Cloudflare, Inc. (NET) Earnings Call Transcript & Summary

All right. All right. Okay. All right. Thank you all for coming. I wish I could have lunch, but maybe I'll try to figure that out. I'm having protein bars all day. So...

Oh gosh. We can get you some real food at some point, Matt.

All right. I thank you all for coming. I've known Cloudflare for a long, long time. I've known Michelle a long, long time. We've traveled together. We've laughed together. We -- our kids growing up together.

I think we've cried together.

I think we were in Canada once we got that...

Okay. Okay.

So yes, I do appreciate the partnership and the opportunity to chat here. We got a lot to cover here. And then, Phil, thanks for bringing the team here. Appreciate you as well. All right. So let's get started.

Sure, let's do it.

So we were just talking kind of off stage a little bit about sort of the new branding that I got to give you credit for because I think you came up with the new moniker here. Because I think in the past, it's always been a little bit of challenge. What is Cloudflare? What are you trying to solve? And you announced it on the Q3 call, the connectivity cloud. So -- and I think from what I heard, you -- that was sort of like -- were part of the naming there. Tell us about what that means to you? Why is that resonating internally? But then even more so, why is it resonating externally as well?

I know. You're right. Before, Cloudflare was described the things that we weren't. We're not this. We're not that. So it feels really good to say Cloudflare is the connectivity cloud. We love it, and I was part of it. But these things take a group and so [ Faye ], my colleague who's here, Phil, many others that were part of that, but it was actually really fun exercise to go through. So Cloudflare is the connectivity cloud. What do we mean by that? We help connect people, locations, data, where they want to have -- make it all work together, the fabric that stitches it all together. In comparison, some other cloud companies that have been around for a long time, in many ways, the business models were built on captivating the data, keeping the -- [ people ] the data walked in. I would think for the next -- what we've been doing and the next innings of the Internet, it's going to be more about connecting everything, whether it's employees, locations, data centers, cloud services. We are the fabric that it connects it all together. So the connectivity cloud it is. And I would say that when we go through these exercises, you never know how it's going to be received. We introduced it to the world at the end of September. And I would say that we've been really -- it's resonated. It's resonated with our customers. It's resonated with investors. I'd say, generally, people like -- this makes a lot of sense and I like it. So that's always a good place to be.

A name to a face. We didn't talk about what the plan B was, but I'm sure there -- was somewhat left on the drawing table.

That's for a...

We can talk about that offline.

Yes, offline. Yes, offline. Yes, exactly.

So maybe building on that, to this day, I still -- like you guys -- I've known you forever, and you guys has been public for many, many years. But I still get the question like why are they so different. What is the secret sauce? And we've talked about this in the past, and I think it's -- there's been a couple of core tenets and you have built technology on top of that. But maybe just for those that are still maybe a little confused about why Cloudflare is so unique, talk about some of those -- the secret sauce of Cloudflare.

For sure. Yes. Thank you. It's interesting. The Internet's about 30 years old. And I actually think COVID has really made us all realize how important the web and the Internet is to our lives, to businesses. It's accelerated a lot of trends that have been happening for a long time. But it's kind of went from 15 years ago, maybe a business had a website for branding to actually now business is having to think about digital first processes and whatnot. So the reason why it's hard to understand -- I'm going to answer your question, but the reason why you're like I don't really get it is there's a huge shift happening underneath all of us, and it sometimes is hard to see. And I think COVID kind of helped say, wow, cloud, the Internet, this is more important than it was the first 20 years of the Internet. And I'm 100% sure it's going to be more important in the next 20 years to businesses and people and societies. And so that's why it's really important to pay attention. And so it's interesting. When we started Cloudflare, we saw this shift. We saw the shift from hardware and software that really made the Internet run. The first 20 years was server rooms, data center rooms with a lot of hardware stacked on top of each other to a bunch of things you couldn't see, like cloud services or SaaS or apps or APIs. You're like what is that stuff. I can't feel it. I can't walk into this server room that's really loud and hot. I don't understand where is everything sitting and it was hard to understand. And so what's interesting with Cloudflare is we've built a global network. And we're 13 years old, and we built our global network in a very modern way that is different than any other global network that was built before us. So we have a differentiation around our technology and architecture. And we have really some strong principles around that, is we want all of the services we offer to run in every point of presence close to our users who are connecting to our network. And that's different than how the first-generation global networks built their network. And it turns out, we think that's very powerful. So that's one unique sort of thing. The second thing is we've kind of just -- James Barksdale has that famous quote of Internet's all about bundling and unbundling. The last 10 years, which is really kind of like the -- first 15 years of the cloud was all about these like point cloud solutions. People built the solution that solved one problem for their customers. And it worked. There was this huge trend. People are buying it to help solve one problem. That's not going to be the next 15 years. And so the second thing that really differentiates Cloudflare is that we're a platform that does a lot of different things for our customers. And we believe an integrated platform is the winning strategy. And that sets us apart from a point solution, where it's, sure, you could buy that 1 point solution or you could buy that plus 3 other things from Cloudflare. You go on a platform and all of a sudden, you have easier operational excellence, lower total cost of ownership. And by the way, there's no latency, which is a really nice value proposition. And the third, I would just say, so there's a technology differentiation. There is our product offering, and then our go to market is interesting too, where we -- again, before Cloudflare, folks really kind of said we are either an enterprise company or an SMB company. You kind of have to declare. And what's interesting about what we do, Cloudflare helps make everything faster, safer, more reliable online. Everybody needs that, everyone. If you have a son in high school working on a hobby project, he needs something in front of his project, otherwise, it'll get knocked offline. He probably uses Cloudflare on the free site; or if you're a small business owner or you're some sort of agency sort of thing, I'm a photographer, I make portfolio online, you need Cloudflare to make it fast, safe, reliable around the world. And we have these price points that help fit these kind of different use cases. At the same time, if you're a Fortune 1000 company, you are looking for, hey, how do we release that fast, safe, reliable, and we have -- 30% of the Fortune 1000 is Cloudflare customers. So it's interesting. What we do is very horizontal. So our go-to-market is very horizontal. We have price points that start at free up to businesses paying us tens of millions of dollars a year. And that is different than I would say compared to 10 years ago where you had to declare. And we kind of said, no, Cloudflare is something that lots of people can use. We're going to fit the right motion, go-to-market motion to the right price point. And I think that that's been really important because it made us easy to use because if you're starting at free, you got to be easy to use. They can't call you. You're not paying for people to have -- field those calls, but it turns out enterprises want it to be easy to use too because they want their teams to be more agile, freed up to do sort of things. So it becomes this circulus flywheel. And I think all of you as investors love flywheels. So do we as business owners.

Yes. That's great. Well, I think it really does then circle back to the connectivity cloud. It really builds the foundation for everything you're doing.

