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

Okay, great. So my name is Rudy Kessinger. I am the lead security and infrastructure software analyst here at D.A. Davidson. With me today we have management of Fastly. We've got CEO, Todd Nightingale; and CFO, Ron Kisling, joining us today. So, thank you both for your time.

Thank you.

Thank you.

Todd, could you maybe just start out for maybe those who are a bit unfamiliar with your business, just who is Fastly? What do you do?

Sure. That's a gift. That question is a gift. Fastly, traditionally, has been a performance-driven dynamic content delivery network and a CDN company and that's sort of where we were born more than 10 years ago. But the journey has been, I think, super interesting. In fact, we see -- we see the reality of this happening across the market right now, but we've always been in the performance side of the CDN business. And from customers, even years ago, we got the feedback that what they were looking for is really an edge partner, edge cloud partner, more than a CDN point solution. And we've built out security technology, including really all of the edge security space what Gartner calls web application and API security, [ WAF ], and that includes DDoS mitigation, bot mitigation that's about to release. It includes next-gen WAF, which is really the crown jewel of that space. The portfolio has expanded to true edge compute able to dynamically deploy bespoke workloads around the world and now edge observability. And those components really build out an edge cloud platform that gives customers who are focused on user experience, focused on delivering the best performance on their web assets. It gives them a platform for that. We say, at Fastly, we make the internet a better place where all experiences are fast, safe and engaging.

So Todd you've been in the seat for just about exactly 1 year now, I believe. And I've got 2 questions there. Firstly, what has surprised you about the business relative to your expectations coming in either positive or negative?

Yeah. A lot of surprises. I feel like I did my research, I was hoping there would be not too many surprises, but [Technical Difficulty]

--the network side and the capital intensity side and then on the go-to market side of the things.

Sure. Capital intensity side, we're going to -- I'm going to ask Ron for some help there. But look, on the network side, by focusing the team on the business outcome, we were able to uncover real optimizations in our infrastructure. One of the core value propositions, the strength of the technology at Fastly, is that our infrastructure is software-driven. It's a software-driven network, it's a software-driven infrastructure 100%. There's practically nothing in a Fastly pop other than data center switches and compute platforms. And because of that, we're able to build enormous amount of efficiencies and deploy it to our fleet without having to deploy a shred of new hardware. So far this year, we've deployed 35% performance increase across our fleet with $0 spent with no new hardware deployed. And that is driving our confidence, #1, that this architecture has real power and like really unique ability to outperform our competitors' technology and that the algorithms behind the kind of performance advantage that we deliver continue to bear fruit, which is awesome. And that's what's driving some of the capital efficiency.

Yeah. What I would add is just, it is that network that allows us to achieve the capital efficiency that we have today. I think if you look at the improvement, it really is a factor of the company not having a good mechanism to really forecast traffic expectations into the future. And so a lot of the investment was made absent that information, and it resulted in a little bit of a overcapacity over buying for the network back in '21. We started to work through that in '22, and you're starting to see us align with what we think is really the go-forward capital intensity, which is probably 6% to 8% of revenue, and that is sustainable into the future.

And then I would say the other big change we made was around really business alignment and aligning the spend. So, Ron put together an amazing budgeting and strategic process to drive clear budget alignment across the company to deliver a OpEx and an operating margin result that we're proud of, we continue to beat expectations there. We hope to continue to do so. And even more importantly, on the growth driver side, taking that budget and really focusing as much of the spend as we possibly can on distinct quota carriers. We opened up a ton of regions this year. We will add up, will open up new regions next year, really designed to focus our spend on driving growth, specifically customer acquisition growth.

So I want to start, I shouldn't say start, but I guess I'm going to kind of walk through the P&L, starting on the revenue side of things. And your largest -- one of your largest competitors acquired some contracts from StackPath and Lumen, who exited the business. I think that -- those 2 deals totaled $60 million to $70 million a year of delivery revenue. There are some interesting things going on with some other players in the CDN space as well. Just what have you seen as far as consolidation over the past few quarters? How has that benefited your business? And when you look forward, do you expect these trends to benefit even more so as we get into 2024?

