Fastly, Inc. (FSLY) Earnings Call Transcript & Summary

Good afternoon, everyone, here on the East Coast and still good morning on the West Coast. It's Tim Horan, the cloud and communications analyst here at Oppenheimer. We have Fastly up here now. Both the CFO and the CEO will be joining in another minute, but we have Adriel on, and Josh will be joining in a couple of minutes here.

And it's great that we have Adriel on to start, because my first question is basically just summarizing the quarter from your perspective, Adriel, and a little bit maybe on the guidance. But from our perspective, frankly, it was one of the strongest quarters we've ever seen with 62% revenue growth in our coverage universe. But just love to hear your perspective on that. And we'll -- also maybe just a lot of the questions you'll be getting exiting the quarter. Thanks, Adriel.

Sure. Thanks, Tim, and thanks for having us. Yes. I mean I think one of the key things, just to sort of state off the bat, is you're sort of really seeing sort of the attributes, in this case, positively of a usage-based business. When you see customers begin to adopt and use your platform in a usage-based business, you see it in real time, which is why when we came into the quarter, call it, back in May or Q2, we were able to put a guidance number out there that I think was likely significantly better than what most folks were expecting for Q2 for Fastly this year. We were -- at the beginning of the year, we had sort of forecasted, for the year 2020, we were going to grow a little bit better than 30%. Just come off of a year, we have grown 38%. And then we just posted a quarter here where we grew 62%. So clearly, there's been an acceleration as a result of the -- as a result of COVID-19. And we were kind of out front and center with respect to kind of how we thought this was impacting us. Artur Bergman, our Founder and Chief Architect, had noted through his research that school closures was a big indicator of sort of driving Internet traffic. We've given an indication of kind of how that was playing out through what we had seen so far at the end of March and then in April. And then you really saw that sort of flow through Satya Nadella who's head of Microsoft and part of maybe some of the questions we've been getting lately has indicated that you saw 2 years' worth of digital transformation occur in about 2 months. For us, that really reflected itself in Q2. And you saw not only from a revenue standpoint and from a growth standpoint across many of our customers, our dollar-based net expansion rate increased 137%. We increased our average spend with our enterprise customers up to $715,000. And then you actually saw a gross margin leverage to the tune of about 600 basis points increase. And then actually our first not only EBITDA, but non-GAAP operating profit quarter. So there is a lot of leverage that I think was inherent in the business that we were able to sort of see come to fruition sort of a bit of a preview just in Q2. And then as we got out of Q2 and back again to May and sort of we've gotten -- as we sort of expected, things were beginning to emulate as different local governments and national governments would begin to implement their own relaxation to shelter-in-place or school closures. And in this case, different municipalities were thinking about how to open schools safely. So we began to see a bit of that into Q3. And normally, you think of Q3 as flat uptick to Q2. Clearly, this year, from a comparison standpoint, it's a little bit different. But we -- but Joshua and I and the rest of the team have talked about, I think by the time you get to Q4, we're hoping to get to sort of normalized seasonal year for Fastly, which is normally our third quarter, is relatively our strongest quarter. And this year in particular, you're going to have sort of increased velocity on the Internet from an election year. So hope that I gave some perspective and sort of our expectations going into the quarter, what we experienced in the quarter, some of the financial applications for that relative to Q2 and sort of what our expectations are for the rest of the year.

And I think we have Joshua on now also.

Apologies for the delay.

No problem at all. I love your backdrop.

I was in the middle of adjusting it, because I was just on a meeting that -- it was a Google meeting. You can put backgrounds in Google meetings. So you learn these things -- new things every day.

Your background looks great. Maybe you're -- Adriel, you might have missed the beginning, went through some of the financials on the quarter. Maybe your perspective on the quarter and the guide. And I would also like to know if you guys can give a little bit more color on what's going on with the volume growth going into the quarter and since leaving the quarter. And maybe a little bit more about how do we get the really nice uplift in the gross margin.