Exactly. It really does.

Let's talk about some of the recent results. There's been a lot of investor anxiety this year about macro and demand and cloud cost optimization and all sorts of concerns. You guys saw a bit of an inflection point in Q2. Q3 was kind of more of the same, even a little bit of sequential acceleration. Can you talk about some of the underlying trends that you're seeing? And I think we're all hinging at like what is AWS going to say or what is Azure going to say. What are some of these trends that you're seeing support the growth that you've presented?

Yes, yes. It's hard to predict the future. And so -- but it's easy to look, okay, what are we seeing. And I mean it's clear. Two years ago -- when you were at this conference 2 years ago, like the world was really frothy. We were all geniuses. Crypto was up every day. The markets were up every day. Everyone was hiring like left, right and center, and it's definitely not that, right? And I think that that's become very clear the last 18 months. We did see, I'd say, things kind of bottom out in Q2. And so it hasn't gotten worse, although I was -- we're here, of course, seeing investors but also spending time with customers, and I was at a customer's office earlier today. And the CEO was saying, yes, we haven't seen it get worse, but I'm not sure it's gotten that much better yet. It's kind of uneven. And so everyone's kind of closing their breath of saying, okay, it definitely hasn't deteriorated. But I'm not sure it's been as fast to [ sent out ] either. So I think that's important. So that's just kind of like the backdrop. Some data that I think is interesting to look at like is, okay, what do sales cycles look like. Are companies deferring decisions? Or are they making a decision? And that's important, especially if you're an investor or for me, running my business. I'm always saying, okay, is the buyer deciding or are they deferring the decision. And we're seeing more companies decide. So it's clear, are we winning or not versus we're just going to push this out. Those are different things. And so we're seeing more companies saying, okay, well, I can't defer forever. I got to modernize my enterprise. I got to find the things that are must dos, not nice to dos. Where can I start? And so sales cycles, win rates and are companies deciding, so these are some of the -- of course, productivity of your sales team matters a lot in terms of running your business. And those are some things where, again, we've seen improvements. Pipeline generation, is the pipeline healthy? Are you able to create that demand? Those are important things for enterprise software company. And so we look at all those metrics. And again, they're not getting worse. And so they were positive in Q3, which we spoke about in our earnings call, but we continue to keep a tight eye on that because what you want is the agility to say, if something changes that you can react. If it starts fast and sent out, okay, great, let's lean in versus, if it's uneven, let's be patient and wait for that moment where -- and keep being very disciplined in how you operate the business.

Yes. That's great insight. The other -- we'll talk about GenAI in a second here.

Well, it's 7 minutes. AI changes every 7 minutes. That's my joke. But it's like true. Every 7 minutes, there's something new.

It really is crazy.

Yes.

And if you can put some sort of an AI moniker on your website, it certainly helps as well valuation-wise. Before we get there, sort of a macro-related question, cloud cost optimization. I know you guys see different trends than other people. We all know what AWS and Azure and GCP said. And we look at some of the consumption companies and some are talking about better things. Some are talking about maybe some more sort of in-line things. How do you think about cloud cost optimization, either as a business and how you think about spending your own dollars or how your customers are? Are we -- to a bottom of that -- does that sort of like parallel back to your comments of like things aren't getting necessarily better, they're not getting worse, but does that mean like cloud optimization trends have kind of stabilized?

I live in San Francisco, and right now, politics in San Francisco is about returning to common sense. And I actually think there's some parallels with the cloud spend, so stick with me for a second. So I said that the cloud is about 15 years old, okay? And I would say the first 15 years is really about being cloud first. And so there's all a rush for businesses to say, okay, we've got to spin up our workloads in the cloud. And that's what we've really seen. We're shifting to cloud smart. We've learned a lot. Okay? And so now there's this awareness, common sense around the decision makers of we have to be cloud smart. There are certain things that work really well here, use what we need, rationalize what we don't. And that's just because this is this new -- it's the shift that I was talking about in the last 15 years. It's a new thing. And so the last couple of years, we were like, okay, let's just take a breath and figure out where do we want to invest, what are our tools, how do we continue to do what's right for the business. So I think cloud smart is really what we're going to hear more of. And I hear different leaders talk about that. Related but slightly different and of course, it's a smart strategy. Like, well, I don't want to pay for more than I'm using. That's just kind of common sense. I don't want to have all these pre-commits but then for it to sit idle for 3 years. That doesn't make any sense. Idle for 2 months, fine for it's there. That's not a big deal, but -- and so I can grow into it, but not if it's going to take me 2 years. And so -- and that's the same with licensing, seats licensing. My employee profile looks different. And so I think it's kind of common sense with tech tools of it was a little bit of we rushed to this innovative services and now we're going to be smart about, well, making sure that we can keep the agility in our business. Related but certainly different is there's a lot of companies that have been around for a long, long time. And in my role, I spent a lot of time with CIOs and CISOs at organizations. And I've been doing -- I mentioned earlier that I've gone to visit a couple of customers this morning before I was here, and we went to see one CEO, one CIO, and both of them brought up how can you help me consolidate my spend with legacy vendors. They want to know that. And that's -- when they're talking about legacy vendors, they're talking about often hardware or other point cloud solutions that are like I don't want to pay for all this. I want to take 6 vendors put it onto a platform and I want to see a 30%, 40% savings in that. And I was in France in October. Carrefour, large multinational business, we went to this -- there was a cybersecurity conference in France called Les Assises and [ Guillaume ] was our decision-maker there, and that's exactly what they did. They took 5 vendors, moved to Cloudflare. He's like I had a team of 7. Now they're a team of 4. I have one control playing all the same data. The connectivity is all there. And I've not fired those other 3 people. I freed them up to do other things, and our business is so much more efficient, more modern. We can do other sorts of things. And so I think a lot of organizations are saying, okay, common sense with the tools we're using. Make sure we're using what we're doing before we bring in another vendor. Can one of our current vendors do that? And then how do I take these hardware or point solutions and move it to a platform so I can get better total cost of ownership. So those -- they're related but slightly different, too. That's what I'm seeing a lot in the conversation. So what does that mean from an investor? I'll let you figure that out, Matt. That's your job for that.

Well, we -- and actually, that dovetails into my question because I was hoping you'd provide some insight here. One of our themes, I think, when we talk about 2024 is the bigger are going to get bigger. And I think to your point, we had a keynote this morning. We were talking about RBC spend, and we want to do more with fewer vendors. I mean that's sort of -- and you said that was the message today. Has that narrative with Cloudflare changed? Were you having those conversations even a couple of years ago? Were you perceived differently by your customers and today, it's more like, yes, I can consolidate a lot of spend on Cloudflare?