Sure. I think you can name the competitors, so Akamai bought the contracts of StackPath and Lumen. The StackPath and Lumen were kind of well known as being very, very low-cost providers. And not that, that necessarily for some out of the market, I think being a point solution just is not -- that is not the way this market is going to go. We're seeing platform, the platform strategy playing out in the edge cloud space, just like we saw in the central cloud, AWS and GCP and Azure, finding homes where point solutions are very, very hard to maintain success. We're seeing the same thing in the edge. And so point solutions are having a hard time holding on. I think it's generally good for the space, having a couple of the very low cost options removed from the market, should making the pricing environment a little better for us, which is great. I think that will -- that trend we'll see echoing through the system for the next year or so. But the -- the reality is, I think it's a real opportunity for us, to be honest. There's a ton of customers that have been using StackPath and Lumen for content delivery. I think they're going to be picking their heads up and making a vendor selection regardless of who bought their contract, and we plan to earn their business.

Yes. That's fair enough. On security, just tell us about where you're at with security today? I believe it's 14% of your business today that's predominantly just with the WAF solution. You're coming out with bot and DDoS here soon. What is going to be different about your products? I guess what is different about your WAF product, but also how are you going to differentiate with your bot and DDoS and come to the market with a security offering that is advantaged versus your peers? And I'll start there and then I got a follow-up.

Sure. I think we do have a great track record here. The WAF solution is market-leading, especially in efficacy, and you see this in the tech bake-offs across the industry, including Gartner. The -- it's great tech, and it comes in through an acquisition, Signal Sciences from a couple of years ago. I love the tech. I love the way it's deployed. It gives customers real choice, but we do have work to do on easing the cross-sell from existing content delivery customers into our security space, and we're working very closely to make sure that, that is really accelerating our cross-sell. We moved all of the Signal Sciences technology into the infrastructure. So it's already running in all of -- on all of our machines in all of our POPs. Our customers are able to leverage it on the same data plane, but the management plane having a single dashboard for both things that will give us a really nice leg up and ease our sales motion there. So we're very focused on that. That will be coming out in the first half. And I think that will be a huge, huge step to making sure we have the best product because it's the easiest to use. Bot mitigation is in beta right now. We have a ton of customers on it. It's going live. [ It will go limited availability ] this quarter actually, so we can start closing some business, but it'll go GA in Q1. Again, we did a very exhaustive survey of the market here of all the antibot technologies available. This is an interesting emerging space. We believe what we will have because it has so much platform information that we have so many signals coming from next-gen WAF, coming from our content delivery will be able to be a real premium offering in this space, which is super exciting. We launched a managed service, security managed service, which is proactive and not reactive. We've had great response on that. And that revenue base is starting to grow where customers are starting to trust Fastly, not just for the security tech, but actually the monitoring and managing of that as well. Security track or security team's track record really has been phenomenal. The security proxy work that they launched for browsers. We picked up our third large browser customer this past quarter in Mozilla, and that's been very successful, too. So I feel, if we look -- when I look at our business, I look at content delivery and network services sort of our core and security as our growth with compute and observability incubation, I feel really good about the security business.

And then just a quick follow-up in terms of the potential financial impact as you lean more into security, do you have any aspirational goalpost in mind in terms of what percent of the business you would like to see come from security over the next couple of years? What be it your new bookings on a quarterly basis, be it the actual revenue mix of the business? And then secondly, just, Ron, if you could refresh your memory, I believe when Signal Sciences was acquired the gross margins were in that -- were said to be 80%-ish plus. And so remind us about the accretiveness of the security business to gross margins as it grows as a percentage of the mix.

[ I will take the first one ]... I think there's an opportunity for this to be. When I look at a customer -- individual customer that is well deployed in CDN and security. I think security can be 40% of their spend. In the fullness of time, I hope security winds up being 1/3 of our business, with compute coming behind and content delivery probably leading for some time, but there's no reason it couldn't be that large, yes.

