Fastly, Inc. (NYSE:FSLY) RBC Capital Markets Technology, Internet, Media and Telecommunications Conference 2022 November 15, 2022 4:20 PM ET
Company Participants
Vernon Essi – IR
Ron Kisling – CFO
Conference Call Participants
Rishi Jaluria – RBC Capital Markets
Rishi Jaluria
Let’s get started, and last session of the day. So thanks everyone for being with us. For those of you that don’t know me, my name is Rishi Jaluria. I cover software here at RBC. I’m delighted to be joined by both the CFO, Ron Kisling, and Head of IR, Vernon Essi, who used to be in my seat, on the sell side, once upon a time, and probably one of the best decisions he could have made was leaving that and going to the corporate world.
Vernon Essi
Don’t say that.
Rishi Jaluria
But great to have you both here. Thank you so much.
Vernon Essi
Thank you.
Question-and-Answer Session
Q – Rishi Jaluria
Maybe let’s just start kind of quick overview for the general listening in the room, right? Fastly and kind of this evolution from being next-gen CDN into this edge computing and security opportunity that we see today?
Ron Kisling
Yes. So I think, you know, Fastly is a leading edge network company. And I think one of the important things to understand kind of our value prop is how this industry sort of evolved when the edge network first came out. It was really about protecting companies, servers from going down, you get a high level of traffic, big news event, and your server goes down. And so off of my, you know, came out, one of the first ones out there and said, you know, we’ll build the data center, we’ll do the servers when that peak comes, we’ll protect your network.
What’s really happened over that period of time is it’s really gone from how do we protect your network to the motivation to go to the edge is really about performance. It’s about the user experience. It’s about e-commerce. How do I get the quickest response on the cart? How do I get personalized information to maximize revenue, reduced cart abandonment, how do I deliver extremely high quality video timely without buffering? And so that’s really kind of the driver today.
And I think, you know, what Fastly did subsequent to going public was they very quickly acquired Signal Sciences to fill out sort of the core security that you need for web applications at the edge. And then subsequently COVID happened driving a big acceleration in the business. And so a lot happened very quickly, a lot of reaction to try to manage that change in the business ahead of that. And so coming to where we are today as we sort of look at the business and say, okay, how do we build a sustainable business and how do we drive toward profitability and leverage in the business.
Rishi Jaluria
Awesome. Let’s – maybe I want to start coming out of the quarter, right. So you seem to be showing some pretty nice resiliency in this environment right? And I think contrasting that with some of your competitors, I mean, we heard Akamai talk about weakness in their CDN business. Agio, the old limelight seems to be kind of having a rough goal of it right now. What is different about your business versus those specifically on kind of the delivery and performance side that’s driving this relative resilience.
Ron Kisling
Yes. I mean, I think first and foremost, I think with the headwinds in economy I think you can make an argument whether, you know, Internet traffic goes up or down. I can say there’s some segments that are probably countercyclical, like streaming. But I think more importantly, if you look at where Fastly is today, we have very low market share and very low penetration in our largest accounts in a very large market.
And so what’s really driving our business today is your market share gains at the customer level and expansion within our customers through taking bigger share of traffic, adding on security products like our WAF or our DDoS. And then emerging, all the not big contributors today is sort of the compute opportunity that allow us to grow within customers. And so that’s the primary driver that leads our growth.
Rishi Jaluria
Yes. Got it. That’s helpful. And maybe just sticking on the macro before driving into some more important topics. I’m sure you’re still, like, you know, at least monitoring macro headwind. If they were to kind of show up, how would that impact your business because you talked about there’s this argument to me. Does it actually kind of — certain parts kind of cyclical? I’m sure commerce may be a little bit more cyclical. How do you think about that?
Ron Kisling
Yes. I mean, I think we’re monitoring it very closely. And I think over the last year, we built a much more robust sort of forecasting process that involved very deep engagement with our largest customers almost on a weekly basis to really understand where the business is going so that we can manage to the direction of where our revenues and the demand is coming from, as well as manage our investment across infrastructure to support that.
I think, again, the main driver is going to be this market share gains. I think if you were to see the economy sort of impact, I think where you start to see it is probably first in commerce. I think streaming remains strong. I think you’d see it amongst a few of our larger commerce. And so you would have maybe some increased weakness in some of our customers.