I would say it's different. And I'm not sure it's even specific to Cloudflare. I'm just saying 3 years ago, that's not was the psyche of the decision-makers of the tech buyers at these companies because it was -- again, things were frothy, right? It just wasn't that. I think that for good or bad, the tough macroeconomic environment that came on pretty suddenly has been a wake-up call, that back to common sense, prudence, like discipline is actually important. There's a lot of pressures on the cost side to do more with every dollar whether that's more productivity but also, okay, how do I make that dollar go further. Like every business -- almost 95% of companies, maybe 5% are not impacted by that, but most companies have just been like, whoa, it's been a wake-up call. Of course, you layer in some of the instability on the geopolitical side, and that kind of makes people weary. And so I think it's -- not to make lemonade out of lemons, I actually think it's been healthy. Back to companies who run a good ship, who teamed who take this seriously are going to end up being in a much better position than 3 years ago, where it was like, I don't know. Just like throw the kitchen sink at it, go, go, go as fast as possible. Don't think. And so 3 years ago, this was not something we heard. I would say, even 18 months ago, it wasn't something we heard because people were still trying to figure out coming out of COVID what was important and what they were going to do. But now -- maybe all of you have been talking about this for a long time, but now the CIOs are talking about it. And they -- especially budget planning for next year, they're all under pressure to figure out how I'm going to do more with less, what's my legacy IT replacement plan, who can I go on a journey with, who's going to go from vendors to strategic partners. That's another way to say it, consolidation. It's how do I go from a list of vendors to betting on a few strategic partners. And so what I love and what I've always loved about Cloudflare is I feel like we're well positioned to be one of those strategic partners for organizations. And we have a bunch of those today, which is wonderful. And so when you can go talk to these organizations saying we obviously are a strategic partner to other customers, we love to have the opportunity to win that with you, and we will show up every day and do the work to earn your trust and show you the ROI in your investments, and I think that's a winning strategy.

Yes. That's great. One more before we get to sort of the elephant in the room, GenAI. But a lot of investors, certainly on the growth side, worry about deceleration. And I think when we're thinking about some of the building blocks for next year, obviously, you guys haven't guided to next year yet. But maybe talk about sort of the building blocks to durable growth and how you think about the longevity of this because, candidly, you've seen growth companies go from 50 to 40 to 30 and some might go to 20 and 10. Like how should we think about the growth trajectory here?

So Mark Hawkins is on our Board, and he was the CFO and President of Salesforce kind of from $3 billion to $30 billion. So he joined our Board a couple of years ago, great person. Many of you probably know him. If you don't, you should -- wonderful human being. And he has a really good quote. He's got kind of a Texan accent a little bit. He lives in Austin now with his family. He's very close to his family and he said -- he's like when there are bumps in the road -- so during this like top macro environment, he's like stay close to the customers because then you feel the bumps with your customers. Okay? And that was some advice he have us to a Board meeting a couple of years ago. And I -- in my role, I spent a lot of time with customers, and it's not just me, our whole senior leadership team, every week at our senior leadership team, one of the executive's talking about the customer story from their customer call. The week before our leaders have now had their direct reports spend time with customers. And I bring this up because you go spend time with a prospect or a customer and you learn a ton. And you realize we are solving real problems for them. And at the end of the day, when you're running a business, it's are you solving a problem for somebody. And I think it's great if your technology is super cool and you talk about all the awesome shiny things, but it's like what problem are you actually solving for somebody. And it's really healthy to go spend time with the people who are using your product to go what problems do we solve for you. I think I know, but let's hear it in your own words. And so the durable building blocks at the foundation is Cloudflare solves real problems for our customers. And those problems are ubiquitous. Every organization has similar ones. They're not bespoke to one. And so just some examples to illustrate that. Cybersecurity, that's a big word. It means a lot of different things. But really, if you were a business today, it is something you care a lot about, and everything Internet-facing is like how do I make sure everything facing the web is fast, safe and reliable. Cloudflare is the market-leading solution not only because we have a lot of customers, but you look at all the Gartner Magic Transit -- Quadrant, excuse me. Like we're top right, which is great. We're like, okay, we have a lot of customers using that, but we know that we don't have everyone because every day, there's some other business facing some attack that's knocked offline, and they're not a Cloudflare customer yet. And so you're like, okay. So it's like great. We know that use case. We're so good at it. We're so good to onboard those customers. We know that we're going to continue to grow there. You talk about Zero Trust, talk about AI, developer platform, networking, all these things that we do. And so it goes back to are you solving problems for your customers. And what I love about Cloudflare is the answer is yes, and it's not one problem. It's many problems, which makes it hard to explain what we do and understand and it's like, wow, I wish there's only 3. And what's the percentage of distribution between those 3? No, no, the answer is there's many, and we're going to have more customers using each of those use cases and they kind of come and go depending on what's going on, and I think that's a good place to be in. So that's what I think is the building blocks of our growth. We solve real problems for our customers. These trends are accelerating. They're not slowing down. Cloud usage, the Internet for internal corporate networks, that was never a thing 5 years ago. No one thought about their internal corporate networks as touching Internet. It was almost like you didn't want to put your corporate networks on the Internet. But now your employees are working from anywhere. They have to touch the Internet. They have to. So that's why these things are like accelerating. Employees are working from everywhere, rise of the cloud. And so it's just -- it's one of these things, like there's real problems that we can help solve for our customers. And I think like that's what we are very laser focused on.

Yes. Let's pivot now to a topic that we didn't even talk about a year ago. It changes every 7 minutes.

Yes.

One of the favorite quotes from this, I got it from Mark down here on the front or my [ DOR ] and his -- the quote that he heard really resonated with me, is AI is not going to replace jobs, but people that use AI will replace jobs. That really resonated with me, I think, when you think about the context of AI and sort of what it means. So talk about -- we'll talk about the cloud first specific drivers. But talk about sort of what does this tech trend mean because we've talked about AI and ML forever. Why is it different now? And is GenAI, in your mind, is it really as revolutionary as sort of we all read about?

