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

Let me just go through some disclosures for important disclosures. Please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If any questions, please reach out to your Morgan Stanley sales representative.

With that, why don't we do just sort of level set in terms of where we are in terms of the state of the business. You recently reported in Q4, which was a better level of execution. You guided 2025 revenue ahead of consensus, also positive. That said, this year, growth has fallen into the low single digits. Can you give us a sense of where you believe the team executed well this year? And where do you think you're looking for in terms of areas of improvement going into 2025?

Yes. I mean I think there's a couple of areas where we saw execution really improve as we got in the second half. particularly around our go-to-market motion. We brought in a new Chief Revenue Officer midyear. And over the course of 2024, Scott built out his leadership team and made a number of changes just in terms of our go-to-market effort. And we started to see some of the benefits of that with new enterprise customer adds in Q4, strengthening pipeline and increased bookings relative to the outlook for 2025. Additionally, we made a lot of progress on the product side. I'll let Kip talk a little bit about that. But I think that allows us to enter 2025 with a much stronger product portfolio and better platform integration.

Yes, absolutely. We focused a lot last year on the product side, on platform unification. And really integrating our products and services into a coherent platform for our customers. And that started to show up in terms of a growing corpus of customers who are taking multiple product lines from us. We also spent a lot of time last year improving our product development engine, if you will. And I think we saw acceleration through the year, concluding with launching 3 new products in the fourth quarter. And I think we'll be able to build on the improvements we made last year and have an accelerated pace of product delivery this year as well.

Yes. So accelerated pace of product innovation that drives hopefully a new product pipeline going in...

Absolutely. I mean if you think about last year, we started the year with 1 security product, the WAF. We're starting this year with 3 security products. So that gives our new CRO and his new leadership team a lot more to work with and to sell and a lot more engagements with our customers.

And probably from a customer perspective, you always sort of hitting the check box of capabilities that you need to compete.

That is exactly right. I mean, we had a number of our P-type situations last year where we didn't fully qualify due to some of the missing products that we've actually subsequently shipped. So we're pretty excited about that.

Awesome. Going to -- so maybe in terms of like some of the challenges that we saw in 2024, that was mostly concentrated in the top 10 customers and in your media business. Can we just sort of dive into what were the factors that these media customers were seeing that caused them to pull back on their spend on Fastly. And to what extent was that weaker revenue a function of lower pricing and not just lower volumes?

Yes, that's a good question. we did see -- it was really a bit of a surprise, I think, in early 2024. When a subset, maybe a handful of those top 10 customers who were very focused on profitability, made a change in some of their strategy. They had a trend of vendor consolidation and they reversed that and added an additional vendors, which resulted in a loss of traffic share. But more importantly, they did that to drive price downs as well. And so we saw both the impact of price downs without increased traffic, which we typically see at renewals as well as a dilution in terms of our traffic share in, again, a handful of these top customers, which really resulted in declines in revenue from our top 10. We saw while we want to reduce concentration, we saw that happen a lot quicker than we wanted with our top 10 customers, we exited 2023 at -- with them making up 40% of our revenue. We ended 2024 at 32%. What we did during the year, though as we built out, improved our engagement with these largest customers with the senior leadership team as well as the sales team. to improve our ongoing engagement across the year and really dramatically improved our visibility into not only what their expectations are around traffic. But what are their strategies, understanding how are they planning to manage their expenses -- and so with that better visibility, we have much better confidence in terms of our revenue outlook. And additionally, we've had the ability, in some instances, actually influence favorably our share of their traffic through that engagement.

Yes. Just quick follow-ups there. In terms of the mix sort of in the low 30s, is that kind of like sort of a near-term trough in terms of how we should think about that mix from that segment of the business. And then from an operational perspective, you touched on it just a little bit, but sort of go into detail of what's driving that improved visibility with those specific customers?

Yes. I think in terms of the concentration, I think that's a healthy level. I think as we start to see stability in those customers and maybe even a little bit of return to growth in the second half, we're going to see that stay kind of in the low to maybe mid-30s in the near term. And I think that visibility increase really has to do with the engagement that we're -- we have with those customers, building relationships with their senior team engaging regularly with them to understand what's their strategy, how can we help them achieve their goals, if they are focused on profitability or if they're focused on new markets being involved early in that process.

You mentioned the sort of pendulum swing between consolidation and then back to multi-sourcing, if you will. If you look at what's happening in the category, there are now fewer players in the delivery market Edgio most recently exiting the market. How do you see this dynamic of a fewer number of players impacting Fastly going forward with respect to both pricing and the ability to grow further share in the category.