And I think from a gross margin perspective, I mean I think that 80% certainly is a good number when it's running kind of by itself. When it's running on top of delivery, it is incremental revenue for some of the same traffic. And so you actually can see margins when it is part of a delivery solution being much higher, probably even in the [ '90s ]. I think where we really see benefits to gross margin is when we're selling across the platform, fully utilizing the platform, getting paid for multiple solutions on either that same traffic or on the same hardware where you start to see really that leverage our gross margins favorably.

Shifting gears to go-to-market and what you've been doing with the channel in packaging. Could you just quickly recap the initiatives you guys have leaned into there and just the early success you're seeing?

Yes. So on the packaging side, we made a move to build a lower friction motion for customers to buy Fastly technology. And in so, in doing that analysis, we found there were a handful of really tough feedback from our customers and prospective customers. It's hard to know which features to buy upfront when they want to make the initial purchase. And it can feel like a little risky to start buying fast and you don't know exactly what the bill will be at the end of the month. And so we instituted packaging, which allows people to buy CDN and edge compute for the first time as -- in what almost feels like a SaaS motion. It's a package with predictable billing, no overages unless you're significantly over your usage limit, and a suite of features that is most of what you'll need for practically everyone, everything you'll need. And we'll put that in the market. We got great response early on because of predictable pricing now. We're getting great response from our sales team, because it's easy to sell and easy to sell through the channel. Standard pricing on each package doesn't have to be custom negotiated standard discount strategy across the board. That's great. We're seeing them pick it up. This last quarter was a big success in terms of the packaging motion. And you can see it in the RPO results. Packaging is great business for us, because it's a monthly -- it's a monthly predictable bill, which shows up as an effectively a monthly commit from our customer, which is awesome. Second piece on the channel side, I've always believed in systems integration channel for a company like Fastly, and the reason is that Fastly is an edge platform. There's so much development work and systems integration where it can be done on the platform and that can be used to integrate into our customers' infrastructure, and that can be done faster and easier by systems integration partners. And so by bringing those partners on board, we get 2 benefits; customers that use them, can find the staff augmentation they need to ramp up Fastly usage faster. That tends to be good for, Ron loves that. It's good for our bottom line. But our partners, more than that, when they become experts at using Fastly, and we're really starting to see this now. They are finding new deals in their accounts and they're registering those deals with Fastly, which is great. That's starting to show some confidence from the channel. And yes, we continue to invest there. We'll push hard in '20, I think '23 is going to be important. '24 is going to be an important year for channel. We're going to push hard next year.

And on the channel, I know you're not disclosing what percentage of revenue comes from the channel. But you have said, I think in this past earnings call, you've seen a 3x in deal registrations from the channel year-to-date versus all the [ '22 ] and also 50% growth in channel revenue. I guess have you -- have we started to fully see the revenue impact from the 3x in deal registrations, or when does -- what is the lag? And when should we start to see the full effect? Because 50% revenue growth versus a 3x and deal regs, that's a bit of a variance between those 2 figures. So when do we start to see the full impact of the momentum you're seeing there on the bookings front?

Deals take between let's say, [ 4, 8 ] months to close. So that's what the lag is on a deal reg to close, and then the bookings ramp from there. So like there is a lag for sure. We're going to see it. It's going to -- you're going to see what 8- to 12-month lag on that stuff, but that's why we're trying to give you visibility, especially in partner activation. So like bringing partners on board, we closed Q3 with 55 partners on board, up from low-30s, I think. That result, that's a really solid leading indicator for me. Deal reg being the next indicator that those partners are being activated and bringing deals into the system, not just closing existing deals and of course, the revenue, but we all know the revenue is a lagging indicator. I hope that we're going to always be seeing this. We're going to be tracking deal registrations that hold a large amount of revenue that will be pushing their way through the system, and that number will always be growing.

And then just a quick follow-up on packages. I know you've talked about reducing the buying friction. Do you have any data yet where you can conclude that the average sales cycle is notably shorter for customers who buy packages, or what does that look like?