So yes, I do believe. So just to put it -- my framing of the magnitude of this is the invention of electricity, the invention of the Internet, kind of this lightning bolt of AI, they're like the same order of importance to the shifts that are going to drive our economies and lives. So I think it's a big deal. What's changed, and I think this is important, is because we have talked about AI for a long time, like RBC, you've been using AI a lot in your business for a long time. I know I've heard Dave McKay speak about that, how RBC is really forward leaning on this. But what changed was -- and you're not the only business, but you've made a strategic bet, you invested in this is up until about a year ago, call it a year ago, 2 years -- a year ago when ChatGPT came out, it took pretty highly skilled labor and a strategic bet by the business to invest in it. You had to hire a lot of PhDs. There wasn't the talent. It spent a lot of infrastructure to make the models work. Like people are spending a lot of money on this. I don't know how much you were spending, but it wasn't free at RBC. You're spending money on it. What's changed is almost like it got democratized. And it's something that was like really tightly held by some PhDs that were really smart with a lot of money to go solve business problems also -- and became -- the building blocks got democratized. And so it's -- all of a sudden, it's like, wait, who can build with this? And so what's happened is consumers are now using it through ChatGPT, but a lot of the building blocks have become much more democratized. So bunch of entrepreneurs are building things with it. Some of those things are fads. Some are not -- don't have real -- they're not actually solving a problem, so they're going out of business. But a bunch of them, because the infrastructure got democratized, they're like, wow, what else can I build with it. And they're building these use cases and ideas that probably will help change a lot of productivity in businesses. So that's kind of what's changed, and I'm really excited about it. It's messy. It's going to be noisy. There are going to be a lot of winners and losers. And someone told me -- I live in the Silicon Valley, so someone told me how much venture has been deployed to AI investments. And like there's no way you're going to get any return on all of that. Like there's a lot of losers in there, but you're just placing bets because it's just moving so quickly. Having said that, if you look at a business, what does this mean? Well, the first thing you have to ask yourself is, is AI a tailwind for my business or a headwind for my business. And you definitely want the answer to be a tailwind, right? And if it's a headwind, you're in big trouble, right? You got to figure out something. You got to do something different. And so for us, we think of it as a tailwind. And why do I say that? Well, we've also been using AI and machine learning in our cybersecurity products for a long time. Actually, that's like the whole premise of Cloudflare. We crowdsource all our cyber data around the world across all of these customers, so we're really good at finding 0-day exploits, attacks faster than anybody else. And anyway, cybersecurity is kind of a data science game. And so AI is really good. How do you predict? How do you know and same on the reliability side? So we've been using it for a long time. So now that it's more democratized, it's easier to get the talent. It's like, okay, great, we can go faster. So we think that's interesting. But then in your business, you're like, well, how do we help run our business processes faster? And there's just a lot of things because these businesses have popped up of, well, how do you do sales and marketing. There's lots of ways. With all people talking to customers all the time, how do you make sure you help pull the right resources to put in front of the customer based on the conversation? Kind of makes that salesperson more effective, right, than without having this tool. So back to this copilot model, which I think is important. So are these business processes, which every business used to be thinking about? And then the third bucket, which I'm really excited about, which is, again, tied to Cloudflare is, well, how do you enable this for the next generation. And so we're doing a lot of cool stuff there that we can talk about. But I do think it's a big deal. I do think there's been a change. I think that it's going to continue to speed up what you happen. It doesn't mean everything will win. Not everything is a good idea, but it is going to make businesses more productive and more efficient. I was at a dinner. Obviously -- and this was at a dinner a few weeks ago. And it was a lot of CEOs of tech companies. And this, of course, had been 7 minutes. This topic came up, and they were saying how a lot of these AI copilots making their developer teams more efficient. And so then the CFO at the dinner said, so does that mean you're committing to 30% more output next year or can I pencil in 30% less spend? And then I was like, whoa, whoa, whoa, don't take my budget. You can't take away. Maybe I'll give you like a 10% productivity gain just because I'm not sure. And so I think it's still early, but I am very, very -- it's a fast shift. And those who are leading in, I think will be the benefactors and those who aren't will be the legacy -- tomorrow's legacy players.

Yes, that's great. So let's talk about some of -- why you think you guys are going to be in that winner's circle. You guys just had your birthday week. You're teenagers now, 13.

Yes, we turned 13. Yes, does anyone here have a 13 year old at home?

Yes. I...

Some over there.

You're getting close.

Mine's 10. I have a 10 year old. Yes.

So let's talk about how do customers view Cloudflare as a landing spot for AI. We've talked a lot about inference models. Like why Cloudflare? In the winners and losers, how do customers see Cloudflare positioned?

Okay. Well, first of all, there's -- it's an interesting customer acquisition. So all these AI companies that have been born fairly recently, some of them are growing like weed. And it turns out, when you're growing like weed, scaling, being fast, safer and reliable, they get attacked all the time, well, guess what? Our bread-and-butter products work really well. And so we have actually a high percentage of the AI start-ups using our core services because it's exactly what we do. We help businesses scale online and they're all Internet-facing. They're all modern. They need something in front of them. They're using Cloudflare. So that's been an interesting segment of customers who've signed up and some are on the pay as you go. They don't pay us very much. And there's others who pay us a lot because they've hit velocity -- escape velocity, right? And so it's great we can grow with them. Love that. Looks like -- that's great. But -- and then, again, we use it in our products, which I think is really, really exciting. We can talk more about that. But the third one, which is unique to Cloudflare and it's back to what sets us apart, as I was telling you earlier about our global network and how we've architected it and how we're really proud of it. Every service runs on every server. It's very modern how we do what we do because we've built in the last 13 years. We learned from those ahead of us, and we feel like we have a distinct -- a differentiated point of view of how we've gone to market with our network. So we've put GPUs at our edge. We've put GPUs into our network. And we believe we are the only globally distributed network that's done that. So why does that matter? Well, all of a sudden, up until today, if you wanted to do this, you'd have GPUs at like CoreWeave. It's like Azure had them and everyone was using CoreWeave. That was all in the U.S. All of that's done in the U.S. So if you're a business in France, and you want to do something, all the data had to come back to the U.S. to do the machine learning and data science. And again, I've heard Dave speak about it. I don't know. He's not here. But I've heard Dave McKay speak about it, and I'm pretty sure RBC had a relation with Microsoft. You did all your machine learning in the U.S. too because that's the only place you could do it before. Apples of the world, they put a lot on the device. I don't have my iPhone on me. It's in my purse over there, but it's a lot of the other like device companies run a lot of inference of the device. All of a sudden now, the companies put GPUs in their network, and we're in 100 countries around the world, 300 cities. We're not in all of it yet. We're at about 50 cities right now. All of a sudden, we can do inference -- there's a new place to do inference that's not the device and that it's not back at the cloud storage place that's currently only in the U.S. There's another place. So will all of a sudden, what can we do with inference now that's distributed much closer. It's kind of like there's a new place where you can do it. And it's interesting. It's early, it's early. I mean we just launched this at the end of September. So we're about 5 weeks in. In the earnings call, we said we've been doing about 18 million requests, where people are developers, are trying to figure out what to do with it. So we believe that the fact that there's a new place to do inference beyond just the end device is going to unlock new use cases and innovations and that businesses can bring forward. And we're really excited to see what they are. And in terms of what makes us different, as far as I'm concerned, we're the only one who currently does this. Doesn't mean others won't, but they haven't yet. And so we feel like we're able to help bring this mark to the market and drive this innovation. And it's really exciting place to see. We feel like there's going to be some really interesting innovations that will be uncovered that just weren't possible 6 weeks ago. And I think that, that -- we've all had a long career as you kind of think of those moments of when something changed and that became this chain reaction to something else. It doesn't always work. There's lots of things where it's like maybe it doesn't, but it does feel like we're well positioned to help drive that forward.