Sure. Let me start at a high level. Obviously, when you're putting 62% year-over-year growth numbers out, the largest number of new customer adds, nice leverage on the gross margin line, incredible leverage all the way down to the bottom line, that's a good quarter. And I'm incredibly proud of the team. So I look at that and I think if we -- like a year ago or 1.5 years ago when we IPO-ed, if we could have imagined quarters like this, we would have been really satisfied, and I think investors would be really satisfied. Now what's exciting about this is we think there's -- it's sustainable, right? I mean we think -- when you look at the guidance that we gave through Q3 and Q4, we were seeing still really strong growth rates. And as the world gets back to normal -- now we don't know what the world is going to look like. Adriel touched on this a little bit. But our assumption is we're not going to go back into great economic lockdown for the next 1.5 years. There are going to be flare-ups here and there, but we think economies are going to go back to working just to an extent in different forms. And so the new customer count is really important in all of this because what we've shown and what we've seen, to your point about the future, is we've seen an acceleration in customers that are looking to digitally transform. And these are customers that are either digital natives and are growing faster than they ever thought was possible. Or they are customers aren't digital natives and know they have to become an engineering culture that drives innovation and iteration. And so we're seeing both in the funnel. We're seeing the sort of early adopters that we've always seen. And what's really exciting is that it's complemented by a whole bunch of companies that have made the realization that digital transformation isn't just a document and imagine that someone could write for you, you actually have to become -- to survive an engineering culture. You have to build differentiation and not buy it. And you need to empower the new decision-makers who are developers. If you look at the usage-based developer-focused models out there and you look at the quarters that all of us are putting up from Twilio, Datadog and others, I think they are showing growth. They are showing various growth metrics, but they're still showing growth. Expectations aside, obviously, and what's built into those stocks. But -- so I think that -- the summer is always slightly slower traditionally from a demand perspective. People are out on vacation. There are certain regions like Europe where August is really, really difficult. But we continue to be excited about what we're seeing in the demand funnel. And we're also innovating in that funnel as well. I mean, you don't have events and you don't have face-to-face dinners, you've got to innovate. And we're proud of how we're innovating to capture the mindshare of that customer. So I will say positive. Summer is summer. And it's a strange summer for all of us, but it definitely has some impact for sure.

And I check your volumes daily at different times in a day. And we -- it does seem like things are flat-lined a little bit off the peak in March. But it was unprecedented, absolutely, unprecedented times. And I mean I -- obviously, we still have a decade ahead of us, so -- of really, really strong growth. But it does seem like you're capturing the who's who of new startups. We'll get to TikTok in a second. But they're a classy example. But to your point, companies that are engineering-driven and that are real innovators are coming to you. Are they finding you or are you finding them? Like what's the sales process like?

So the first 8 or 9 years of our business, they all found us. We really didn't market in a traditional way. We didn't have demand then. We didn't go out to the market and seek customers. So that is a relatively new muscle for us. It's really improved over the last 12 to 18 months. But it's still a new muscle. So still a lot of organizations find us. We are changing that. The stock price run-up has been good for brand in a sense as well, because people now know who have a slightly better idea of who we are. But Fastly isn't a household name, and we stand behind the best of the Internet. We're not the ones out there sort of putting our hands up and saying, listen, we're changing the world. Our customers are changing the world, and we are happy to facilitate that through our product. And we really want to put our customers front and center in that story. But to your point, we -- if you're an innovator, and you just look at the fastest growing and the best technology, even public companies, we are very proud of having almost all of them. And certainly, the most innovative ones is customers and partners. And we see a couple of different phenomena. One phenomenon that we've seen, which is really interesting, it was articulated by Nick Rockwell, who was the Head of The New York Times, is he brought Fastly in, in order to be a part of a change of digital transformation, right? So it wasn't about I'm digitally transforming, and I'm now at this point where I can bring Fastly. It was like, I think if we bring Fastly, and it will facilitate digital transformation. If you look at Gannett, the USA TODAY -- runs USA TODAY and many other newspapers, they digitally transformed. And when they started to run up against all of the barriers that their provider put in front of them, they brought us in. So we're being brought in at the start of digital transformation, we're being brought in the middle and the end, depending on the organization. And we're seeing use cases across all of them. But to your point, if you want to deliver a scalable, secure and perform a workload on the Internet, be it an API, a website, I think we offer a very compelling story currently.

