Fastly, Inc. (NYSE:FSLY) Credit Suisse 26th Annual Technology Conference Call November 29, 2022 10:55 AM ET
Company Participants
Todd Nightingale – Chief Executive Officer
Ron Kisling – Chief Financial Officer
Conference Call Participants
Rich Hilliker – Credit Suisse
Rich Hilliker
Well, excellent. Hi, everyone. My name is Rich Hilliker. I am a software analyst at Credit Suisse. We are excited you are all joining us, and we are particularly happy that Fastly could be here today. So we have got CEO, Todd Nightingale; and CFO, Ronald Kisling. Guys, thank you so much for being here.
Todd Nightingale
Thank you.
Ron Kisling
Thank you.
Question-and-Answer Session
Q – Rich Hilliker
So, maybe Todd, first and foremost, congratulations. We are excited to have you up here. 90 days went by quickly. If you are in the saddle, you are doing really well, it sounds like. So would love to hear your perspective, joining the company, eyes wide open, kind of drinking from the fire hose, how has it been? What were you most pleased about? What were things that you are going to spend a lot of time focusing on? Maybe give us the lowdown of your first 90 days?
Todd Nightingale
It’s not 90 days yet, man. Give me a couple of days.
Rich Hilliker
One more day? One more day, 88, 87.
Todd Nightingale
Yes, it’s been amazing. I think you when you are considering taking a job like this, you do as much research as you can, meet as many people as you are allowed to, I guess. I was lucky to have known Fastly to some extent in my previous role at Cisco. And I thought I knew a decent amount about the business, but I spent a good amount of time and kudos to the founder, Archer and the former CEO, who went way out of their way to make sure I got every possible question asked – answered. But – yes, when you come in and all of a sudden, it’s – you are doing it every day and you are going through every ounce of data, there were some super interesting pieces. I think the biggest one for me, the biggest maybe positive surprise was the innovation roadmap, the velocity of innovation at Fastly and the way the roadmap is playing out, I think was in a lot better shape than I thought perhaps from the outside, especially in that there is sort of a core business in network services and content delivery and there is a growth engine – a growth business in security, especially with the Next-Gen WAF and Signal Science acquisition.
And there is an incubation technology area in edge compute and we just did a very early launch of another incubation area in observability. And I think from the outside, it was hard to see that these modules were sort of like well-formed and that there was kind of core growth incubation, which is that durable innovation strategy that I want to run. And so coming in, it was just really nice to see that in the engine, the sort of innovation velocity was in such great shape. It was a gift. It was my like biggest question mark coming in. And I feel like maybe what maybe comfortable joining is that’s sort of where I come from. I come from the innovation side of the house. So yes, that was amazing. And then maybe my surprise on the challenge side, not the negative side, was maybe also kind of tied to that with sort of how well the products are packaged together and how these modules are sort of demonstrated and brought to the market, which is part of the reason I think I had a hard time seeing it from the outside. And so we are working feverishly on packaging right now to get really kind of teed up a new way of bringing technology to market in the new year, but yes, it’s been great. The only other thing I’ll say is coming back to an organization of this size, just it just feels great. Like, it was just incredibly welcoming. I just spent the last week with my family in New York. Thanksgiving is my favorite day of the year. And almost all of them told me, you look so happy now. So that was really nice to see.
Rich Hilliker
Well, that’s great. Well, congratulations again. And I think this is a perfect segue. I would love to talk about – you outlined a number of priorities on the call a couple of weeks ago, one of which was product packaging. So maybe could you help us understand a little bit more of your vision here to kind of really reveal to the market, as you mentioned, this durable growth engine, right? There is such a line of sight towards all these different products. So how are you thinking about going to market and the steps to bring this to life and maybe the time to value there?