I [indiscernible] from consolidation. I think first and foremost, I think it does drive pricing stability. I think the fewer players that you have. And there was also Edgio had a lot of capacity, and I think that drove some fairly significant discounting. And I think as we enter 2021, we're going to see a lot more pricing stability. I think our outlook is for pricing to be more in line with what we've seen over a longer period of time back to sort of normal trends. . I think the other opportunity with leaving the market is an opportunity for us to gain some of their customers that are looking for a new customer. We've signed a few in the fourth quarter. We're engaged in negotiation with a number of those customers to bring on new logos that were Edgio. And in the case of multi-vendor, there's a really good opportunity to capture a portion of their share of traffic.

That makes a little sense. I think I'm going to jump on -- Maybe I'll direct it at Kip since you could probably have a perspective. And then Ron, add-in for your color, but when we think about packaged deals being up 150% in 2024, it's up 60% in Q4. Can you tell us what's included in these package deals? Who's the target customers for these package? And how do you expect this pricing and packaging strategy to help the overall growth equation at Fastly?

Sure. We have had 2 generations of packages at Fastly, the second generation that we had I launched actually one of the first things I did when I got to fastly about a year ago. It was really oriented towards simplifying the purchase of the products. In a lot of cases, it gives our customers more of a fixed price, predictable pricing for them, which is also predictable for us, which is great. It's also helped drive adoption of our Compute product line, in particular, as a lot of the packages cross the different product lines. We've seen great -- I mean you mentioned some of the numbers. I mean we've seen good results from this initiative. And it's really strengthened our conviction that making it lower friction and easier to consume our offerings is key to our growth. And in fact, with the 3 products I mentioned earlier that we launched last quarter, we started launching all of our products with a self-service capability where you can actually go to the website, sign up with a credit card and start consuming the value more or less immediately. And that's something that we've seen a good response to as well. So I put those packages in kind of a broader umbrella of things we're doing to make it much easier and lower friction to do business with us, and we're going to continue focusing on that because it's been particularly fruitful for us and helps us grow the customers that are not the top 10 type of customers, but the customers where they're consuming multiple product lines, and we have a stickier relationship.

Yes. It also seems to help unlock the new logo acquisition.

Absolutely. No. It's so much easier for a customer to just pick up a package and start running with that. And so you'll see more and more from us along those lines.

And just sort of to tie the bow on the 3 new products, for those less familiar with the Fastly story, can you describe what those 3 products are?

Yes, sure. Last quarter, we launched Fastly Object Storage, which is the object storage capability. We've actually seen great uptake on that through the self-service channel in particular. We launched our DDoS offering, which takes some really world-class DDoS technology that we've developed over the last decade of running a CDN. I like to say the same technology that protects our CDN clients like The New York Times at like on election night as you can now get as a Fastly customer for your own services. So that's been huge. And we also launched our Fastly AI Accelerator as well last quarter with support for open AI, Google, Gemini, as well as Microsoft large language models. So as I said earlier, I think it's a reflection of the accelerated product delivery that we've established at Fastly.

While we were talking about how this packaging strategy kind of improves the sort of visibility and sort of budgeting process for your end customers. What about for Fastly itself? Like can you describe what you're doing on the pricing and contracting side to improve visibility into Fastly's business and in quarterly revenues?

Yes. I mean in addition to the packages, which give us a SaaS-like revenue stream that makes it easy to predict, we have with some of our customers had annual commitments. And we're putting a lot of focus in terms of aligning those commitments with what we think the revenue expectations are. So there's a tight alignment, adding multiyear commitments to those customers as well because that additionally provides better predictability -- it also helps drive customer engagement in terms of how many products they buy. We like to tie commitments if they're looking for discounts so that we can get that trade-off. And as we continue to increase packages and drive more commitment, that's going to substantially increase just the predictability and longer-term visibility that we have for revenue.

Yes. That makes total sense. Let's kind of dive deep into the product strategy. First, on from like an executive standpoint, there's been a number of executive changes in the past 2 years. Kip you came on board last year. Can you share what your top priorities have been since joining Fastly about a year ago?