And so how do you compare to the competition? I mean are you like an order of magnitude better? Like is it just a totally different product, because companies can come in and use your platform to digitally transform and no one else -- there's no other platform out there to really do this?

Yes. So I think what gets confused in this business is there are a variety of different markets, and the markets are actually different. So we've got a commoditized video business. In that market, it's highly competitive. It's really about price. It's difficult for you to monetize a video website, you care about how much it costs to deliver. That is a highly competitive market. There are the likes of Limelight in that market and EdgeCast, which is bought by Verizon. Akamai plays there, CenturyLink, Level 3. Like that's a highly commoditized market. If you look then at every other enterprise market, we're talking about the largest in the world, they're effectively dominated by Akamai, and you start seeing us having single-digit penetration. It's really a 2-horse race at this point. There is no one else at the enterprise. And so you -- in order to really understand that question, you have to really understand that there are 2 very distinct markets. One is highly competitive, one isn't. And then in the highly -- in the nonhighly competitive market, us versus Akamai, it's simply a story of customizable, programmable, not customizable, not programmable. If you think of the cloud experience that a developer expects, you go on to a website, you sign up. In a second, you can start using the product, right? You start using the product. You start integrating. All the documentation is live. You don't need an army of professional services people that are offshore. I mean every experience is different. So in many ways, I think the way to think about this is like cloud and anti-cloud. And anti-cloud is not a great place to be right now with anti-software, and I think that speaks to our growth.

So you added, to your point, 114 customers in the quarter. Last year, second quarter, you only added 6, but you were kind of averaging, like, 50 per quarter. So you've doubled your kind of run rate. Are you at a point now where customers can come in and just like basically almost self-provision and use a credit card to sign up and then start using your platform like right off the bat? And the reason I ask that, I mean, can this be the start of a trend where you can really go out there? And it seems like you'll need another new branch of the go-to-market strategy. But is that going to be a focus to really start to get this viral?

Yes. So the answer is we -- you've always been able to go online. Anyone could right now start a Fastly account, use a credit card, get a developer account for free. I mean that's part of who we are in our DNA. The approach for us on the small and medium business market, and particularly the small market, is a little bit different, which is we believe your entry point in that is actually one of our partners. So we really firmly believe you're going to start an e-commerce shop like Adobe Magento, Shopify, go to Wix, go to one of these. If you're a developer and you want a page, go to get HubPages. So instead of building a muscle to go and monetize these millions and tens of millions, hundreds of millions of un-unique domains, we've taken the approach that we think the future is in platform. And so that's our approach to that market. So when you look at the customers that we sign and bring on ourselves, those are often the larger sorts rather than the smaller ones. Because if you're a $20, $50 a month customer, we think you're better served through one of our partners. So we do have this dual strategy. I think the answer to your question is we believe in that virality of the business, and we will continue to invest to have developers start with us, use us, play with us. And that translates really well into the enterprise business. But we do have a strategy of a small and medium size, which we think is really powerful.

So in that SMB market, do you know what percentage of your revenue comes through channel in total or maybe SMB specifically?

We don't disclose that. I don't think we have that, Adriel. I'm not sure that we've said those numbers. No. We haven't talked about either. I think these partnerships are really important for us, because we do believe the SMB needs to be served. We think that they need security, performance and scalability, just like everyone else, and we want to be in a position to do that. We firmly believe that they also want the simplicity of what partners can bring. And one of the really nice mechanisms that we're seeing is as a customer grows on one of these platforms ought to now graduate to using that platform and us in a direct relationship over time. So actually a really nice sort of feeder system for people who want slightly more control than you might get through one of these partnerships. So we do see a nice flow through there, and that's something we're just trying to spend more time on.

And do you think you're the primary partner for someone like a GitHub and a bunch of the -- and Wix and a bunch of the other companies you mentioned?

