Cloudflare, Inc. (NET) Credit Suisse 26th Annual Technology Conference Transcript
Cloudflare, Inc. (NYSE:NET) Credit Suisse 26th Annual Technology Conference November 30, 2022 3:00 PM ET
Company Participants
Thomas Seifert – Chief Financial Officer
Conference Call Participants
Phil Winslow – Credit Suisse
Phil Winslow
All right. All right. Good afternoon, everyone. Welcome to the Final Day of the Credit Suisse 26th Annual Technology Conference. My name is Phil Winslow. I’m the software analyst. I recognize many of you from prior sessions, so thank you joining us as always. Very excited to have what has been one of our favorite stories for many, many years, Cloudflare, and a longtime friend, Thomas Seifert, the CFO. Thomas, thanks for coming down, appreciated.
Thomas Seifert
Thanks for having us.
Phil Winslow
And so – so I was actually – when I was prepping for this, I started to think I was like, you know what, you’ve just celebrated your five-year anniversary at Cloudflare just this June. And I guess just to level set things for everybody, could you sort of just reflect back since the time that you assume this position. Can you walk us maybe some of the key issues, initiatives that you and management team that have addressed? And how is that evolving to where you are right now in the go forward strategy?
Thomas Seifert
Well, first of all, five years, that’s the longest I’ve been in for a while, and it has been quite a ride. Every time we get ready for an earnings call and we pick what are the customer samples we want to show and how do we talk about new products, and we said, gosh, we did all this in a quarter. Now you want me to summarize the five years. I – we went public. That was a huge initiative. And it’s not so much the fact that we went public. It’s I think what differentiated it is how much effort we put into the pre-IPO getting ready for being a public company. And there’s always this don’t worry about this one day in your life, worry about what comes after it and what systems do we need and what processes do we need, not only to be a public company, but how do we scale, what is life going to look like when you are running for at this time – at that time $500 million of revenue.
So that was always a big part of us thinking ahead and getting ready to scale system-wise, process-wise, people-wise. And when you grow north of 50% for compounding it for so many years that is the true challenge. I think what is always amazing to me even after five years is the amount of innovation that happens. I still remember when we try to find a way to convey our business model in simple terms and have people like you understand what Cloudflare is all about, and we came up at that time with this bumper of Cisco – Cisco as a service, right, and that was quite a way to get started, but you have added – we’ve added zero trust. We added network services. We added storage products. There’s workers and edge computing. So the sheer amount of innovation that happened over the five years, I think, is astonishing and keeps us up.
The network went from 270 points of presence in many cities. We are in more than one location. Now this journey, I think that from a premium model to a pay-as-you-go model where people give us a credit card, to place now where most of our revenue comes from enterprise customers and large enterprise customers, customers that pay us more than $100,000 per year is for more than 50% of our revenue. And we still had this reputation less than three years ago. Oh, you’re only serving small businesses and developers. So digesting this as a company in terms of go-to-market and messaging and branding, I think, was a big part of what kept us up and will keep us up on – on the journey to the next milestone. How do we move up the enterprise stack from our customer size and structure perspective and still keep what we think makes us competitive and efficient as we scale up. So there was plenty to do and there’s still a still a lot that needs to get done, yes.
Phil Winslow
Yes. So, well, you’re still growing, you’re growing fast and you’re driving the train at scale. So it only gets it’s harder, but that’s a fun part.
Thomas Seifert
It’s harder.
Phil Winslow
It is. The – so we’ve talked about over the sort of the past five years. Now the next question is one that I’m sure you wouldn’t have guessed is the near-term. So the – because we’ve been asking everyone with obviously they’re these cross currents and headwinds, macro, geopolitical. How are customer conversations changing? You talked about this a little bit on the last call, but what is being prioritized, maybe de-prioritized where are people saying, look, I need to invest in Cloudflare now versus maybe taking a little longer.