Again, I think the bigger drivers are going to be the expansion opportunities but that could be a mute our growth rate a little bit. We’re not immune from those, but it’s not the biggest driver of our growth.
Rishi Jaluria
Yes. Got it. That makes sense. All right. So maybe let’s kind of think about three separate buckets of the business, although I don’t necessarily think about it that way but investors do. Right, of the core CDN and performance, there’s security, and then there’s kind of the compute side, right? If we fast forward out five years, what does that sort of mix look like over time?
Ron Kisling
Yes, I mean, I think that’s probably a good way to frame it because I think internally we sort of look at it similarly, which is kind of our core business today is the delivery business, the platform. The growth engine is really the security business and then the sort of emerging technologies, the incubation is really compute and observability that we announced last week.
I think as you move forward fast — move forward five years, I think if you look at sort of the business today, your delivery business is growing probably in the high teens. Security business as a whole, probably around 30%, compute very nascent today, but a higher grower. I think at that sort of 2025, I’d say you’d look at the non-delivery piece to be roughly a third of the business.
Rishi Jaluria
Okay. That’s great. Let’s drill a little bit down into the security side of the business. So you know, you talked about Signal Sciences, that does seem to be doing well. Maybe can you start with, you know, how integrated has it become with the rest of the security portfolio, especially because there were areas that there were actual pieces of overlap between that two. And how should we be thinking about the non-Signal Science as part of this?
Ron Kisling
So I think if you talk about Signal Sciences from an integration perspective, I think the first thing we got to integrate was the sales team so that we could sell the delivery with security that allowed us to get a better attach rate.
The second thing that we did earlier this year was we effectively migrated the technology or the WAF onto our edge compute platform, so now runs on Fastly. What remains to be integrated is, if you will, the management console, how you, you know, control and configure it. There’s still a Signal Sciences management console to configure. There’s still the Fastly delivery platform.
And so we’re currently working on ringing that all into one platform. You can configure the whole system, you get integrated reporting across delivery and security from that one platform. And so that’s really the – I think the next step that takes away a lot of the friction in terms of driving, adoption or cash from security by being able to just install and manage one platform.
I think if you look at the rest of the – the way I look at the security platform is, you know, the security we focus on is web application platform. What do you need at the edge for that. And it really is three main products. It is the firewall, which is the Signal Sciences product, which I think really is an industry leader. It’s — you sort of measure that by – how many — what percentage of your firewalls are turned on in blocking mode? Lot of them are monitoring mode just because of the false positives. We’ve got well into the 90% of the Signal Science WAFs are in blocking mode.
The other component is DDoS or Distributed Denial of Service protection. We have a strong product offering there. I think we haven’t maybe packaged it as well as it could be. I’ll come to that in a minute, but we have very strong DDoS protection.
And then on the bot side, which is the third component of your web application software. We have a product in that space. I think there’s, you know, still some opportunities where functionality we need to improve, that’s on the roadmap for next year, where we have customers that have really specialized need around bot detection. We have a partnership with human, and then we bring them in where there’s really specialized needs around bot.
Rishi Jaluria
Got it. Got it. You said you wanted to get on the DDoS. I guess when we think about just packaging a security, what needs to be done there to get better market awareness?
Ron Kisling
Yes. I think there’s two things. Coming back to delivery, Fastly heritage has been selling into the very large complex customers with dynamic traffic patterns. And so, we’ve typically sold it sort of all part here’s a list of all the features and functionality. And you can say, which ones you want to turn on and how much you want to pay for each one. So each proposal is with folk. It works great for the top 20 or 30 customers. But beyond that, it creates a lot of friction in the sales cycle.
It elongates sales cycles. It creates frustration for the sales team. And it makes it challenging for customers to say, I wanted my delivery what do I need for it? And so, one of the things that Todd shared this on the call, I think he – one of his observations, you know, in the first 24 hours, if you will, was we need to package the product and create packages that have what companies need to do to deploy at the edge with security that provide predictable cost that really applies across kind of the rest of the enterprise space.
The other thing by having packages is you can’t really sell of the spoke quote into the channel. Signal Sciences has, and we still have a really nice channel motion with the Signal Sciences product right. By building these packages, with predictable revenue, we can bring delivery into the channel as well. And so, we take away a lot of friction in the cycle. We create a set of revenues that are more SaaS based or predictable as a portion of our revenue. And we can leverage the channel to expand our presence in the mid-market.