Yes, sir. I mean that's -- I think that they see the performance and the scale and the security and our ability to manage the scale. I mean scale is really important. These companies are huge organizations, and they're growing extremely quickly. They need enterprise-grade security and scale. They can't have large outages. They can't afford to have huge security breaches. Like that's not in the game for them. So yes, we certainly find that those that are sophisticated does.

And maybe just stepping back, it was a great example with The New York Times, how they kind of use your platform to digitally transform. Were they successful in that? And maybe how did you really help them to transform? It sounds good on paper, but practically, how does that work?

Well, I think when you see the future, and they did, they saw that their business was advertisement-driven exclusively and they saw a future where that advertising would be a significantly smaller percentage of revenue and digital subscriptions would be the way that they would create value. So for you and I to go spend, I don't know what it is, $20 a month, whatever it is, you have to get value, you have to get consistent value. And it's not just about reading that one article. It's about consistently seeing value, because you're going to consistently see that charge on your credit card. And maybe it's $20, because I'm in Canada, it's more expensive. But the -- and so I think you have to look at digital transformation through the objective goals. And the objective goals for that organization were I need digital subscribers to come and stay. What was really exciting about their last earnings call or their last earnings was they just announced that digital subscriptions have overtaken advertising as a -- as their -- in terms of revenue. And so when you think about 5 years ago, digital subscriptions -- or 8 years ago, near 0, you now have more of your business in this really solid -- I mean we have very strong relationships with The New York Times. They can't imagine canceling my subscription. I think that's the case for a lot of people who source the news. That's a complete transformation of the business, right? In one case, you're serving advertisers, which you still do. In another case, you're serving users except very different perspective. So I think if you just look at the end goal, I would say it's transformatively successful. And they're one of the examples that I think every media and reporting organization looks up to and says, "I want to be that," because everyone else's advertising revenue is also dramatically shifting and changing in a way they can't control. And so I would say, absolutely. And if you look at the culture that was built there, we certainly aren't -- the culture was built by the people. We were a tool in that process. We're proud to be part of it, but by no means can be claim victory. That was just tremendously hard work by that group and incredibly visionary work to know that your underlying business that's been around for hundreds of years is about to completely change, right? I mean that's a huge change.

And that's happening, obviously, to every industry. That industry got hit hard early. And they've done a phenomenal job of managing that transition. Are there other companies or industries that you can point to that are just sort of realizing right now, we have to do this transformation, and you're starting to help them?

Yes. I mean I think if you're a restaurant, if you're -- if you have a place in a mall, if you're a chain store, we're seeing those organizations wonder if they will exist. Travel is another area that is obviously being hard hit. I think what's really interesting about travel is we're starting to see these sort of green shoots of growth coming out of those groups. So really innovative ideas. I was talking to a CTO who was saying in the March time frame, carts were empty. April time frame, cards were empty. In May, people started to put like that anchor hotel into their cart. They didn't fill it up with the rented car and the flight. And they're starting to see in June and July, those trips come together, right? This is like a December trip to the Caribbean, December trip to Mexico or a ski trip to Europe. And so I think we're starting to see that dreaming aspect in that business. And we are seeing a huge push for digital and innovation, how can we make this easier? How can we give people this peace of mind? So I think we're seeing a number of industries, even financial services. I mean what happens when you close every branch bank that you have. Does this huge footprint matter? Is that really your biggest asset? And so yes, we are seeing across the board in payments, in financial services, in traditional, retail -- Adriel talked about Satya talking about 2 years of digital transformation acceleration. There's a great survey of CIOs done by Twilio where they got 6 years. And so you could imagine thinking about how fast that's changing. What's really exciting and humbling about our position is we get to witness the change. I mean we witness it before the world sees it. We see the people that are saying, I need to change, because they come to us. It's a little bit of a crystal ball. It makes our jobs really exciting.

So I did want to just touch on TikTok, and I could talk to you for hours, Joshua, but I know you were -- I think you were on TV this morning or last night, I just read some comments this morning. Any more updates or color? I know you gave a bunch of metrics on the call.