Thomas Seifert
Yes. Well, for us, this was a picture that started to form really early. I mean I think we – the first time we talked about headwinds that would be building up was as early as the fourth quarter of last year already and people just – we had a little folks at that time gave us a little credit for that. But we see a lot. We see a lot of data and that influences business decisions. So what has certainly become a theme over the last two or three quarters is that new logo acquisitions have become significantly more difficult that the more ACV is under negotiation, the longer it takes people measured twice before they sign. We’ve seen that. And we have not seen many changes to the better over the last two quarters to be very honest. Where we see hope, especially unique to us, is when it comes to security and the threat landscape has not – has become more relentless. The war in the Ukraine does its thing to heighten the threat landscape. So there are certain tailwinds that we are seeing. But overall, from a macro perspective, the themes that showed up in the second quarter continued in the third and have not changed materially in the fourth quarter.
Phil Winslow
Got it. So we went from five years, the last few quarters. Let’s talk about today now, actually sort of the blog post that I woke up to this morning. And so obviously, in the blog post, you talked about this is going to be your first price increase in 12 years, 25% beginning January 2023. But you also announced the option of – to pay annually rather than monthly, but do so at a discounted rate. So the question I’m beginning in the hallway is, hey, could you help us figure out sort of the why behind the change and also just the kind of the math on the sort of the impact so to speak.
Thomas Seifert
Yes. So there are – if you look at our business, there are literally three parts to it. There’s a freemium business, where you have millions of customers that use our products and services for free. Then there is a business that we call pay-as-you-go. This is literally how the company got started, where folks gave us a credit card, and we charged them in the beginning $20 a month and then later $200 a month.
And you bought literally eat-as-much-as-you-can plan, right? Everything that we offered in terms of products and services was included in this plan. And then there is an enterprise business, which is today the majority of the business. So when we talk about this price increase, it refers to the pay-as-you-go business, where people gave us a credit card and give us a credit card, and we charge them monthly.
So this is a business that is less than 15% of our revenue. So, it used to be all of Cloudflare, now it’s significantly less. And we’ve had discussions over the years a lot in terms of when do we increase prices in these segments because what you get today for $20 is very much different from what you got for $20 10, 11, 12 years ago in terms of number of products, quality of products and service levels.
So we have been doing a lot of work over the years, when would be the right time, what is the price sensitivity in this segment and we decided to do it now. Well, there is inflation, but cost is not really the important part. It’s also not so much a revenue topic to be very honest. We’ve been quite vocal about our push to get to free cash flow breakeven.
And the important part in this program of raising prices is not only that prices will be increased, but that you can lock in existing prices if you commit to an annual plan. And I think this is literally the main motivation behind this move is turning another 15% of our revenue, probably not fully, but to a larger extent from a monthly to an annual billing concept with all the advantages that comes upfront, which gives us an opportunity to reinvest more and earlier into more innovation. But that’s pretty much the motivation behind it.
Phil Winslow
Excellent, thank you. All right. So let’s zoom back out here for a moment because Cloudflare at a high level, addresses multiple markets, application services, Zero Trust, network services to develop the platform.
Thomas Seifert
Yes.
Phil Winslow
Can you really touch on the opportunity in each of those markets, specifically, where are you seeing the most growth and traction near term, but what offers the greatest longer-term opportunity?
Thomas Seifert
So, we started application services, Cisco-as-a-Service is how we got started. This is obviously today the main portion of our revenue. The IPO we sized it, this is a TAM of about $35-ish billion that we are going to disrupt. The second wave of product is what you call Zero Trust or is Cloudflare One for us. So that’s e-mail, browser isolation, access, gateway and now a fifth product feature called Threat Intelligence that falls into that category. It’s probably another $30 billion to $35 billion market that we are disrupting. We call it a second wave product. So, it’s gaining momentum this year. It will be a significant part from an ACV new business generation opportunity next year. So, this will drive growth in the short term.