Rishi Jaluria
Yes, yes got it. But let’s turn to edge for a second right? I think, we’re still very early in the edge computing opportunity. How do you think about sizing what is a realistic market opportunity for edge?
Ron Kisling
Yes I mean, it is really early days, and I think a lot of what is going to drive that is really what are the use cases and I think from our perspective, I really view it as we start to see it deployed and we’ve seen a couple of our largest commerce companies deploy it. We announced a large deal in Q3 where they deploy it.
And I think as people start to see the types of use cases, I think they’re going to come up with additional ideas on use cases, and I think you’re going to see kind of a rapid acceleration. There’s a lot of different opinions. I know there’s, I looked at a couple analyst reports. Someone had sized it at $2 billion I think still early days, but I think there’s an enormous opportunity when you really look at compute.
Because as I said, a lot of people are going to the edge, to create fast immersive personalized experiences, whether it’s to bring people in to their video platform, whether it’s commerce, whether it’s finance, and the more opportunity you can to personalize that using compute. As experiences become more and more immersive, it’s just a natural way to build those experiences very responsibly at the edge.
Rishi Jaluria
Yes, so on edge, I think one of the top questions I get from investors when we’re talking about edge computing as an opportunity is where does edge computing end and where does central cloud be right and what are the use cases for each and how do they kind of play in each other? How should we be thinking about that?
Ron Kisling
Yes, I think, what you’re going to use, over time, I think a lot of the data that is, not tied to that immediate experience, the stuff that doesn’t need to be served up immediately. There’s a large amount of that the central cloud plays. It plays a big role. The edge is how you deliver it to the user very, very quickly. And with, caching and dynamic data management, you can over time given those patterns, sort of cache at the edge, the stuff they need to see and keep the stuff that’s not being used in the central cloud.
And so being able to move that back and forth, which, one of the things our product does with dynamic data management is really tries to bring that content to where it’s being used, and they work very well today. So ultimately, that’s the big ecosystem in terms of how do I get data and experiences to the edge. And I put the data that’s used close to the edge and the data that’s less used is a lot more efficient to keep that at the central cloud.
Rishi Jaluria
Yes no, that makes a lot of sense. May be alongside that if we if we think about the central cloud providers, right, AWS, Azure, GCP. Two of them have some offerings in CDN or they may call it Azure at this point. Do you worry that they’ll become more competitive in edge over time or is this going to be a kind of different vendors for different use cases?
Ron Kisling
I think my sense is, at least for the foreseeable future, I think it’s going to be different vendors for different use cases. I think, the cloud providers still have a lot of work and focus on the cloud side. And I think, as the edge becomes more and more immersive, the end solution is a very fast, resilient, global network with compute and with a full suite of security to be able to really sort of deliver that.
And so I think in the foreseeable future or in the foreseeable future, I don’t see them being sort of big players in that space. I think they’re great partners. We partner with some of the big central clouds on some of our deals, and I think ultimately that’s an ecosystem of edge and central.
Rishi Jaluria
Yes, that makes sense. I’ve got a bunch more questions, but I did want to open up to the audience if there were any questions for Ron or Vern?
Unidentified Analyst
When you guys talked about bringing everything onto one platform, how would that change your pricing model?
Ron Kisling
So today, all of our security runs on, the delivery platform. We do price for the various components separately. It also runs on the same platform. I think the bigger driver that may influence pricing is really how do we package it and what services are included in that package and how do we price those versus pricing them separately. So you get a bundled price for security and delivery.
They’re certainly – even on the delivery platform options across, low balancing media shield and other services just on the delivery side. And so, I think as you package them, the goal would be to provide one package pricing that the customer sees for a solution versus buying individual products. And I think the other benefit as we roll in other products is, each of those does have a value. So if I’m selling, my firewall, and delivery. I’m essentially getting paid for my traffic maybe at different rates twice. I’m paying for that traffic once.
Rishi Jaluria
The questions are fielded. Oh, sorry yes there is a mic coming there.
Unidentified Analyst
The streaming industry, on its own are going through – a change, they’re facing more competition among themselves. Does it matter to you? How does it impact you if it does?