Todd Nightingale
Sure. Yes. I think that’s incredibly top of mind for all of Fastly now. I think every organization is a product of where we come from. And at Fastly, we are sort of born in the publishing, media, entertainment customer base. And those are customers where the network service content delivery that they are buying from a vendor like Fastly so important to their business that they want to lean all the way in and they want to buy it piece by piece and negotiate every feature in every product. And they like the kind of pure consumption-based utilization model, which is great. And we are – that’s how our business runs. Our core business all runs that way. The security business runs as a SaaS model, but the core business runs that way. The problem is that there is a lot more of the market that wants a reliable, predictable price. They – every month, they want to know that they bought the whole solution that they are not – that their engineering and development teams aren’t going to find some new feature they want to use that will be priced differently or that will be a surprise. They want to buy the entire kind of package and have that bundled with enterprise support, all the bells and whistles. And so by providing that as an option, we bring something to the rest of our customer base that they kind of sorely want and we hear this from the rest of the customer base, like it could be easier to buy things from Fastly. And I think we have a real opportunity to do that.
There is a secondary effect of it, which I think is really, really important. And that is that it radically simplifies the operation within the company. It means that we can do standard package, standard discount – sorry, standard package, standard discounting strategy and hierarchy throughout the team with a standard price and it affords us the opportunity to approach the channel in a new way, because when you have a standard package, standard discount deal registration that opens up the opportunity to run a high-velocity motion with the channel. And so just making our operations simpler, increasing the speed with which our sales team can operate, I think, has a huge lift to how efficiently we can run. And so our teams are kind of, I think, excited about this across the organization.
Rich Hilliker
Absolutely. So as you touched on the partner ecosystem, what would the vision be here? I mean, a lot of ideas, I am sure you are having and coming up with as you move through here. But that streamlining, lowering the barriers to adoption in some ways, how are you thinking about – or which areas of the business do you think are most ripe for the partner involvement?
Todd Nightingale
Yes, that’s a great question, because this is a discussion that’s like happening at Fastly right now. It’s incredibly important. There is a lot of partnership opportunities for Fastly. We can OEM technology, we can become an OEM to people that we call a platform partner, etcetera, where Fastly – where their site and their offering becomes powered by Fastly. We have the opportunity to do different kinds of co-marketing, etcetera partnerships. All of those types of partnerships are – there is all levels of different opportunity and cost and risk, etcetera. But I think it’s become clear to our leadership team in my first 90 days here that the channel partnership, the systems integrator channel, has enormous upside for us in that so many systems integrators come with the software expertise, the development expertise, to onboard on to Fastly so quickly to drive all the value that our end customers want out of an edge cloud platform. And they can bring business into Fastly and then they can help service those customers for higher customer set. And so that is just a – that’s an amazing kind of limited downside high upside potential kind of opportunity. And so we have decided to like really lean into that channel partnership. And so the packages are being built standard with standard pricing, standard deal registration etcetera to run a partner program that can be a high velocity motion. I will also mention like that’s sort of where I come from. So I am like eager to sort of rekindle that motion.
Rich Hilliker
Awesome. Well, that’s exciting.
Todd Nightingale
But one other thing is that Fastly in the Signal Science acquisition brought in a business that ran through the channel very efficiently. And because of that, the security product line at Fastly is like ahead of the game. It’s already packaged in a SaaS motion. It already has predictability in pricing. It already has a channel motion that works and channel partners that are engaged. So we don’t have to start from scratch. We just have to take those channel partners and really make them Fastly wide partners. And that’s amazing.
Rich Hilliker
Got it. Well, maybe on Signal Sciences, you have talked about it as sort of the growth area of the business, right? We have talked about they have a channel motion. I would love to know your thoughts in terms of – if you were to give coming in – giving Signal Sciences, Fastly combined here, like integration, like what sort of score would you give it? How happy are you with the progress that exists now? And then maybe looking forward as to the vision, how far are we along in completing the vision and then also kind of executing against that vision?