No, I'm going to get that data, but I don't... What I have right now are anecdotes. The one -- there's a -- I certainly have data on this. There is a friction that's removed with predictable billing for sure. So like more than half of the initial tranche of customer wins in those first 2 quarters, more than half of the customers mentioned predictable billing as a reason they went ahead, which is awesome. So that lowers the friction from the customer side. The sales motion process side, I'm going to find some data for you on that. I believe there's a huge win here in standard pricing standard discount, but yes, I don't have any data on it. It's a good question.

Got it. Ron, I want to come to you on gross margins. I know you have never faced questions on gross margins before. Could you quantify, I know you're facing some near-term headwinds from some ramping traffic in some less mature regions. Could you quantify the impact to gross margins in Q4?

Yes. Yes. So I think in Q4, those headwinds are probably -- it's about 100 basis point headwind in our gross margins. And it's really driven by increasing traffic in some of our smaller remote international regions where we've had very low traffic in the past. But this is part of a good process going forward because as we see this spike in traffic, we're able to go back into these markets, renegotiate our bandwidth rates. As we see traffic come up, we can increase peering and turn those from what a relatively higher gross margin or higher cost markets into more in line with what we're seeing across the platform and apply those gross margins to all the customers in that region.

And what -- I'm not sure exactly how many regions you're seeing that. But how many regions have you repriced those contracts thus far? And how many do you have to go?

There's not a specific number. I would say, certainly, we renegotiated one of those regions in the third quarter. Our Head of Infrastructure was in Southeast Asia, last week, renegotiating contracts there. I'd say there's a [ hand-fold ], and I think we'll work through those over the next quarter or 2. At that point, I think these headwinds tend to abate, and I think we're back on our trajectory of seeing our incremental 80% gross margins that we talked about at Investor Day.

Yes. So I want to come to some of those gross margin targets at the Investor Day. Just to be clear for investors as they look at their models, that 100 basis points impact you feel it will be fully abated by Q2 of next year.

Yes. It's sometime in the first half.

Okay. And then coming back to those targets you gave at the Investor Day, again, it was 80% incremental gross margins on a year-over-year basis going forward, understanding quarterly, you've got some seasonality that, that can mess with things. But -- and 65% gross margins by 2026. Are you guys still confident in achieving those?

Yes, we are. I think implicit in that 80% gross margin is one of the benefits of this traffic is it does allow us to optimize some of these more remote regions so that our global network is more efficient. As we drive new customer acquisition, expand the number of products that customers have, we talked earlier about the gross margins that security has, particularly where you have multiproduct deployments. And we continue to increase the diversity of our customer base. All of those are going to drive improved gross margins. And that's what drives the incremental 80% gross margins on a year-over-year basis that achieves that 65%-ish gross margins in 2026.

Maybe just [ rank order ], like what -- to get to that 65% target, what is -- what's the most important factor? What's the second most important factor between security between what those -- what you just named?

I think what I would say is underlying those are foundational is sort of the rigor and the forecasting we put into our overall network costs. Todd spoke about the efficiencies that we've been able to deploy via software. There's probably another round of those efficiencies that we're able to deploy into the software, continuing to increase peering and negotiate rates broadly are foundational to achieving these results. I would say then it really comes down to continuing to grow the business at the rates we spoke about, increasing the deployment of products across security and compute. So we get higher attach rates across multiple product deployments, this is kind of the order. And then I think continuing to diversify the verticals where we deliver product and building out the mid-market. Those are all contributors to that higher gross margins.

And as you look just step it down to OpEx and getting to those operating margin, free cash flow margin targets you laid out as well, where should we see the most operating leverage over the next couple of years?

Yes, you're going to see the biggest operating leverage in G&A. I think here what we've seen is efficiencies in terms of our overall operations and most of that, you're going to see in our general and administrative costs. I think that's where we're going to see the biggest leverage. But as we drive efficiencies in our operations, it is going to fund improved efficiency across product and engineering and sales and marketing, which helps fund the investments we're making in product and go-to-market.