No, absolutely. And I joined in January of '24. I spent a lot of time traveling to visit our customers. And I certainly spent a lot of time with the internal teams, but I wanted to hear it straight from our customers what their impressions and experiences with Fastly were. I will say, I -- some of the most remarkable meetings that I had with customers during that time were meetings where the customers where agree with us for not selling them more, which in my several decades of experience I had not encountered. I've been in a number of difficult customer situations, but that was a unique difficulty. They literally like we have budget and we want to do this, how can we do more with you. We love your technology. And as I dug into things, really my impression from before I joined that Fastly was a company with an incredible technology platform and an amazing engineering team that was in my view, not earning the market share it deserved was kind of validated. And as I dug into my own organization, I lead the product function. I made a number of changes and in fact, replaced all of the leaders in the product organization, the middle of last year and took advantage of the restructuring that we have as a company to really make some accelerated change, I'll say, in the product organization. We spent time focused on some of the fundamentals, having a much stronger, a more structured product development process. I saw incredible innovation and incredible energy, but not necessarily as focused on solving important customer problems where there's really a significant differentiated opportunity for us as a company. So we've retooled the way that we manage the portfolio and the way that we manage our product delivery. And like I said earlier, that's resulted in opportunity now to deliver more products more quickly into the market. And we tripled our number of security products through the year from, as I mentioned earlier, from 1 to 3, which really gives us a much more complete security capability. And we've really, I think, established an accelerated cadence with the launches and with our go-to-market organization. That's -- I mean, that's my area, but I think the change with our new Chief Revenue Officer, is a huge piece of the puzzle. I think Ron mentioned that earlier, but midway through last year, we had a new Chief Revenue Officer come in. He's fantastic. He's actually got his new leadership team in place now and was able to put in compensation plan that he wanted for this year, which emphasizes some of the cross-sell and new logo acquisition more. And he brings a much stronger security focus. So I think that's going to be super helpful for us this year.

Let's talk a little bit more about Scott, your new Chief Revenue Officer, sort of kind of pre and post, you spoke to like a philosophical change in the product organization being more focused on solving customer problems. What's the philosophical change that Scott is going to bring to bear in the organization?

Yes. I mean I think one of the areas, and Kip touched on it really is a security focus. He has a lot of experience selling security. And I think as the -- our space has really moved to more of a platform sale where security is a key part of it. I think that's part of the overall sales effort. And I think that's going to have a big impact I think bringing in a team that has that experience. And additionally, I think as we've sort of evolved focusing on new customer acquisition and that bringing in new customers outside of kind of that top tier where we had a lot of early success is now building out a diverse customer base with discipline, strong focus on aligning where we have resources with the opportunity in that region, which means redeploying salespeople in some areas to really focus on what that opportunity is in that region is going to drive, I think, efficiency in that sales organization and return. And as I indicated earlier, we're starting to see some of that impact in terms of the stronger pipeline, stronger bookings, as well as an acceleration in customer acquisition.

And it makes total sense. Let's talk a little spend some time on the security opportunity. In terms of just sort of level setting, how many customers you're taking on the security projects today and can Signal Science WAF, be a material driver of customer expansion going forward.

No. I mean, definitely, although I will say Signal Science WAF now just part of our security portfolio and one of the nice things about the 2 additional security products bot mitigation and DDoS protection that we launched last year is that those are built organically by Fastly. So it demonstrates our ability not just with the WAF that we acquired. It's a fantastic product that we acquired from Signal Science but our own team is developing new security products going forward. I track a number of metrics related to our platform strategy and because that's really the foundation of what we're trying to do. In many ways, even more important than our security strategy, our compute strategy or delivery strategies, how to solve it together. And one of the things I track is what percentage of Fastly's customers are using more than one product line. And that number, we don't share it externally. That may -- maybe we will at some point in the future, but we don't share it externally at this point. But that number moved materially up last year. very much in favor of security being picked up by more of our customers. And I can tell you that number is well above 50% now. So the majority of Fastly's customers now take multiple product lines. And that's so important to us because it shows the cross-sell and the value of the platform, but also because customers who take more than one product line have really materially lower churn and a materially lower price compression or price pressure. So it really enables us to build a much, much stronger foundation of our business. The other number I track with respect to security and our -- really our platform strategy overall, is our new logo acquisition. And for each new logo, for each new customer that comes to Fastly which product line or product lines are they taking in that very first order where they become a Fastly customer. And historically, I think if I went back a year ago, that was predominantly delivery. When I look at those numbers now, it's far more balanced. That was actually funny. I had someone on my team who focuses on the delivery business. disappointed that the percentage of new logos who took delivery as part of their initial purchase with Fastly had dropped. And I pointed out to him that this just meant that he could cross-sell some of the security customers just as security has been cross-selling our delivery customers. So we're seeing that really play a bigger role. And I don't think we break it out externally, but if you were to set aside those top 10 customers and those large streamers and downloaders and look at what I would call more of the core of our business. The percentage of security revenue is substantially higher than the percentage of the overall company just because we have the majority of those customers taking multiple product lines.

Yes, that makes a ton of sense. If I kind of take you back to the 2023 Investor Day, and where you sort of outline where the product road map is headed. Frankly, you've seen a lot of execution against that product road map there. Is there anything...