Todd Nightingale
Yes. So on the go-to-market side, I think the integration, I would say it’s 60%. I’d give it a B. And the reason why is that internally in Fastly, we’ve really – we brought the teams together. We have one go-to-market motion. That’s amazing, but the channel still is kind of security only, while Signal Sciences has now sort of been brought in to become Fastly Security and we’ve added some Fastly products into that product line, etcetera. It’s still only the security products that are running the show. So when we are able to bring the channel across the portfolio, I’ll give us a A. But on the product side, to be honest, I agree, wouldn’t be as good. There is still an enormous opportunity for us to unify the user experience. One unified UI, one unified API, a unified user experience, a unified developer experience.
And by doing that, we sort of open up the door of a land-and-expand motion where every single content network service customer at Fastly, with a single like click, they can onboard the security technology. They can enjoy best-in-class Next-Gen WAF and DDoS and the security technology that’s built into the platform, 100% feels like one experience, one – almost like one continuous use case. And then the teams are working on that right now because I think there is a huge opportunity here. Once that path starts to be well trodden, we also, I think, will – that will sort of demonstrate this durable innovation strategy of landing, expanding from content delivery to security, to edge compute to observability and what comes next.
Rich Hilliker
Got it. Okay. That makes a lot of sense. Maybe about developers. You’ve always been developer-centric. Archer – that was some of the key founding principles. You’ve acquired Glitch, you’ve reestablished the fast-forward community. So clearly, a lot of effort and focus here to continue to drive velocity there as well. What have you found? And how are you thinking about targeting the developers to continue to capture that mind share and kind of move down that multiproduct expansion motion, have them standardizing. So where are you going to be focusing time?
Todd Nightingale
Well, I’ll tell you, the Glitch acquisition was hard to understand externally. Internally, the incubation of edge compute at Fastly is so critical because in so many ways, it is sort of the next generation of the core value proposition. Content delivery – like, traditional content delivery had been for years about sort of protecting the web server from loads, so under high surge load, your web server would stay up and you your content would still get delivered from the CDN. Fastly’s value proposition is about user experience. It’s about how reactive and how engaging the application, the website, the web experience, the streaming experience is for the user, for the end user. And in order to deliver that, edge compute is sort of like the next big step where you can have fully dynamic content, driven by custom data, etcetera, delivered right from the edge, and we can bring the latency down to the point where it’s just completely immersive. And that sort of next step in how Fastly delivers our core value proposition is so critical. In order to deliver that, you just have to be close to the developers, right? And edge compute is about pushing real custom workloads right to the edge as close to the user as possible, providing the lowest latency experience and really delivering experiences that are compelling enough that they change the user experience.
They change how frequently carts get converted, how frequently e-commerce customers click on a recommendation or media customers look on a recommendation. They have – edge compute has the opportunity to change the way we deliver gaming or digital IoT services. And in order to do that, we have to stay super close to the developer community. And that’s why Glitch has just been incredibly valuable resource. And so we’re we are really focused on continuing to grow and sort of curate that community and really use them as a resource to drive our edge compute road map, which is exactly what we’ve done. We’ve focused on language support and API support to build the most developer-friendly edge compute platform in the world, and that has been very, very successful for us.
Rich Hilliker
Excellent. Sounds like more to come there, too. Well, very good. So maybe we could talk a little bit about consumption versus recurring services. Would love your thoughts, like longer-term, as we think about the business, we’ve talked about – you’ve got a lot of irons in the fire, right? A lot of opportunity, things that you’re working on. I mean, how should you encourage us to think of that, right? Because historically, the core business is obviously consumption. You mentioned SaaS model for security. So as we think about like the mix moving forward, how should we be thinking about that? And how does – how do packaged services that you’re working on, how does that impact that mix?