Ron Kisling
Yes it’s a really good question. I think, with the broad presence in gaining customers, I think we’re a little bit agnostic. In terms of, where the traffic goes, whether it’s, Disney Hulu HBO, it’s if it’s across a customer, we’re a little agnostic in terms of who’s driving the traffic. And I think, some of the, earlier noise around sort of Netflix traffic is down. You know, was that people streaming less or was it, as more and more providers are out there, the bigger volumes across a lot more players.
Unidentified Analyst
They are facing lot of cost pressure because they’re just competing themselves to gain market share. Will that ripple through to you and maybe you may have to reduce the price that’s what I meant?
Ron Kisling
Yes, it’s a good it’s a very good question. And I think what we’ve seen particularly in streaming, but I think across all of it is that sort of on an annual basis, there is some price declines. I think what we’ve seen over the last say 12 to 18 months is maybe a decline or slowing of that annual decline in pricing. And I think some of the drivers around that are, I think across the industry, all the players are being a little more disciplined around outstanding, what are what the costs are and what their pricing is.
Offsetting that is, if we negotiated our bandwidth costs, our data management or bandwidth management, traffic management, gets better every year. Our service get more efficient every year. So we have a similar sort of downward, it’s just like with computers every year they get more powerful. Because maybe moving a lot more data with the same equipment. I think the other data point that when you start to look at those dynamics, which is a real dynamic, is we have a very efficient network that is very capital efficient.
And so, we have an opportunity with our cost structure and I’ll talk more about gross margin, but we do actually have an opportunity to have a lower cost to deliver edge to these streamers than some of the other players in the market, which gives us an ability to compete equally with them and still have favorable gross margins.
Vernon Essi
Yes, if I could add one thing on this too. On the streamers, I mean, another point which really is part to your question is to differentiate. They’re getting more into the live streaming game, and sports, is a big piece of that. We would argue we’re probably the best CDN to carry live sports. I mean we have very low latency. And so you’ve seen a lot of that content and activity going on of late.
So we’ve benefited from that. And I think you’re going to see more of that go on in the future. I think I’m not mistaken this year is the first year we’re streaming actual television is overtaking live right now. So there’s a big change happening instead of over the air or broadcast versus streaming. It’s flipping right now that’s going to continue.
Rishi Jaluria
Any other questions out in the field? Yes all right. Well look, I’d be remiss if I didn’t ask some margin questions. I am having CFO up here with me.
Ron Kisling
I’d be disappointed if you didn’t.
Rishi Jaluria
So let’s start on gross margins because you did just mention that right. What is the long-term kind of gross margin profile of the business, can we see something that’s more typical SaaS call it like mid-70s or is there a structural inhibitor to that?
Ron Kisling
I think the way I would answer in terms of our trajectory, I think there’s certainly an opportunity to be a lot more efficient around our investment in our capacity. And I think we’ve talked about some of the things we’ve put in place about a year ago. We’re starting to see some of those benefits better bandwidth management. So we’ve indicated we expect to see a couple hundred basis point improvement going into Q4.
We expect to see continued improvement in 2023. I think we exit 2023 within striking distance of 60%. And I think even based on the current product mix that gross margins in the low 60s are very possible. I think as you add in more and more compute, you know, you add in more and more securities, those become, roughly a third of the business. I think there is an opportunity to get into the upper 60s. That’s probably the visibility that we can have today across that.
Rishi Jaluria
Got it, got it all right. And then, the second part of that question is going to be on the long-term. Yes our margin going to follow the business? How should we be thinking about that?
Ron Kisling
Yes, I think it’s one thing we’ve spent a lot of time talking about. I think as we look at the business, we’re focused on growth. I think the gross margin opportunity gives us some tailwinds in terms of getting to cash flow breakeven. There’s, a lot of operational efficiencies that we’re going to be executing across OpEx next year.
I think when you pull all those together, well we haven’t shared some of the specifics. I think as you look to sort of the – if you get to the end of 2023, I think a breakeven adjusted EBITDA is very feasible.
Rishi Jaluria
Got it, okay that’s helpful all right. Going to the competition question I really want to drill specifically into cloud right? They’re obviously making meaningful investments in edge company seems to be doing well. How do we think about the competitive dynamic there?
Vernon Essi
Yes, I think when you look at kind of the delivery side, we don’t typically see the most, I might say, 99% of the time right – percent of time, we really are going up against Akamai. They’ve built a really good business across the mid-market, across the entry market, their products really well tuned to that. They haven’t really made significant inroads into, I think, the enterprise space. And I think – the reliability, the scale, our network.