And then network services come next. And then there is storage. And then there is Workers. Workers is a product that allows you to literally write code and deploy code at the edge of our network. That is how it started. Now it has become significantly more. I remember discussions as early as five years ago where people said, you have to push this for revenue and we are not going to monetize it? And Matthew, thank God, was really strong pushing back and said this is not about revenue, this is about adoption, how do you get more developers on the platform. So, even in the next year, this will not be a significant revenue opportunity that will come towards the end of this five-year horizon.
Phil Winslow
Yes.
Thomas Seifert
But today, we think we address with the products we have without workers, a TAM slightly north of $100 billion.
Phil Winslow
Got it. Yes. And so when we started discussion, you talked about the $0.5 billion, now $1 billion. And obviously, you’ve given the target of $5 billion. But just like how Cloudflare’s revenue streams have evolved to your point or call it past five years, when you think about sort of this next target $5 billion, how do these revenue streams evolve from here? Like how does the contribution mix change when you think about sort of backing into that $5 billion?
Thomas Seifert
I think our core business will continue to grow. Zero Trust will be the second wave that builds on that, then we will see Network Services and Storage and then Workers will come at the end. I think what makes this discussion so interesting for us is that we think we can get there with the products we have or in the features that are in the short-term pipeline. Because, first of all, the innovation engine is so strong. And then, of course, the TAM that we address is quite significant, and we have penetrated only to less than 1% at this point.
Phil Winslow
I definitely say that Cloudflare has a very big menu. To choose from there, multiple pass and multiple different options to choose from that are kind of connected, disconnected, but actually you all do provide. And very unique in the sense of what you provide versus other people that maybe are just focusing on a sliver of that. When you think about just multiproduct, call it bundling or go-to-market, how do you expect this to evolve at Cloudflare? How is this sort of changing or impacting your competitive position?
Thomas Seifert
I think it will impact it rather favorably. I mean when we – the first part of the journey was literally all-you-can-eat plans. And we sold you everything you want in one piece or in single pieces. As product complexity increases, you have to find way to make it digestible. So we thought like all successful software companies that went ahead of us and in front of us, we bundle our products in groups application services is one zero trust or what we call Cloudflare One is another really important bundle. And you become this one-stop shopping place where people can bundle point solutions in one platform, through one control plane with all the efficiencies that come from a cost, from an ROI and from a performance perspective and especially on Cloudflare One which is zero trust security bundle. We want to be this place, where customers and enterprises go for anything that concerns enterprise security, similar to you go to Microsoft 365 when you think about enterprise productivity. That’s the kind of opportunity, I think, we have.
And the products are put in place now. There are five now that make up this bundle, you can buy it in one.
Phil Winslow
Yes.
Thomas Seifert
But we go where the customer is. We pick you up wherever you are in your journey.
Phil Winslow
Yes. And so when you think about this, let’s double-click on Zero Trust and call it Cloudflare for a little bit because to your point, you have this bundle. So when you can talk about sort of the key products, obviously, you talked about access, but you’ve remote browser isolation, you have added e-mail.
Thomas Seifert
Yes.
Phil Winslow
So you have this bundle. But you also and if you think about your – you can actually have a more modular, you can just take RBI, you adjust your components of it. So how does that impact both you competitively, the go-to-market and how customers can adopt.
Thomas Seifert
Well, it gives us and the customer, the utmost flexibility, right? If your gateway on-premise licenses up for renegotiation, we can just sell you that and pick up the other features as need moves or as budget dollars moves or as existing contracts expire, we can sell also the whole bundle.
What makes it so interesting from our perspective is that once you are on the network, with one product and one feature, literally every additional service we offer is one mouse click away, right? This is the neat part about – there are no professional services. There are no Cloudflare employees that need to sit you and get on promise.
And you see the benefit of moving to the next product immediately. We have now a feature that we call Fly Fishing, where for enterprise customers, testing out all features that we have is just an opportunity. If you’re an enterprise customer, you have five products. You can test every other feature that is on the platform for a certain period of time.