Ron Kisling
Yes. So I think that pivot does level set today, security being one of the recurring revenue. And essentially all our delivery is on a consumption basis. With the packaging that Todd spoke about, when you get below kind of just our top high-traffic customers, as they start to buy packages, that’s going to be sold on a recurring revenue basis. So it’s going to provide predictability to those customers. It’s going to drive a bigger percentage of our revenue is going to be on that sort of SaaS model and recurring revenue. And then also, from our perspective, going to reduce some of the revenue volatility that we see from consumption as we see more and more business coming as recurring revenue-based contracts.
Rich Hilliker
Got it. Okay. That’s really helpful. And then as we think about how traffic on the network might evolve, like how are you thinking about that, right? You talked about some really interesting partnerships and opportunities. You’ve talked about private relay, traction and edge compute or Compute@Edge rather. How are you encouraging this evolution? As we – and then relative to the mix of recurring relative to the packaging, kind of bringing it all together.
Todd Nightingale
Yes. I think for us, this sort of innovation trajectory this land and expand motion, it gives us a kind of differentiation, the kinds of services we’re offering. It also – which changes our traffic load and actually makes it more efficient, the more diverse the cost, the product suite is, the more efficient our infrastructure to run, which is great. Maybe it’s not the best – maybe that not a reason to do it, but that’s a nice lift. But as far as the kind of vertical mix goes, I think it does open door to a broader set of customers because it’s a more and more complete solution, right? And to me, that’s sort of the long-term road map at Fastly is building the most complete edge cloud platform. And in that – in doing so, every customer who’s looking to build an immersive like responsive, engaging user experience, will turn to Fastly. And the more complete our offer, the easier it is for development teams and DevOps teams to onboard. I think the faster we’re going to adopt those new customers and drive that motion. So I guess to me, part of this portfolio completion play is diversifying the customer base.
Rich Hilliker
Absolutely. I think all of these questions so far have led us up to probably around your favorite topic, gross margins, right? As we think about the completeness of the platform and establishing this multi-product to market motion, how are you thinking about the progression of gross margins, right? There is been a lot of debate on topic, it’s been topical. Maybe walk us through what the path forward looks like from here.
Ron Kisling
Sure. I think one of the things when you look at say, 2022, we’ve had a lot of sort of kind of noise on that. If you sort of extract that and kind of look at kind of the long-term trajectory, going back to ‘20 and ‘21, our gross margins were kind of in the low 60s. Coming out of that high spike in traffic and some of the supply chain constraints, we made sure we had adequate equipment. We deployed some of that a little bit earlier. Some of our processes around aligning traffic expectations going into ‘21 and ‘22, those processes weren’t quite as well defined. And in some instances, we overbuilt capacity. And so the gross margin declines that you saw largely in ‘21 and through 2022, were largely driven by just kind of overbuilding the network, some international expansion, which initially when you set up a presence in a new market, typically, you do have an impact on margins until you build enough traffic to accommodate that site.
And that’s really been the driver. What we put in place over the last year is a robust process around planning, engaging with sales to do a much better job of understanding traffic expectations from our customers and really deploying in-line with that traffic expectations so that we’re driving more optimum utilization in our network. And so we’re still in that process of seeing that utilization get to where we want it to be. And that’s what we’re starting to see in the second half of this year where we saw gross margins improve 300 basis points in Q3, where we expect Q4 to be another couple of hundred basis points higher.
And then as we move into 2023, there is other levers around we’ve renegotiated some of our bandwidth costs to reduce those. We’re increasing the use of peering networks, which is extremely low cost as that percentage goes up, our costs go down. bandwidth is really our biggest cost. And then we continue to work with the engineering department and making our servers more efficient. And so as we increase that efficiency 40%, we can use 40% less servers. And so those drivers are going to continue to allow us to see year-over-year improvements in gross margin through 2023. And I think if you want to kind of level set in terms of what that opportunity is, I think as you get to the end of ‘23, we’re sort of within striking distance of 60%. And then I think on top of that as you get beyond that, the packaging around delivery, the growth around security which again that diversity of traffic allows us to utilize network more efficiency or all opportunities to drive gross margins even higher into the low and mid-60% as products become a bigger piece of our revenue pie and we start to sell a lot more of our delivery around packages and recurring revenue.