Yes. I mean I think the thing that has not been seen perhaps as well as it should is just how innovative and how fast these customers grow. And so when you actually look -- Bill Gates was on -- and I love how he thinks of competition, especially given his history, where he says, the fortresses of tech look really thick the walls until they don't anymore. And so you think about this entire generation that's grown up with their parents on Facebook like my kids. And I can tell you one thing my kids don't want is to be on a social media platform with me. I can tell you that for sure. And so when you look at my son, 14-year-old son, shows me more TikToks than anything else, frankly, so as my wife. And it's because it's innovative, right? It's an interesting form. The content is great. And so I think people are losing sight perhaps. So one of the reasons this is so -- such an interesting dynamic is because hundreds of millions of people over the course of 6 to 9 months have decided that this is a way that they want to share and work with information. This is a classic story of digital innovation. And so I think the first part of this is digital innovators work with Fastly. So we're humbled to be in a position where those who need to scale and grow work with us. That's one. I think the other point that we made really clearly is we can't control what the outcome is and we obviously will comply with any government requirement on us. I think one of the things that we see is opportunities to either, as we talked about in the call, backfill that traffic because we have a lot of customers that have pent-up demand that were willing to provide content of this type of traffic to. Now we have a lot of network capacity. This isn't a capacity issue. We want to mix our gross margin to continue to get leverage. And so we need some video business. We need some API business. We need some performance business. And when some of that traffic declines, we often have ways to go and -- go to customers, particularly on the video side to get more, because we often cap those customers in order to get the right mix. I think the other side of this is we have great relationships with a lot of the majors that would be in the running for something of this asset of this size. So we have a couple of experiences already with Microsoft, LinkedIn and GitHub. If you think of a GitHub experience, huge customer of ours since 2013, in 2018 obviously acquired, and our relationship is larger and more innovative and we're doing more now than we were in 2018. So we are seeing that the CEOs of these companies, and Satya, I think, is a great example, has seen that it is really not in their interest to take this very gust of growth and innovation sort of force feed it old tech. That doesn't work very well. And so I think we've seen examples of this already. So whether -- no matter what direction this goes in, I think we feel that we are in a position of strength. I would say we weren't forced to disclose this. We chose to do this voluntarily. And we did it because we think it's information that we want to be really transparent with our investors and make sure everyone knows exactly where we're at. So I do think this is something I'm happy that we talked about publicly. And I think the goodwill that we have with that customer, along with all of our customers, when you look at our net retention rate and our churn rate and our Net Promoter Score, will pay off by just being direct and upfront about the situation.

So I have a longer-term question. I think it's the key to the company and the stock, because I am getting some questions online here. And you're answering a lot of the questions, by the way. Is the sell-off in the stock warranted? And that's not my question. But my question really is, your stock is going to look really cheap if you become the platform company for edge, right, where we're sitting right now. The question is, how do you execute on that, right? This is the goal, to be the platform company. Are there other platform companies that you've seen have been successful, you're trying to emulate what they did things right? And what do you need to do to become this platform company?

Yes. So we look at it through the same lens. We have 300 enterprise customers. There are a couple of hundred thousand out there, and we feel like we're just at the start of this journey. And what we have seen is that the innovators have a new way to deliver applications, and we are at the heart of it. So we have the same optimism. I think your question of how you execute is the barrier between -- we are in this unique position that the barrier is not a product market fit. It's not a question of is our product better, is it faster, is it more secure. I mean those questions are being answered everyday by the brands that use us and whose choose to reject the old technology. The real question is how do we get out there, how do we market and penetrate into these organizations. And I think we saw a lot of really nice signs in the last 2 quarters, which is an uptick in number of customers, people are coming to use. They're coming to us. So I think the answer is more of the same, how can we just keep having people come. The reality of the technology with the 2 major investments that we're making just around security and around Compute@Edge, that is where the future is. And we are innovators in both areas and we'll continue to invest in that. So it's a bit of a unique position to be in the whole sea disrupting not just one multibillion dollar market, right? This isn't just about content delivery. In fact, it's sort of less about content delivery than it is about content security, than it is about programmability, than it is -- you imagine this data centers disaggregate. You have 5 appliances on routers that are sitting in that data center and load balancers. Where is all that going? And how are you going to manage it in a completely new architecture? Billions and billions of dollars are being thrown up in the air for the innovators to grab right now. It's almost a plethora of riches. And I think part of it is we have to be extremely disciplined to invest in a disciplined way. I mean, I have never been in strategic meetings where I have more multibillion markets to disrupt. It's a unique place to be, right? When you think of load balancers, think of the DDoS, think of the security appliance market, think of the firewall market, think of the edge security around WAF, think about delivery, video delivery, programmability. One of the things that customers said to us, which I really like this way of thinking, was for every $100 they're going to invest in the cloud, to the central cloud, Amazon, Google, Microsoft, they're going to invest $5 to $10 of that in security and $5 to $10 of that in edge. So you add that up, if this opportunity is $20 for every -- high $20 for every $100 this customer is going to spend in the central cloud, this is a big market now and a really big market in future.