And if you are a firewall customer and you switch on bot mitigation, you probably see that 40% of the traffic that hits here, your firewalls or your load balances as malicious traffic. So enabling the feature automatically instantly reduces whatever the 30% to 40% of the traffic. It lowers your cost, it increases your security posture. So in terms of flexibility for the customer, it’s huge. And for us in terms of finding a path to expansion is very close at hand.
Phil Winslow
Let’s talk competitively a little bit about the Zero Trust and sort of just SASE market because there seems to be sort of a split right now when we think about sort of, call it, the architecture. So you’ve got people that maintain their own their PoPs, own networks, NetsGo, Cloudflare, et cetera, versus those who are trying to leverage hyperscalers really Palo Alto Networks, Checkpoint. Can you walk us through sort of why running your own network is the optimal architecture versus – and how this sort of impacts the competitive dynamics versus these others that don’t.
Thomas Seifert
Our CTO, would sit here, I would say, the speed of light that is the reason and that is – that’s an early answer, but it’s true in a way. That is the thing we cannot make faster, so the way…
Phil Winslow
There’s no more law it seems like.
Thomas Seifert
There is – at least not in our universe. We need to get as close to the eyeballs as possible. This is what we have done with our reverse proxy.
And our Zero Trust offerings run through the same network, through the same points of presence. So the performance, the latency you have by being this close to the eyeballs, regardless of where you are in the world, not in big cities, where a lot of knowledge workers live, but even in remote places is one of the big advantages of our architecture not sitting on a hyperscaler, owning our own infrastructure.
And everything we do over the 12 years, you talked about what else did we do is getting closer to the eyeballs that want to connect. And today, we are less than 100 milliseconds away from 99% of the things, things that want to connect to the Internet. That’s a clear performance improvement and the advantage that we have in the architecture we offer.
Phil Winslow
Got it. Let’s talk about another advantage. I think it’s a little maybe unique and esoteric, but is often overlooked as Cloudflare versus let’s say other Zero Trust or SASE players that use, let’s say, forward proxy networks to secure users access to applications or a Cloudflare added these four proxy capabilities to your core reverse proxies. So you’re forward and reverse. You’re both ends of the connection theoretically.
Thomas Seifert
Yes, yes.
Phil Winslow
What can that do for you as you sit in front of more and more of the Internet and more and more of the origin service, but also more users? What can you do security-wise connection – the performance of the connection, et cetera?
Thomas Seifert
Well, in many cases now, you don’t leave a PoP anymore, right?
Phil Winslow
Yes.
Thomas Seifert
So the connection to the application and the application and the data you want to address in some cases, if its cash is siding in the PoP that you’re addressing, and that is huge from a performance and from a latency perspective, a huge benefit. It has some other benefits to our business model. Because you have to imagine that the reverse proxy, all the first wave of products is paying for the infrastructure of the network.
So how the network is cost structure, it’s the size of the pipe that is moving data is determining the cost and not the amount of data that goes back and forth. So the size of the pipe is already sized by the reverse proxy. Zero Trust is a products are bringing traffic back, the size of the pipe is there.
So we are adding these products at literally 0 marginal cost to our infrastructure. They are very high margin, and this gives us an advantage in how we partner and how we can compete in this market. So there are significant security and performance advantages for the customer, but they are inherent competitive advantages just in the architecture of the network itself that allows us to be super competitive in the space.
Phil Winslow
Yes. Pay for the higher egress and ingress. Egress is up here. So therefore, it takes a long time to close those two and it subsidizes the other. I think that’s another – I’m glad you brought that up. That was actually my next question. Because I think that’s an aspect of it to the people, it’s not just the performance, not just security be on both ends, but egress and ingress, there’s a pricing dynamic here and a gross margin. And so let’s talk about extending Cloudflare’s network directly into building. So point it to be as close as possible with Cloudflare for offices. You’re talking about moving Cloudflare is basically backbone directly into it where millions of people work. So how does – is this a multiplier when you think about your global background, your SaaS initiatives, edge computing with workers, et cetera?