Lastly, for you, Ron and Todd, I want to come back to you for a moment. You guys have been making some discounted repurchases of your convertible notes at a pretty rapid clip. I think you announced [ $240 million to $250 million ] repurchases just within the last week. I know you're being opportunistic there. Is there any -- like in terms of your net cash balance, or your gross cash balance where you don't want to go below.

I think we've spoken about kind of given our business, and of course, we expect in 2024 to be cash flow breakeven, that the right cash balance for us is probably between $300 million and $400 million. We're just under $400 million with these last repurchases. So we're -- I think we're in a good place in terms of where the repurchase are. Again, we looked at this sort of opportunistically, and it was a good opportunity to bring down at a big discount by more than half our convertible debt at a par savings of over $100 million. And so I think we feel really good about where we are. And I think the cash balance feels really good right now.

Yes. So Todd, I think for investors who have followed the story for a while or even myself and my counterparts, we've been hearing about edge compute for a long time. And I think there's still a lot of confusion out there about what exactly is edge compute, but also when does it actually become a meaningful portion of your business? Because today, it's not. And so what -- maybe try to help give some clarity what is edge computing? Why are you guys well positioned to address that market? And then when does it actually impact numbers? Because it seemed to be a thing that you and various other competitors have talked about for a while, but actually becoming a material driver seems to escape the various players out there?

Yes. Look, I think it's an incubating market. There's no doubt about it. Wishing that the growth of that kind of technology would happen in 2 years instead of 5 years doesn't, doesn't make it true, but it's an incredibly important piece here. If I think about the reason that people move to a technology like Fastly to deliver performance, it's deliver user experience and content delivery can only get you there for static content. Fastly has deployed the most dynamic, the most sort of real-time content delivery possible. In fact, we have technology that's border lines on edge compute that's embedded within our delivery network. And that gets a ton of personalization built, built into that. But to take that to the next step to get truly dynamic content, specifically personalization, for your web experience, what kind of content is recommended to you? What kind of purchases and products are recommended to you? How is your checkout experience or shopping card experience? What does it feel like on the high-tech side, what a real-time learnings feel like in true real-time, true performance side to look at the high, at the dynamic part of the Internet to deliver that with the same kind of performance as the static content, you need edge compute. And that is why we're so focused on it. That's why customers are moving to it. I'd love to give you like a date when I think it will be a substantial part of the business. But in the next 2 years, we are going to see an enormous transformation in this space. There's no doubt about it. It's why we're investing so much in it Fastly. It's why my competitors are as well. I wouldn't discount that. And it's why we're seeing this influx of customer acquisition. And that's how I measure that business. I measure them on customer acquisition, not on product releases or press releases. We are bringing customers onto that platform, and those customers will drive a significant revenue stream in the near future.

And in terms of the macro, I know you called out on this last earnings call, it showing up in some elongated sales cycles. Was that new? Had you guys not been seeing that up until Q3? Or it sounded like it got worse. I don't know, what did you see and what are you seeing?

Well, I got that question for your earnings call for the last 3. And I really haven't seen anything before that. There's a lot -- I know that there's a lot of weakness in like SMB customer base and maybe some of the mid-market, but we just don't have a lot of exposure there. And there's some exposure, there's some of our competitors saw it in like financials. We don't have a lot of exposure there to maybe telco again. Our customers tend to see Fastly, it's not a nice to have. It's not something that can be compromised. It's a must-have part of their infrastructure that drives their revenue, that drives their user experience. So we hadn't really seen anything. It was the first time I caught a hint of it and that was deal elongation right at the end of the quarter. We had 3 or 4 deals slip. They've all closed now, and it looks pretty clean. And we'll see maybe at the end of this quarter, I'll have an update for you, man. I don't know, but it's been -- our business has been pretty, pretty well sustained through the macro effects that other people are seeing.

Yes. It certainly appears that it has good. Well, Ron, Todd, thanks again for the time.

Thank you.

Appreciate it, and we'll go ahead and wrap things up here.

Thanks, everybody.

Thank you.