So it's one thing if -- like an RBC is leveraging a Cloudflare platform for that. But to hear some of these AI leaders -- GenAI leaders, give us some examples of what some of these companies are doing with Cloudflare right now because they obviously have choice, and there's -- like you said, Azure is doing a lot right now. But why Cloudflare for some of these strategic thinkers from a GenAI perspective?

It's back to, okay, you have a device. You have doorbells. You have cars -- again, in San Francisco, there are literally self-driving cars now on the street. I took one to Jerry Seinfeld on Friday, went to Jerry Seinfeld with my mom and dad on Friday, and we took a self-driving car there and just like driving you around. And of course, that car has a lot of power, doorbells. But now if you want to do something like predict, hey, do this. If you're the delivery man, we're moving the package at my door. Here, do this and know how to do it. Instead of going back to the edge, it's like can Cloudflare's edge help? All of these IoT devices around the world that are floating around that have limited memory and limited battery, it's -- I want to use the data to make predictions or do more, make them come more alive. And now all of a sudden, we believe with inference at the edge, you can do that. Whereas before, coming back to like the central location, could have been too slow because when you're at the doorbell, if the request has to go all the way back to, I don't know, Seattle and your doorbell's in New York and back, like the person already left with the flowers or the package. And so it's like, well, what could you do? But it's like making the prediction, thinking through this. So it's a lot of things that were like IoT -- the next generation of IoT, making them more smart and powerful, so they're not just a doorbell with a video camera. They're a doorbell with a video camera that can then do something more intelligent. That's an example. Or these self-driving cars, it's like, okay, how do you make them more intelligent? There's a lot with like IoT devices that folks are thinking about. But there's also just around the AI start-ups themselves. They're doing a ton of things, the models, they're trying to get it distributed. And again, there's just a lot of interesting early innings of the use cases. And again, not all of them will work. Some of them will be fads, but there's, for sure, some real things in there, and it's early. And so I think the question to keep asking us and others are what are the use cases that you see that are interesting because as soon as you can explain that and then it's easier to make it replicable.

We talked about that this morning on our panel, and I guess that -- it's a great segue. So the IoT use case makes a ton of sense. What about some just like Main Street, U.S.A. use cases where it's not as like maybe intuitive? What are some of those use cases?

It's -- we co-mingle these, so I'll just put it apart. So we have a development platform that runs on the same network as AI that we're talking about that's inference, where we run, store, compute at the edge. And actually, that's been in the market longer. It's been in the market 6 years. So we take a lot of learnings on what we just put in the market 6 weeks ago and what's been in the market for 6 years, which is early there, too. That's a whole generational -- it's kind of the next shift of computing platforms in our opinion. And so we have a big bet there with something called Cloudflare Workers. We have a storage. We have a database, SQL database now. And so we look at a lot of things that we're learning about use cases around workers and this is our serverless compute platform. And what we're seeing are things that are hard to do back at the centralized store, compute location. And again, you don't think about this. The first innings of the cloud were so fast. But now it's like, well, actually, you don't want to put that in a centralized place. You want to put it back at the edge of the network where it's much closer to the device, so some really very common things that seem silly. A/B testing, before, that was very centralized and you'd sign up for the service and everything ran in the mainframe. Well, actually, the users are doing all this stuff much -- around the world, the bus shelters around the world. You want to be as close to the device as possible. And so what's interesting, businesses are running a lot of A/B tests using Cloudflare Workers because just we're already everywhere. We're already where everyone is, where their users are connecting. Let's run the A/B tests right there. There's a lot with fraud, checkout. There's like a lot of things that we see, which are, again, plain vanilla. You're like that's not that big of a deal, but it is if you're running a business. Turns out you care a lot about these things where it's like this is just more lightweight, dynamic and powerful. And so I think that's really interesting. Another example are the things that used to be really expensive and hard from a technical standpoint that are getting easier to do. And so there's a publishing company where they're running a lot of content, like gated content. You see all the media sites now. It's actually really hard to put those gated pages up. They're slow. They slow things down. It's another one where it's like, well, actually, the licensing, digital, the [ DRI ] rights, the digital rights licensing, it's easier if you do it in the network closer to the use case. So some things that used to be -- and that one, that drives revenue, right? That one is, every time we leak some content to someone who doesn't have access to it, I lose money. And so that's something that we see as, again, every business has some use case that looks like that of I want to control access to something. And all of a sudden, I can do that much closer to where the user is. So I have less leakage because there's no more latency around that. So again, it just kind of goes from being in one location that's really powerful to saying, actually, I can do a bunch of this closer to the device or the user and actually, it becomes lightweight, more flexible and actually solves some problems that weren't working so well in that one place. So I think it's just this next inning of the cloud and back to cloud smart and what do you do there, what do you do at the edge. And more and more things are going to move out there in our opinion.

The other thing that you announced at Birthday Week, along with Workers AI, is partnerships with some pretty strategic vendors, Microsoft, NVIDIA, Meta, Databricks, Hugging Face. Why are they partnering with you guys? Because they can do whatever they want in some instance. But why cloud -- what was so unique about you guys for them?

One thing that I love is that I get to work with startups, like all the cool startups, like Hugging Face, like that's -- they're newer companies.

Hugging Face is just the best.

[indiscernible] Hugging Face, what is that? Just imagine someone trying to explain that to their mom, it's like, "What's that," or their aunts. But anyway, so like NVIDIA and Microsoft. And again, I think it speaks to this how big of a deal AI is, like again, electricity, the Internet, AI. I put it all on the same level of magnitude. And so what's interesting is Microsoft has been investing on this for a long time. NVIDIA, obviously, is a huge -- doing a lot -- has benefited their business -- has benefited a lot from this part of the market. But then the rise of some of the newer, cutting-edge, innovative companies, where the models are running and the LLM models so that it gets distributed faster. What's great about -- back to like why Cloudflare is because we bring something unique to the conversation. We bring the globally distributed network. And there aren't very many companies that have a globally distributed network. And there aren't very many that run GPU at the edge. And by the way, like GPUs at the edge plus the store compute capability, there are very few. Right now, they might be precisely one. We won't be the only one forever, but we are right now. And so it gives us these opportunities to partner with both the really equal startups. And some of those will become the next Facebook, awesome. And the big companies have big bets there. And I think that, that is interesting. And as a founder/entrepreneur, I know a lot of entrepreneurs, I know as shareholders, you need a lot of the entrepreneurs, you kind of realize that's not normal. We kind of sit in rarefied air. And I think it's partly because what we do is so unique and differentiated. It's very hard tech. But we have a unique asset. We sit in some -- we sit in a very strategic place online that can add a lot of value to the conversation.