Thomas Seifert
Well, in the – if you keep the philosophy of how can you get closer and closer to the eyeballs that connect, then pushing your – the edge of your network out as far as possible is hugely important. And today, we are in north of 270 cities, north of 100 countries in many cities in more than one location. This gives us now literally an opportunity to move into additional hundreds and thousands of office buildings.
So you get closer to what connects you, you allow a smoother on-ramp of our customers into our network. But it also allows you to turn this around and get moving forward our services and products closer to the eyeballs that want to connect. So that’s a huge opportunity for us in all fairness. From an investment perspective, we slowed this down a little bit because the post-COVID world needs to figure itself out, which offices will grow and which office locations will move forward. But it’s a really interesting concept for us to take – to move the edge further out in…
Phil Winslow
Exactly. I love the line of your blog post, the most global network, but also the most local. And so those are great line. So we’re going to go on to what is one of my favorite topics. You know this. I mean, I’ve obviously been – so I’d call it the original fan boy of Workers five years ago. And obviously, we remain huge believers in just the rise of the edge is really a third tier of compute and storage. Now you think back to the Cloudflare Connect, Matthew highlighted 450,000 developers are using Workers, with the goal of having million developers by year-end. Can you walk us through just sort of think the growth trajectory? You touched on this a little bit earlier, thinking about adoption first. But how do you think about this platform? What are the key puts and takes?
Thomas Seifert
Well, for Workers was – you talked about the five years and the initiatives work has started actually before it. But it was one of these key innovations that allowed us to be what we are and to keep this speed of innovation, right? Worker is literally this product that allows you to write and deploy code at the edge of our network. And it was originally developed for us, right. Because we saw that in our traditional development processes, we were slowed down, how we – how fast we could deploy, how fast we could test.
And our innovation rate came down. And then in one of those interesting sessions, night sessions at a restaurant, not far from our office, this idea of workers was born and then further developed. And then we opened it up as a product for our customers. So the original application was actually enabling our customers to use our products and use Workers in conjunction to make our products the most customized version of whatever they wanted to buy.
So you had load balancer, your business model was about uptime. You attach to Workers and you made this the most customized load balance on the planet, and that was true for every product. Today, it’s much more. And the idea has changed over time. This is really, really interesting blog post that Matthew wrote a couple of – two years ago, I think it was now I said when we built workers and we looked at the use cases that could benefit from the ability to deploy code at the edge of our network. And you have to – you think about before workers, there were two places where you could write code. You could drive it on your – on an endpoint on your mobile phone and your laptop wherever compute speed was latency dependent and important or you could deploy code in a data center where speed and execution was the most important parameter the edge of our network, you try to optimize for both, right, almost at the latency as if it were on the device because you are less than 100 milliseconds away from the device, but at the speed if it were in the data center. So this is how this all started.
And so we saw it in the beginning that the most interesting use cases would all be about speed and taking advantage of that. We learned over time that it shifted that actually be deploying code at the edge of our network, together with the fact that we have this very global, but very local network allowed for use cases where the controlling the flow of data was the most important criteria, especially in a world where data serenity and data privacy is of increasing importance, whether it’s Europe, South Korea, Brazil.
So the ability to wrap something around the piece of data and saying, this is Phil’s credit card number at Credit Suisse. You can only be move between data centers in Switzerland and only in data centers of that security class became almost – all of a sudden, the most important use case. Today, this has evolved. It’s about controlling the flow of data; minimizing, maximizing speed; and minimizing Egress Fees data that has high moving philosophy needs to be moved between different clouds. And adoption has exploded. We are on our path to million developers, as you just said. And the diversity of use case is just incredible from controlling flow of data, building and rearchitecting complete business models on workers.