I think provides – I think we’re better positioned in that space. To- date, we haven’t really gone after the mid-market. I think there is an opportunity as we start to work with partners to maybe start to see them more often, head-to-head. I think those would be pretty competitive. I think they’re particularly good in that space. I think we have also a very good network, but I think we’ll start to see the more, I think, in that space.
Rishi Jaluria
Yes, yes got it, got it. Maybe going on kind of the go-to-market side now – I think one thing that really impressed me about Fastly back in the day when I informed you guys since you were small little private company, was really just kind of that developer adoption right?
And one of the big things people liked was, self-service and kind of this PLG strategy and now it’s becoming the acronym as you’re in the valley, but you had been doing it for the longest time. May maybe can you talk about maybe re-returning to those PLG with Fastly and what’s the plan to inform them?
Vernon Essi
Yes, I think there’s probably a couple of things one, I think a lot of things sort of play together. I think the packaging makes it a lot easier with self-served to actually, makes that lot easier for adoption. We’ve certainly, particularly focused around compute, really focused on our developer community. I think the acquisition of Glitch was a key part of that. They’ve sort of grown their development community.
I think now it’s about 2 million. And that is one of the – those are really kind of three levers to really, I think, sort of take advantage of that sort of self-serve. It’s going to be particularly important as you get to sort of the mid-market. And I think as you look at kind of what the, the complexity or how easy it is with us to sort of onboard a new customer, it’s extremely easy. And so we’re a lot easier to do that sort of self-service with the ease of onboarding than maybe some of the other players in the space.
Rishi Jaluria
Yes. Got it. And then maybe turning a little bit back to the security topic, right? So obviously, Signal Sciences, we’ve talked about the rest of the portfolio. As you think about kind of broadening out into security, are there other areas where you see opportunity to expand into and maybe alongside that, how do you think about that don’t versus buy debate?
Ron Kisling
Yes. I think if you look at security, I think particularly around web application security is where we’re focused. And I think today, we’re not really looking to move beyond that. And then I think if you look at what’s needed in that space, I think we have the technology around sort of the three key pillars of that security that we need. We need to do additional — some additional more work on bot but that’s internal work.
I think as you start to look at new areas, right now as I sort of characterize how we look at the business, that’s sort of the incubation or new business is compute. You know, that’s essentially opening up what we run our security, our edge platform on to customers to run their own applications that’s internally developed.
We just announced the durability. So I think in the current incubation there are technology gaps that we need to go out and fire. I think three to four years from now, security probably moves into the core, which helps sustain growth levels there, compute, maybe observability moved into kind of the growth driver. And then that new incubation could be a combination of either new technology, tuck-ins or own development.
But in the near term, there are no significant gaps where we feel we need to go out and buy technology to plug a gap in our solution.
Rishi Jaluria
Yes. Got it. And then maybe if we just think through the customer base, right? Maybe the flip side of the enterprise is having meaningful customer concentration, right, in contrast to a cloud flow that’s very long tail. Do you see this as a major risk? Or maybe the more important question is how do you manage that sort of potential risk, right? I mean, we saw that with TikTok in 2020, unnamed customer last year with the outage. How do you think about just managing that and —
Ron Kisling
Yes. I think it’s — a lot of managing it is, of course paying attention. I think a lot of it is the increased forecasting visibility we have for talking to those customers in terms of traffic patterns. So we understand those a lot better. I think that we did a year ago. Our top 10 customers in the last quarter, we’re about 35% of revenue. So meaningful concentration, but we don’t have a single customer over 10%. So there is a concentration.
I think the way we think about it is, we’re happy to have these very large enterprise customers. How do we reduce our dependence on it? You know, it is through accelerating new customer acquisition, which is also one of the key drivers of accelerating growth. And we do that through better alignment of the sales organization, increasing our quota carrying sales reps, packaging, which takes a lot of friction out of the organization and being able to use partners to sell into that mid-market dramatically over time will reduce our dependence in customer concentration.
So there’s direct efforts at reducing that as much as we’re really thrilled to have these really large customers that we built up.
Rishi Jaluria
Awesome. I think it’s a great jumping off point. We are at time. Ron and Vernon, really appreciate your time. Thanks everyone for being here, and I think that wraps up day one. So, thank you.
Ron Kisling
Thank you.
Vernon Essi
Thank you, Rishi.
Be the first to comment