How -- I think one of the questions that we get from investors all the time is, "Okay, it all sounds great, but how is it going to drive revenue?" How do you -- a year from now, 2 years from now, how do you expect to talk about success in driving real -- maybe it's part of that -- the durability component to it.

It is. We've got to find a use case. Again, it's -- again, I think it's -- when new tech comes out, I love that people are placing bets. But at the end of the day, are you solving a problem that companies have? That's where you make money. Like I just said, you can't forget about that. And so we are very much keeping our eyes open to figure out where the use cases that are solving real problems. But it's been 6 weeks. So it's going to take a little bit longer. And that's okay. Having said that, when I go back to Workers, which has been in the market 6 years, again many of you -- some of you have been investors for a long time. Some might not know us yet. I know, Matt, you've been covering us for a long time, and we really appreciate it. I think that sometimes people think, "Well, Workers should be huge out of the gate," or like, "AI should be huge out of the gate." And maybe it will be. Or it might be more of a -- like a durable growth sort of thing, where it keeps building and gaining momentum every year. And then all of a sudden, it comes really fast. Because everything always takes longer and costs more than you think. Because just like entrepreneurship, we only kind of realize that. But what's interesting with Workers, it's been there for 6 years. But how many people are using it? Both obvious developers to big companies, like the Major League Baseball, they run the whole MLB app using Cloudflare Workers and WebSockets. I don't know if any you are baseball fans, all the 3D stuff, all built on our technology. And it was new this year. Because 3 years ago, that wasn't really possible. And I think like that's really great. So now that they're doing it, who else is going to follow suit? And you kind of keep building all these sorts of things. And so what we see with Workers from a revenue standpoint is at first, people use it, hobbyists, they're just trying to innovate. They start to find real use cases. You make those repeatable. And then you become a real part of their development cycle. And that's a really -- that is what drives the growth long term. R2 is competing with S3, which is one of our kind of storage products as part of our developer platform. And what's interesting is, I think, at first, there is a little bit of rhetoric of like, "Oh, we think we'll just shift from S3 to R2." Nobody wants to do that. No business wants to just go and take S3 and move it to R2. That's just not -- they have lots of other things to do. But how do you make it easy to go from S3 to R2? And so we've built really interesting migration technology to help businesses as that's happening. And why you want to do that is also your data is in a neutral place, it's not captive. So whether you want to do machine learning on it with AWS or Google Cloud, you can, or Microsoft Azure. And it's a way to get it out of there. And so we see a lot of net -- businesses saying, "Okay, I'll shift slowly using your migration technology, but that's going to take some time. But anything that I'm going to put in R2 because it gives me way more flexibility and optionality." And so -- and then we have D1. Anyway, so I think a lot of that similar -- we've kind of seen how it's played out on the developer platform. And so I'm just thinking the same thing on the AI side. Although if it hits, it will hit fast. Again, we're well positioned. So we're looking for those use cases that are solving real problems for customers. And by the way, that's what Mistral of France is doing and Hugging Face is doing. Everyone is looking for the real use cases that are durable.

So it's a segue into another question. Take your -- maybe take your Cloudflare hat off for a second. I think a lot of us struggle with how do we pick winners and losers in gen AI? So really, do you pick the piping layer? Do you pick the app layer? Do you pick the reference models? How do you define winners and losers in your mind if you're looking out at other tech or other sort of trends? And how can that help us think about winners and losers?

Well, again, I'm going to -- and that might be a bit above my pay grade, Matt. But a couple of things. I know I've said it a lot, but I just -- because I've known -- we're 13 years older, Matthew and I at least started this business, we're still running it. I know so many businesses that have ran out of legs. They've gotten bought by private equity firms or they've just been outcompeted and they don't really -- they've kind of become zombie companies really is what they've become. And so you've just -- I've learned a lot through the other entrepreneurs kind of through my class or era. And at the end of the day, if you're solving problems that are growing, it's a good place to be. Can you expand your market share? Like can you expand your total addressable market? Like why you versus something else? And I think in Cloudflare, we started doing one thing. Today, we do a lot of things. And that's a good place to be. Our opportunity has expanded over time. Most investments narrow over time. That's -- one of our venture capitalists shared that with me. He's like, "Look, the best investments I've ever made are ones that expand over time." So what that means for picker -- like winners -- picking winners and losers is, okay, what problem are you actually solving? Do you have real customers that are paying you to solve those problems? Why are you unique? And over time, can you expand what you do for your customers in the market? Or is it going to narrow? Are you going to becoming like a much more of a feature versus a company or a platform? And I think that if you just kind of look back, we're 13 years older, there's a bunch of companies that started at the same time as us. The ones that are still doing really well are the ones that have expanded, solving real problems. They have a real reason why, like technical differentiation, which is wonderful structural difference, versus the ones who were really hot but kind of fizzled away. And it's often because the opportunity narrowed and because they didn't know what to go to next. And that's a really bad place if you're an investor in a company and they're not sure what to do next. You just don't want that to be the case.

So that's a great segue into what to do next. One of the -- it's cyber. And you guys have -- it's one of the questions that I get asked all the time about Cloudflare. Because I think we all know you guys have been doing DDoS and bot mitigation for a long time. But all of a sudden now, SASE comes around. And I think a lot of people wonder like, "Can they really compete with some of the pure plays in SASE?" Talk to us about that evolution. And maybe it's the functionality there. Are you winning head-to-head with the Zscalers of the world and vendors like that?