We just announced together with a large group of venture capital firms a $2 billion fund that supports start-ups that are founded on Workers, and we just priced out the first and awarded prices and support based on this. So it has become this unique tool, so to speak. And I’m glad we didn’t drive for revenue maximization, but for penetration and adoption. And it will it will get to revenue by itself.
Phil Winslow
Exactly give a time. Give it time, and so I can keep talking about Workers in R2 and D1 all day, but okay, let’s focus on go to market because I do get this question because obviously, that was one of the big announcements on the Q3 call. Obviously, Chris Merritt, incredible run do you know [indiscernible] Cloudflare. Matthew, I think, said less than 10, and I did some math here, and I could only find 7, it’s how people that have gone from zero to…
Thomas Seifert
If I have to bet, I would take Matthew’s number; no offence.
Phil Winslow
Okay. So maybe somewhere between 7% and 9%. 7% to 9% okay. So I must have forgotten somebody along the way. But point being very few. The – but what was the thought process I call it the skills, the experience that Marc now brings to when you think about sort of the next evolution of Cloudflare from a go-to-market motion perspective when you think about in particular that $5 billion target, what needed to evolve?
Thomas Seifert
It’s a good question. So it was a rather smooth process because Chris was rather proactive when we signed. So we had lots of time to get ready and find the right person. We think we’re convinced we found the right person with Marc. But if you – if you just go back from the person for a second and look at the business model, right, so there is freemium, there is pay-as-you-go, there is enterprise. So in our evolution you need a go-to-market leader that understands and is familiar with all three motions. It’s really easy. It’s a lot easier, I shouldn’t say easy. It’s a lot easier to find experience enterprise salespeople, but we are more than just enterprise sales. Our efficiency, our cost of customer acquisition, the ease of install of use for the product literally comes from the freemium and pay-as-you-go model, right?
If you sign-up with your home block or your home page, it takes you five minutes. If the largest bank in New York signs up, it takes five hours. So there is a reason why it only takes five hours. So you need somebody who is familiar with all three go-to-market motions and is comfortable in them. That was a big part. I think the – of course we needed somebody who has seen scale before this is getting to you learn that whatever gets you to 100 doesn’t get you to 500, will not get you to 1 billion, will not get you to 5 billion. So you need somebody who has seen that before. I think the other thing that makes us really unique is that our sales motion on all three go-to-market venues is a very data-driven approach. So this relationship is important, campaigns are important, personas are important that we target, but data really drives our go-to-market motion.
So you have to have somebody in this lead revenue responsibility position that is highly comfortable with data science and making use of the data, and we think Marc is. So he hit all the prerequisites and the chemistry was good, and he’s a fun person to work with. So he’s a couple of weeks in his job. I’m going to get his arms around it, because I think we are off to a good start.
Phil Winslow
Awesome, alright. Our 30 minutes is up. But I always like to end with one final question, which is, let’s say we’re sitting up here two, three years from now, what do you think that you’re going to look back on and say, hey, sort of this technology, this product is trend or maybe even call it something in the business model that it was more impactful to Cloudflare and its customers and maybe people were giving a credit for back in 2022?
Thomas Seifert
I think there are two parts that I would answer, for sure Workers. I think the opportunity, the potential, the disruption that will come with workers and how it develops and what we now call super cloud computing is going to be very disruptive. I still think that the architecture of the network itself that you have this off-the-shelf hardware in a software stack that allows you to deliver every product and service we have on every server. And now imagine the server is not in 200 cities, it’s in thousands of buildings. It could be on endpoints. We talked about having in virtual SIM card, without their SIM card is I think there’s still even after 10 years, the most underestimated part in the competitive moat we have built and we will look back to that and say this drives, and being global and so much local at the same time is I think what makes us…
Phil Winslow
Well, the network is a computer. All right, fantastic.
Thomas Seifert
Thank you.
Phil Winslow
I appreciate the time, my friend as always. Good seeing you.
Thomas Seifert
Thank you.
Question-and-Answer Session
Q –