Rich Hilliker
Excellent. That was really helpful, a lot in there. So, maybe we will touch on a few of those. You talked about traffic planning. And I know that you have done a lot of work on forecasting and understanding – trying to understand the traffic, the patterns and being able to build for what’s appropriate, right. I was wondering if you can kind of help us understand some of the changes you have made.
Ron Kisling
Yes. I think the biggest thing when I came in on forecasting whether it’s on revenue or traffic patterns was it seemed to be pretty siloed. Finance was sort of doing their work, sales was doing their work, and engineering was doing their own. And so one of the first things I did was bring together initially sales and finance to come up with kind of the single forecasting process. And one of the things that we did because we are consumption based business is we engaged with all the sales reps to talk to their customers on a regular basis. And for our top 100 customers, we talk to them regularly and we get an estimate from that sales rep in terms of what is the outlook for that customer. And I think that’s what’s important in a consumption business, to provide better visibility and reliability. That was really kind of the first priority. And I think that’s been working very well this year. It’s also just driven really great partnership with the sales team, but given us better visibility. And then once that was in place, we put in place kind of a build plan, meetings with infrastructure that included finance and the sales team sort of leverage our revenue forecasting, which was based on traffic as well as sales outlook in terms of what’s coming down the pipeline, what are some of the new deals, where are they located. And have those discussions around what those are, where do we need to invest, when do we agree to invest in the network based on pipelines and opportunities. And so that’s really kind of how that’s evolved. I think with our new leader, we have a new leader running our infrastructure team. I think that process is getting even more refined with a better understanding of even traffic patterns and how we even layer traffic of low priority traffic, how do we optimize that to the network, how do we route traffic during peaks to minimize bandwidth cost.
Rich Hilliker
Right. It sounds like you have been busy, Ronald.
Ron Kisling
It’s been a really great 15 months.
Rich Hilliker
Excellent. Well, maybe another topic that you originally touched on was peering. I know Todd, we have talked a little bit about this, too. How are you envisioning your peering strategy relative to what it’s been historically? And how does this, in your mind, really help – you talked about it being low cost. Like, how do you attack the strategy? How do you establish it and how do you drive the most value here?
Todd Nightingale
Yes. It’s a good time for us right now when it comes to peering. I mean I learned about peering in my grad school when I was learning how the Internet is built. I never thought it would be such a core part. Yes, we buy our biggest driver of cost is network costs. And so when our traffic gets to a certain scale, it makes sense for different types of network providers to peer with us and just have a direct – usually costs for your very low-cost link to connect our two networks. And thereby, we can offload the traffic that’s going to their users directly. We don’t have to pay the very high bandwidth costs. And so every time our scale increases, we become kind of a more attractive peering partner, and that allows us to offload more traffic in those low cost – to those low cost lines and basically take track off of the very high-cost upstream link. That’s – so because of that, we just have this natural opportunity to be able to peer more aggressively as we scale. Every time fast as traffic doubles, we become a more attractive peering partner, that’s good for our business. But the other side of that is also technology side. The network automation, the network engineering that’s required in order to orchestrate all of this – all of these peering links, in one, it makes the Fastly offering the premium offering that it is. Like, the most well-connected POPs in the world, driving the closest proximity possible to the user, the best possible user experience. And the network automation, the network engineering there, like this is an area I am super comfortable from a technology point of view. And it’s amazing. I mean I think we have an amazing team there. And so that’s been a phenomenal step forward. The automation, it makes our offering better with better connectivity, but it also – we have a lot of opportunity moving forward because as we get more and more peering links, more and more well connected POPs, we have the opportunity to do cost optimized routing and service level-based routing. And that’s a huge opportunity. We haven’t really even scratched the surface of yet. So, as we continue to scale we have the sophistication of the network automation to add more and more peering, which is going to be good for our cost of revenue. But I think it’s really important. It’s something that we talk about a lot internally because it has the added benefit of increasing of making the actual offering better, the actual responsiveness of the system better, which is great. I mean how many times do you lower your cost and provide a better offering to the customer at the same time. Yes, it’s amazing.