Yes, yes, exactly. Okay, so this is a new category that's kind of emerged in the last 5 years. And I kind of think of it Gen 1 zero trust or SASE players and then now the current market, and it's changed so much. Or said another way, 4 years ago, if you were looking for a zero trust provider, Zscaler was your option. That was really the only option, really viable. Today, there are many more players, including us. And so it's way more competitive today when you're going through these people. Like as a -- there's just more options than 4 years ago. And so it's interesting. It's -- sometimes it's good to be first to market. Sometimes it's not good to be first to market. And so we have a great offering in this part of the market. We are winning more than we were 3 years ago. And we will continue to win in the future. Because it comes back to the power of our network, the way that we're architected, where we can -- where we're already everywhere and we have this -- we use our own serverless infrastructure, where we run every service on every server. And so we have a great SASE offering today. And if there's some feature we don't have yet and the first generation player does have, it's like we'll put it on the road map, we'll ship it. And we have a good reputation for shipping things on time and which is different than 10 years ago. 10 years ago, when enterprises said -- when software companies said they were going to ship something, maybe it made it in -- they're only doing releases once a year or every 9 months. Our release cycle is daily. So it's just a very different sort of setup than even 10 years ago compared to like other companies that maybe you would think about before. And so we feel well positioned. So every day, we have many customers using the service. And we see ourselves as part of the consideration set for more and more opportunities, which is great. And more and more companies are getting -- are leaning into this because it is a growing trend in the future. The last -- the second thing I'll just say, which was not obvious to me, and so I don't think it will be obvious to this room, is there are some of the companies who pick the first-generation providers and they're 4 years in and they're unhappy. So they're going back out to say, "Okay, who's new?" And so we're now doing migrations from some of the first-generation cloud SASE providers to Cloudflare service. We have something called the Descaler Program. And so whether it's Palo Alto's Prisma or a Zscaler or a Netskope, we are migrating from those -- some of their unhappy customers to Cloudflare solution because it's been around for 4 years. And again, not everyone is happy with those services. And so we're seeing that. And so it's both net new businesses who are trying to figure out their zero trust, SASE journey plus the ones who already have, who are unhappy with who they picked at first, saying, "Hey, I want to see what else is in the market and the market has moved since I made a decision 4 years ago." So I think that's important framing for this room. And it will continue to move. This is a -- it's a journey. SASE, zero trust, you look at what that means, it means a lot of things. And again, I think when it comes back to like networking, security, access, a lot of the things made us really good around some of the web stuff, it's like, "Wow, let's port this to users and employees. That makes a lot of sense." And so we're excited about that.

Well, I think SASE and zero trust is all about connectivity, too. And so it double-clicks back on...

Yes, exactly, exactly, yes.

Makes a lot of sense. The other -- if we think about the whole product opportunity, one of the other big initiatives this year is go-to-market. You have a new CRO. And you've grown up a lot as a go-to-market motion from a lot of direct sales or over-the-web sales to now top-down. Talk about the importance and sort of the progress on building out these sort of seasoned enterprise reps. Because it's been a process, right? And I think we talked about it earlier in the year for you guys. But where are we at from a go-to-market perspective? How good do you feel about the positioning today?

The best part of my job is the people I get to work with. I feel like I work with so many talented people. They care. They set the details. They believe in what we're doing on the go-to-market side, on the technical side, on the G&A side. I mean, anyone at Cloudflare can work anywhere. They choose to work at our company because they really believe in what we're doing. And they feel like we're an important part of the next 10 years of the Internet. And that's just the next 10 years, it's going to be a lot longer than that. But that's -- like that we play this a vital role. And so on the go-to-market side, of course, we're constantly -- we're at an inflection point as a business, where we're going to do about, what did we guide to, $1.2 billion this year, I think. Yes, yes, yes, big one. Yes, okay. I want to make sure I get that right. That's an important detail, especially for this audience. And just we have multiple products. We're very global. We service multiple customer segments. Actually, all of the things that make our business much bigger and the opportunity bigger, like I love that. It's just very different than when you're running a $200 million business, selling kind of $150,000 contracts of the same thing to the U.S. It's just different. There's more takes and puts. And so we had to bring in a leadership that understands the scale, which has been great. The good news is we're able to get the talent to come in. Everyone is super excited. And then they're bringing -- transforming the teams, evolving the teams to make sure that we're set up for success in this next phase of growth. And I -- as a business owner, as an entrepreneur, I think sometimes the mental model you use about thinking -- building a company is like it's like this. And if you do a really good job, maybe it's more like that. But it's more like this, it's more like stairs. You kind of hit these inflection points, things that we're working, you're like, "Okay, I've got to change it up." And you go change it up and then you get going again. So it's like this, so the go-to-market, we were at a stair step. And I feel like what's been encouraging is we brought a new leader as we've been able to get really great talent on the enterprise side to join -- I think 0.5 million people applied to work at Cloudflare. It's like we're a B2B company. That's like a consumer service. Like it's crazy. We're not that big of a company and that demand for talent. And if you look at some of the new sales talent we've brought in compared to the previous cohort, we see them -- again, it's early, so I don't want to -- like don't model off of this because it's early. But it's -- you just see all the metrics of what you want. They're all improved. Let's say the better onboarding experience, they're able create pipeline, get in front of customers, what they're saying resonates. And so it's been positive, which is great, which gives us confidence to continue to invest, that we're doing the right thing. That's good. And if you go talk to customers, it's like they're happy. We're solving real problems for them. And so making sure that we can match the sales demand to have these conversations to the market is really important. And I feel like we've done a -- we've been able to make those changes while keeping the stability of our business, which is something that doesn't always happen, so we don't take it for granted. And so we're trying to be disciplined and thoughtful about it but feel pretty good about where we're at in that story arc.

That's great. I mean, you spoke to sort of you guys being a destination of talent from a workforce perspective. And that's always been the case. And I think this builds on a lot of what we talked about for the first hour here in terms of why it's so interesting. But maybe talk about the importance of D&I. Because I know that's a huge focus for you. And does that also help attract some of the best talent out there?

So all the data shows you that more diverse teams are a better place to work and drive better business results. And I think Cloudflare is one example of that. We've always been a really diverse team, both like gender -- I'm more of the founder and I'm a woman. You don't see that with very many infrastructure companies, all though Cisco also had a woman founder, which is kind of cool. The -- that even like the early people who worked at Cloudflare, at one point, we did the -- it's like the average engineer had moved like 1,500 miles to work for us. Like we pull talent -- technical talent from a lot of different nationalities and backgrounds in the early days of Cloudflare. Because from very early on, we were global, meaning that we have this free service from very early on. And if you looked at where our customers, where the traffic was coming from, it was like all over the world. It was not just U.S., it was like really all over the world. And so we had a bunch of engineers from South Africa come work for us and France and Switzerland and Asia. It was like awesome. Like I mean, all of the -- so we were kind of a melting pot early on. And so when we say DEI, of course, you think about gender and you think about underrepresented minorities, and we care a lot about that. But this also just means diverse background. We had a team happy hour last night. We had a bunch of people in the area. And we hosted a happy hour last night. And there was a newer member of our sales organization. And she said to me, "Oh, my god, I've never worked in such a diverse team before." And I was thinking -- she was a woman, so I was thinking like, "Women," I was like, "Oh, awesome, tell me more about that." She's like, "Like for example, some of my team has kids, I've never had that before." And so I think like diversity means a lot of different things, where it's not all looking the exact same is maybe the way. And so I do care a lot about it. I think having more women in the tech industry, especially in deep tech, what we do is pretty geeky. Like we make a lot of DNS jokes at Cloudflare. But like I love that we have a lot of women. I love that we have underrepresented minorities. We were at AFROTECH in Texas, African-American -- AFROTECH to have more Black people join. We need more underrepresented minorities and women in the tech industry. Because it's the future. These are great jobs. And we have to make sure we're building products that have that point of view in, so we don't miss the mark and insulated. And you want them to participate in the economic upside. And so it's something I do care a lot about both at Cloudflare -- and if you just look at our numbers, about 34% of our team are women, which is really good for a B2B software company. It's very good. Is it 50-50? No, but it's very good, like we are like market-leading. I mean, about 16% of our team are underrepresented minority. And if you just look at positions of leadership, 33% of our leaders are women, which is again great. And people when they see themselves, they can come. And so I feel like we're playing a role. And I went to see a customer, it was a European customer and it was the CTO. And he kind of -- we had a team from Cloudflare, which we had a couple of other women and men in the room. And he came up to me and he said, "Can you help me be more diverse, please? Like how do you always have women work for your company? I need some in my team." Because this is -- I know from another CTO, he's like, "I don't want work to feel like I'm working in a submarine. Like a bunch of dudes in a submarine. Like I want to have some diversity." And so I think it's -- we have a long way to go. But I feel like we're helping [ DEI ] place where that's possible. And so -- and again, I think that it's great. We're not the only company. But I love that there are companies who are paying attention to it. And I would say to people, vote with your talent, go work somewhere, where that's important if it's important to you because vote with your talent, that's what you can do.