More to come.

Yes. That's essentially the question is like with respect to like API security or anything else what other capabilities are you looking to bring to market to drive to those customers that want to buy more products from you? What are sort of the focus areas on the product roadmap?

Yes, absolutely. I mean that Investor Day was before I was with Fastly, but it was prescient and it's wonderful to be able to deliver on that and API resilience, I think, was mentioned there. And you certainly -- we have customers using our platform today for that. But similar to the story on DDoS, I think there's an opportunity for us to package some world-class technology and deliver that to our customers. We have some of those security products. I hesitate to launch security products that -- or launch products at an investor conference. But we will have...

First time for everything.

We will have to -- we're experimenting and marketing, but I'm not sure that's the experiment we want to run. But we'll be launching some new products very soon. In terms of focus areas, one thing that's kind of popped in the last couple of quarters for us, and I think is an area that you'll see us doubling down on is a security technology called Deception. And when you look at our DDoS capabilities that, as I mentioned earlier, grew up over our decade-plus of CDN experience. We take pains when we block a DDoS attack to make sure the attacker thinks that it's not blocked and that the attack is working. And that's really critical because the they waste their time and resources on that. They don't try something else or try something different. And the research is now coming out showing that, that's an incredibly compelling capability more broadly in security. And we have some expertise in that area, and I think you'll see that is a theme across our security offerings just because it's incredibly effective. And in fact, research has shown that attackers hesitate to attack targets if they even think there's Deception deployed because they can end up attacking something that's a fake target, they can end up thinking their attack is working and spending their time and resources, and it's just makes it so much more difficult for them that they will often simply pick a different target. And so we love that for our customers.

Super interesting, super exciting. Maybe sticking on the FEMA security. When I mean been a tech analyst for a while. When you reading security, the channel and partner seems to be like an important piece of the puzzle to solve. So maybe talk about some of the changes you've made to the channel partner program to build your relationships with the important sort of elements of the security channel to help drive that cross-sell, upsell motion.

Yes. I mean we had a very strong channel presence that we sort of inherited with Signal Sciences around security and WAF. And over the last year or 2, we've engaged in training that channel to sell more than just security to also be able to deliver our self-delivery. So the channel remains an important piece of kind of the strategy. Again, focused on training of that team and really focused on kind of the deal reg that's coming out of that to be able to actually convert those into customers.

Awesome. So we've talked about security. And I guess the -- and to Kip's point, I mean, ultimately, this is just a drive like a broader unification across all of these opportunities. But if we think about the edge compute service opportunity, how do you guys think about that opportunity today I've been talking about a lot in the ecosystem. When is it going to be the year of edge compute has been kind of a famous question. But where do you see Fastly sort of differentiated and the sort of edge compute serverless opportunity.

Yes, absolutely. We have, right now, I think, an incredible serverless edge compute opportunity. We were pioneers in using the web assembly or Wasm technology to provide a language agnostic, high-performance, highly secure edge compute capability. And we have a lot of customers using it to great effect. I think New Relic is a great public reference customer. I would note that we actually use that same compute facility that we make available to our customers to build a lot of our product. So the AI accelerator is actually built on top of that or WAF, for instance, is actually built on top of that technology as well. So we're super proud of that. And it's not a big part of our business, but we've seen growth rates in excess of 50% for the last 2 years in our compute business. And I think the addition of Fastly object storage last quarter will complement that and open up additional use cases for us there and also provide even more connectivity between our delivery service and the edge compute. That said, we run into customers who have workloads that don't lend themselves to a serverless modeling longer running process workloads and things like that, that they still really want to be able to run in our high-performance distributed infrastructure on our network close to users. And we've been exploring models around that with those customers and in fact, we've have been running a workload like that in production for one of our large customers now for some time. And we're exploring how we could productize that and make that available so that more customers could run more kinds of workloads in the Fastly network and on our platform.

Can -- in terms of trying to connect the dots, you've just rolled out an object storage capability you've had an edge compute business has been growing 50%. I mean, are we headed towards like an edge inferencing opportunity that you can monetize in terms of as AI gets switched the edge, how are you thinking about edge inference for AI workloads?