And do you think, ultimately, all the security that you're talking about or an IT organization's entire security can be handled in the cloud away from on-prem?

Yes. We're seeing it today. So yes, I mean, the answer is the most innovative organizations that we work with, they have no on-prem anything, full stock. It just doesn't exist. And so we're already seeing that. Now I'd make the differentiation. There are 2 security markets here. There's an IT security market, which is protect zero trust environments, a Zscaler type market, what CrowdStrike does, even what Okta and Duo do. Like that's about securing one side of the business. When we talk about security, we're talking about the APIs, the web applications for the organization. So I want to acknowledge there are 2 security markets. We play in one and not the other today. It's very interesting to speculate about what we could do protecting the enterprise, but -- the enterprise assets, but that's not a market that we're in today.

So I mean, ultimately, other platform companies, it seems like, in your position, kind of get the developers onboard, they learn the platform and those -- it starts to go viral. Where -- what do you have to do to kind of get that virility going in terms of really getting more and more developers and more organizations onboard, so you're not just at 300 enterprise customers, you're at 3,000? And then you have those people almost becoming a sales channel for you.

Yes. I mean we have to just keep doing what we're doing. We have to keep marketing. We have to keep selling. We have to keep innovating. I mean 62% growth speaks to a pretty high growth rate. It will be interesting to see when all the numbers are printed, where we stack up. I think we're going to stack up in the top group. And so you also have to ask yourself how fast you want to grow, how fast can you sustain the kind of healthy growth that we need. We think we're growing really, really quickly. So I'd argue that we're there. It's happening. It's in the middle of it. It's going to take time. I mean these are enterprise sales cycles. One of the things that people don't fully appreciate about our business is we are selling against a cloud company, right? So if you think about what a Twilio does, they go and replace hardware, right? That's -- in many cases -- or they think about an innovative use case that when you already have a cloud, Akamai is the first cloud. I mean Tom was talking about this in his earnings call, and it's true. They pioneered this idea of a multi-tenanted cloud. And the opportunity could have been there for them to compete with the likes of Amazon or even us in the modern application. They haven't, but that's -- but unlike -- we are like Twilio's next competitor. We're selling against the cloud. They're not selling against the cloud. So that's the other element here. So when you have to replace your entire data center and all of your hardware is going away, Akamai moves with you. I mean you don't have to move Akamai. It's a cloud. And so I think that, that's misunderstood, and that creates a slightly different sales cycle and a slightly different enterprise market for us, because we're not replacing a piece of hardware.

Great color. And is there any particular company out there you're trying to emulate that's kind of gone through this process, like Salesforce.com or anyone else, GitHub, LinkedIn or...