I think you said historically, they're not -- this is not a nice-to-have, it's a must-to-have and...

I think if you don't do it, you're going to get left on the wayside. Again, right now, it probably doesn't matter, but eventually, it will. Because the talent will go to the places where they take it seriously. And it's one of those things where it's a little bit like I said before, you do these things and it's not like you're getting credit for it at first. That's not why you're doing it. You're not doing it to get credit. You're doing it because it's the right thing. It's a good strategy, a better place to work, better business results. That's why we do it, okay? And it's the right thing. But it's like really there's a good business reason for this. But I do think the companies who start to do that will end up paying dividends in the future. Just kind of like entrepreneurship, it's like things aren't big until they become huge successes overnight.

Yes. You guys have -- from a product development perspective, you've been so innovative. I think you're not afraid to fail on products, too, right? I mean, it's, I think, been a part of your DNA to be so innovative. When you think about M&A in the future, does it -- do you think because of the scale that you've got and sort of the global nature and sort of the talent that you're attracting, does M&A ever become part of the story beyond the smaller tuck-in-type deals?

We are so innovative partly because our opportunity is so big. Back to like do opportunities narrow over time or expand because of what we do, I think it expands. So we're ambitious like, "Let's go and get at it." So that's one. And two, we've lowered the cost innovation internally. We've made it easy for our teams to innovate and ship things. And companies who do that are well positioned versus the ones who don't. If you make it really expensive to innovate, then you've got to be really careful before you make sure that you're not failing, so you don't waste all these resources. But for us, we're like, "Okay, we've lowered cost. If it doesn't work out, no big deal, a couple of engineers, a few weeks, not a big deal but no more." And so I think like both those have been helpful. We have that muscle built. Some businesses don't have that muscle. Because they had a big winner, and they don't know how to build anything new. And that's a hard place. You've got to start to build up the muscle. Just like going to the gym, the first time you go, it's hard. But if you keep going, all of a sudden, doing those lunges or squats become a lot easier. And so it's the same in tech development. So that's on the innovation side. What was the...

Just M&A.

Right. I was like you asked me a question about M&A. So the thing about M&A is again, it's hard to predict the future. And I've talked to some folks upfront. I saw a chart. 1,000 private companies are listed as unicorns. That's too many. 1,000 is too many. I literally wrote that paper on unicorns. It's about 10 years since I wrote that out. I think it was 2013 that it came out. I remember reading about it. We were a unicorn. At the time, it was 8 to 12 a year, and we were one of them. And aligned that, we were like us and Dropbox and couple of others as kind of part of an era. So now to think there are 1,000 private unicorns -- private companies that are listed as unicorns, like the valuations are too high. I don't need to say that to many of you, but that's the case. And so what's interesting is the funding to do extra funding is not there. Like the venture capitalists are like, "Really?" My sense is most things aren't getting funded and aren't getting follow-ons that they need. And it's been a long time. And it doesn't feel like it's going to be the fastest out of this tough macro environment. Some of them have taken venture debt. That's going to come due. So the question is will there be interesting opportunities over the next year? We think there's going to be. We think it's going to be a little bit these high -- like some of these things are going to become opportunistically, where we're paying attention to say, "If there's something interesting, we should act." And so again, it feels like valuation has been high for a long time. Is this finally the year that they come back to reality? I said [indiscernible] don't really like reality. But it might be back to reality time and kind of hard, hard knock. And if so, like that's an opportunity for those who are well positioned for the future, which again I think us are. There's others, it's not just us. But it's like who are the companies that are the monster in the making over the next 10 years? Again, I think we have the chance to be one of those. And so looking into saying, is there something opportunistic? Because of this, what we believe, is going to be a shift, different than 2 years ago, where everything was just so expensive, you're like, "That doesn't make sense to get a deal done." This might be the time.

Yes, interesting perspective. Maybe -- we've got about a minute or 2 left here. Aside from generative AI, which again we didn't talk about this last year, I mean, if we're sitting here next year or 2 years from now, what are we going to be talking about that's like -- that like, "Wow, I can't believe we didn't talk about that 2 years ago or a year ago"?

Okay. What we're going to talk about is networking, modernization of the networks, which is really sexy. That's so important, networks have been around for a long time. And again, I spend a lot of time in the market. And just even like today versus a couple of years ago, as you've moved your workloads to more places, as you've moved your applications to more places, networks, which are a big part of [ making sure it works ], are going to move to a service model. They're going to modernize. They'll have it modernized. They're going to. So what does that mean? Anytime you hear the word MPLS, SD-WAN, I think those are not modern. That's all going to get modernized in our opinion. So how do you take both corporate networks of like office buildings and data centers and warehouses and manufacturing plants connected to their cloud and their employees? All of that, you need a global network that can move a lot of bits of traffic faster, safer, more privately, more agilely than anyone else. And that's what Cloudflare's connectivity cloud does. And so it's -- those are big transformation, big modernizations. The market wasn't really ready a few years ago. But if you start to talk to these networking teams, and they're just -- the warehouses are falling over the manufacturing, how they have it designed isn't working for this modern way, the next wave of the Internet for how they're impacting businesses or evolving businesses. And so we're going to be talking a lot about that. And people are going to not realize how important this is, but...

[indiscernible] sometime to that. But great, we've covered a lot of ground. That was super insightful. Michelle, thank you again. So from all of us at RBC, Phil, thanks again, and best of luck.

Yes, thank you so much for having us. Thank you.