be 2030, 2035? I mean maybe, Jay, you've got a better answer.

Also it's true in the long term. Your question is valid, but there are so many other factors playing into it. Automation is happening more and more. Auto -- software code is being written more and more on its own. So I think so many factors will play into it. So we are working on all of those.

Final question. On competition, specifically in the newer emerging products and cloud products that you're offering, cloud protection products in market. How do you think about competition in those segments of the market specifically? It's very fragmented. And is the sales -- because it's more of a greenfield sale presumably, how do you establish an incumbency position, leveraging what you already have?

Very good questions. So for cloud protection, the market is confused in general. Investors are confused, too, okay? And it's an emerging market. So for simplicity, you should think about cloud workload protection in 2 buckets. Bucket 1, when workloads talk to Internet or workloads talk to each other. Traditionally, companies are trying to do what they did in the data center using firewalls as virtual firewalls and do it that way. That's a lot of it is being done. For the first time, Zscaler has brought in a Zero Trust approach built around ZIA, ZPA, the switchboard approach. So the only competition in that space is legacy firewall companies. And a couple of quarters ago, we announced, our offering made it generally available. We are actually counting on that big time. It's not a greenfield because the same CISO who bought Zscaler for protecting users is actually helping buy Workload Protection when they communicate with each other. So that's one part. Second part is probably that you hear a lot more about from a lot of these new companies, and Gartner throws those acronyms, CSPM, CI, EMC, that kind of stuff. There are lots of newcomers. Over 100 companies started in the past 18 months, especially when money was very easily available. [indiscernible] Okta, the [indiscernible], all those names come up. These vendors are almost like CASB vendors who are there. They are not sitting in line for communication. They basically make API calls to see if a Workload is configured properly or not. Does workload going to have proper permissions or too many people are allowed to access those. Do these workloads have vulnerability. So this is all done by scanning and API calls. It's a relatively easy entry. There's no big barrier to entry. Tons of competitors in that space. I think you'll -- eventually both sides will come together. We did 2 acquisitions in the latter space work to bring them together. We are integrating that along with workload communication so we become fully differentiated part. And I think that's a big growth opportunity for us. Early indications are very positive, and we're bullish on it. The last comment we'll make is, many times, companies struggle from going from one point -- one product company to a multiproduct company. When we talked about ZPA at the time of IPO, that was the first time they're going to prove to ourselves, to the world that we can go from ZIA to ZPA. We did that successfully. Now we've done it very successfully with ZDX, very rapidly growing faster than ZPA did. And now the cloud workloads and all is the next time. We're doing it the fourth time. So we think we have learned how to do it well. We learned how to expand the product portfolio, and we've done a good job in selling it. I'm totally focused on execution. I understand that execution at different sizes of the company needs to be done differently. So we are proactively doing a number of things to do it well.

It's a great place to end that. Thank you so much for your time, Jay, Remo. We really appreciate it.

Thank you, Gabriela.

Thank you.