I mean we really admire -- we certainly -- the usage-based business is an important part of our model, right, which is where -- if you think -- if you sort of fast -- if you sort of rewind, you remember the Salesforce model of, hey, Cisco or Oracle or Civo, whatever, has to earn your business once every 5 years, because you're sort of POC-committed, right, by buying into Civo. Guess what, folks, buying to Salesforce once a year and our earlier business the next year, right? We heard this is a big part of like the no software knowhow, the no hardware, no software like the SaaS model. If you think about what's happened in the usage-based models, we actually earn our customers' business every day, because our customers only have committed to us about 50% of the revenue. So if you think you're going to do $20,000, maybe you commit $10,000 to us. So we have to earn that business. They can go somewhere else. And so I think when you look at the emulation, yes, of course, Salesforce and these other companies, but they were in a "I sell to you once a year" license business. I think when I look at the usage-based models, they're quite innovative. I mean, obviously, Amazon has a really strong usage-based model. Twilio has a really strong usage-based model. Both have developers buying into that platform. Those are the ones that we'd look up to and say, wow, they've really figured this out at scale. And we admire what they're doing. Google would be in that category as well around the cloud. And Microsoft, I mean, Microsoft with Azure is doing the same thing. They're capturing this mind share, and they're growing at breakneck paces on large numbers, and our numbers are getting larger.

By the way, do you get customers that come in from Azure and AWS?

We do. Yes. We've got -- many customers will come through a variety of channels that we have out there through direct or indirect relationships, depends on the situation. But yes, absolutely.

Yes. Sure, sure. [Operator Instructions] I can keep going on for hours and hours. The -- I'm hearing we're going to have -- and everyone is basically talking about this. We're going to have one of the strongest e-commerce quarters ever here in the fourth quarter. Can you talk about how important e-commerce is to your business? Any kind of metrics would be great.

Yes. We haven't shared metrics about where they are as a percentage of revenue. I think when you look at the website and you look at some of the rumors that are out there, you can see that we are a strong player in that market. The most innovative companies use us. And we are also anticipating a very strong quarter. We have seen, over the last 4 years, the spike up every year in what e-commerce is doing for -- as a percentage of holiday sales. I think it's hard to argue that this year, we won't see something that is significantly larger. And we thought about that when we're giving guidance.

Great. So I got a great question here. Can you talk about just new product introductions coming up, especially on edge compute? Yes. What should we be expecting that you think will be very important over the next 6 months?

Yes. So I'd point to 2 areas. One is edge compute, right? And we -- what you're going to see is us moving into GA phases of that in 2021 or getting out of that beta phase. The beta phase has been really important for us, and we cautioned the market this was going to be a slower process. When you're working at the scale that we work with the security requirements that you work, we don't do what some other vendors do, which is just rush products out to get a marketing headline. I mean that's not in our interest. We don't have a lot of chances to screw up the largest web fix in the world. And so we are very cautious, and I would say, thorough because that's the expectation of our customers. So what you're going to see is enhancements in the number of languages we support. We already talked about some of the work around logging and visibility. But this is -- our world is not about just a walled garden. This isn't like AOL, right? We need to integrate with all of the systems that are out there, the analytics systems, the build systems. If now we're writing code, it is a system that we don't traditionally -- you don't traditionally write code, you need to get APIs and connections to those things. So a ton of ecosystem work. A lot of that actually isn't obvious to people, except for the developers who work in the code and expect it to work. So a lot of behind the scenes, it -- Compute@Edge is like an iceberg where most of what you don't see is below the surface. And so a lot of below-the-surface work. I think where you will see above-the-surface work is the other iceberg of security. We today are not known as a security company. I think that's a misnomer, because it's built into everything we do. And it's very rare, if not ever, that a customer will come to us and say, I want you to deliver my application and not secure it. So there's sort of a number of tenets of security in terms of what we deliver, TLS, DDoS, WAF, bot. And I think you're going to continue to see a lot of investment on the WAF and the bot side of our business. We deeply believe -- we're hearing from customers more and more, particularly during the pandemic, about how important security is to them. This is not a new trend, but it's accelerating, and it's actually accelerating at the same pace as digital transformation, because it's forcing people to completely rethink their environments. And as part of that, it forces you to rethink security. The same way as I was describing, like the network -- we're like the next competitor of Twilio. Well, security isn't that, because you're getting rid of every appliance in your data center, you don't have a need today. Let's go. I used to have this. I don't anymore. So you're going to see continued investments, significant investments on that side, because that's a very much near-term opportunity that we are taking advantage of and can do more.