Rich Hilliker
Want some exciting opportunity. And maybe at the time, if anyone has any questions, feel free to raise your hand. We have someone who is running around a mic, so we will give you guys a second to think on it. And if not, I have got a couple more here we can go through. Okay. We will keep on going then. So, maybe Ron, I was wondering if you could talk through your guidance here a little bit that you issued. Nice results. And then as we think moving forward here, like what’s your level of visibility and confidence given all of what we just talked about earlier, right? How are you feeling about that? There is a little bit of conservatism from other companies in the market in general. But we saw really is – what I thought, a strong guidance from you. So, I am wondering if you can walk us a little bit through that?
Ron Kisling
Yes. I think to kind of level set kind of how we have approached guidance is given kind of the level of visibility we have. And I think at the beginning of the year, we provided full year guidance. And the visibility two quarters, three quarters, four quarters out was much less visible than so at this point, we are kind of looking one quarter out. And so it’s a lot easier quarter to guide. But if you look at what’s been really driving kind of the growth this year, it’s really been expansion with our existing customers and new customer acquisitions. And so I think a lot of times we get the question about given some of the economic headwinds, how are you feeling and why are you confident about your growth. And I think a couple of dynamics are, one, we have a relatively small market share in a relatively large market and also relatively small penetration in a lot of our largest customers. And so – that really has been the driver of our growth is expanding within our existing customers and market share gains. And so we see that continuing. And personally, I think from an Internet traffic perspective, where how customers actually deliver their product to the end user, whether it’s streaming, whether it’s commerce or financial. So, we are kind of like the utility. So, we are one of the last ones that certainly will shut off. So, as you take those sort of dynamics into place and then look at how we have built our forecasting process, I think our guidance strategy is the same. It is we are providing our outlook based on what we have high confidence in terms of success in Q4. Q4 does tend to be seasonally higher. There is a number of sort of high-traffic events. It’s a high traffic quarter for e-commerce as well. And as we have talked to our customers, this is how we came up with kind of the guidance that we feel really confident about and continuously reflects that expansion within those customers and new customer acquisition.
Rich Hilliker
Excellent. Okay. And maybe, Todd, busy 90 days, it sounds like. But I think you have had so many conversations at this point, customers, partners, a number of investors. Where is everyone maybe agreeing on – or in your conversations, where would you say there is a misunderstanding amongst those groups? Is there any sort of area that you think needs to be clarified?
Todd Nightingale
Yes. I mean I think there is a – there is definitely clarity around the margins. So, I am glad we got to talk about it today. I get tons of questions on that from the investor community. And look, I think we have a huge opportunity to correct our margin just on cost control, just on getting the costs right, building an efficient infrastructure and sort of leveraging the technology that we already have developed. So, that’s maybe like the biggest sort of misconception. And then I think the vision is key here. I mean I think in 3 years, if people think of Fastly as a CDN business, we would have failed, like that would be a huge issue. We are an edge cloud platform. That’s where we deliver value to our customers. We are not trying to relieve load from people’s web servers. We are trying to deliver the best possible user experience and edge cloud platform is going to be the answer for that for years and years to come. And so I mean I think just unifying the message around that vision is something I am super focused on.
Rich Hilliker
Well, I think that’s a great note to end on. We are grateful for you guys both attending, for Fastly being here, for everyone in the audience and online. Thank you guys so much, and I hope you enjoy the rest of the conference.
Todd Nightingale
Thank you very much.
Rich Hilliker
Thank you.
Todd Nightingale
Thanks everyone. Thank you for being here.
Be the first to comment