So Joshua, I have 5 questions here in 2 minutes. So I'll be quick. Do -- does Azure and AWS see you as a competitor or not or a partner?

I mean we see them as a partner. I can't speak to how they see us. But based on how they work with us, I think they would say partner, but you'd have to ask them.

Do you track -- or do you disclose much the number of developers working with you guys? And what feedback do you get from them?

We don't disclose the number. We do talk about the feedback, and we've talked about that in earnings calls around Compute@Edge and other areas.

Sure. Do you see your edge product being able to kind of be used with autonomous driving?

We need more time to talk about that. I think there are use cases, whether that's the case, but not the way people think. Cars are going to be like data centers on wheels. If you imagine when your call gets dropped driving into your house, you can't have the security of your car be sort of completely based on how good your cell phone connection is. So the answer is, yes, but for more background update and other types of tasks. None of us want our cars to hit somebody, because we got into a bad cell reception range. And we won't solve the cell reception story for 10 more years.

I think you answered this, but what do you think your wallet share is of your enterprise customers now for the products you sell?

It varies by industry. In some of our industries, it's sub-single digits. And in some of them, it's 100%. And so you imagine that, for example, in e-commerce situation, a lot of customers will not lock 3 vendors. In a video situation when you're doing the English Premier League or the Super Bowl, you do want 3 to 4 vendors. So it very much depends on the industry, and it varies from 100 to -- we have $100 million to $200 million wallet shares where we're 2% to 3%, a lot of growth in certain numbers.

And last question. I've been a proud owner of a Tesla for a long time, and I think Elon Musk has designed that car obviously for electrical vehicles, but also for ultimately autonomous driving. And I think he's light years ahead of everyone else. So the question I'm getting is, have you designed the company from the ground up for edge cloud that you can do security on this and you can do compute at the edge? I mean have you thought through this process the last decade?

Yes. It's why we built the company. And we built the company to be built by developers, for developers. And inherent in that statement is just a completely different way to build a business. So when you look at our metrics, the metrics are -- because we're doing it better than everyone else, the way they do, we're doing everything differently. We don't have any really expensive routers in our data centers or load bouncers. We've written software every single part of the stack. We have completely changed how we go -- approach our customers and how we work with them. If you're a customer of ours, use Slack, you get an answer right away. That's a special relationship from someone who knows what they're talking about. I was just on a call with my bank. There's nothing more frustrating than how customer support is done across the world. I mean it's just -- I can't even swear anyone, because I can't talk to someone.

Let's not start there. All right. I have one last, last question, which is a good one. Cloudflare is claiming 0 nanoseconds on edge compute on latency. Do you compete with them on this? Or does it hurt your value proposition? Are they better than you, I guess, that's kind of the question.

Yes. I mean there are really 2 spaces here. There's the SMB space, and there's the enterprise space. And because we are in the enterprise and they're in the SMB space, we can take a different approach to market, right? For us, we can't sort of pre-set up all of these -- 100 of these things, because we may have a customer that needs 1 million tomorrow in its packets. And so we actually have to work on technology that scales in real time. When your scale isn't as large, customers aren't as large, you actually can use other means to do that. So I would say it's sort of comparing apples to oranges in the sense that we -- in our Compute@Edge, we don't limit how many you can do. We don't have all of these restraints around them, because it's not what our customers want. So I don't -- I think it's just they're building for their customer. We're building for our customer. I don't think it's a matter of like a number. It's just -- it's sort of a -- if I went to a customer and said it takes me 0 seconds to start a car, and you're a developer, you're like, that's not actually the way car starts. And so I think that there's some marketing in there that speaks to the different audiences that we have. We can't go out with an architecture that simply doesn't state, nothing is 0. It's not -- if it's 0, it's not a start-up time.

Excellent answer. Joshua and Adriel, thank you so much. We've run over, which is the first time I've done this morning. I could spend hours more with you. I really, really, really appreciate the time. Thanks so much.

Sure. We appreciate it, and thank you all for listening. Take good care. Stay safe.

Bye, everybody.