Yes, absolutely. I mean we -- we have a lot of CPU resources at the edge. I mean we run a single platform. So the same server and network infrastructure that is delivering videos or downloads or e-commerce sites or running our security products also run our compute workloads. So we have a lot of capacity in terms of processing power there, and we have a lot of customers looking at using it for various forms of AI workloads, whether it's our own AI accelerator product or things that they're developing themselves. And I think what we're finding is the edge is just very useful just as it's been useful for security, just has been useful for content delivery. There's use cases in compute around personalizing content and so forth. There's use cases in AI as well. I think a really interesting example is something one of our customers is doing with Fastly Object Storage. They have a large multi-tens of petabytes, image library. And they have a new business model with their image library, which is licensing it to people for model training. And you think about Fastly's network and our platform being close to end users and devices, well, it's also incredibly well connected with all of the clouds in the world because that's where applications and content often originate. So they came to us and they're storing that image library with Fastly so the people that they license it to do training in the cloud could have a high-performance access to that library at very good economics. And so I think we're seeing the edge will play a significant role in AI. And we're working with kind of pioneer customers, if you will, on various use cases and how they apply our platform. And we'll be evolving our platform, whether it's the underlying capabilities of the software to enable those workloads.

Just a follow-up question to Kip's answer. Like when will we see opportunities that look interesting over a long time horizon. As CFO, how do you think about the capital allocation decision to fund those opportunities versus what kind of like, I would say, security like that needs to be really successful, right, to get where we want to go to in terms of growth and margins. And so the trade-off decisions in terms of capital allocation decisions for these incubation opportunities versus the core. How do you think through that?

Yes. I mean, I think when it comes to -- from a road map in terms of development, we have a really good process around kind of what the opportunity and use cases are I think what we've built is a platform that allows customers to kind of apply kind of their own use cases. And I think when it comes to actually deploying it from a capability -- we have, over the last couple of years, really improved our ability to deploy that fairly quickly. So we can actually do much better planning. And as we see the opportunity, we can deploy the capital to meet that need. And I think if we have the actual demand, that's one of the things I'm looking for is do we have the contract? Do we have the deal? And then we can support it with the needs once we have it rather than having to invest well ahead of that, as I think we may have done in the past. And we're looking to get better at being able to do that more real time so that we're really investing in line with actual traffic, actual business that's there and we can meet those needs.

Yes. I mean I'll say one thing that helps us is unified software-defined infrastructure. So other words, we launched AI accelerator, there wasn't like discussion about a capital investment because it's a new workload on our existing platform. Now as that grows, maybe it drives more needs for our platform, but that capacity if AI accelerators not being used -- can be used to deliver videos or to process other things or do security workloads. And so it's one of the advantages we get our unified software-defined platform gives us more agility in terms of pursuing new opportunities.

Yes. Awesome. That's a great point. We've been talking about kind of the strategic opportunities and the product late opportunities. Let's get back more towards the near term in terms of the outlook. Q4 revenues came in better fiscal year '25 guidance also came above expectations. Ron, can you talk to us about the growth drivers of the next 12 months that support that above consensus guide that you reported a couple of weeks ago. And particularly in the full year outlook, why are you expecting to see an acceleration in the second half?

Yes. I think the drivers are -- we talked a little bit about that is the improvements we've made in our go-to-market motion. The -- entering the year with a much broader complete security portfolio, but above and beyond that, I think we've got better visibility. The increase in enterprise customers. As we look at those tend to add revenue over time. And those will start to add meaningful revenue as we get into the second and third and fourth quarters. So that customer acquisitions are going to drive some of that accelerated growth in the second half. Additionally, with the bankruptcy of Edgio, I talked about the opportunity to gain share from those customers. Some of those customers as they meet commitments with their -- one of our competitors bought their contracts as they meet those commitments. There's an opportunity for us to gain those traffic. So those will come on board across the full year. And as we get into the second half, given the headwinds and challenges we saw in some of our larger customers, you start to get the easier comparison. So you can see within our guide, which we feel that as we get to into the second half, we can exit at close to a 10% growth rate is implicit in our outlook.

Yes, makes total sense. Going back to that Investor Day in 2023, maybe this is where we can wrap up. You laid out a 3-year operating plan targeting revenue of $800 million, $900 million gross margins, approximately 65% op margins. about 7% calendar '26. Given where you stand today entering into 2025, what's the latest thinking on how achievable that model is today? And just maybe sort of comment on where we are ?

I think the model and the incremental model around our gross margins and the investment we're making in the business are still intact. I think given some of the headwinds that we saw from some of our largest customers, the time line has changed for that you pushed it out, I think, a year or 2. But I think the model is still intact. But we do see incremental gross margins of north of 70% in the medium to long term, driving margins north of 60%. We're continuing to focus on improving our operating margins. We're going to reduce that operating loss by more than half at our current outlook, and we're hoping to do better than that this year. That same trajectory would get us to profitability in 2026. And we remain extremely disciplined on managing to bottom line results given the revenue visibility that we have.

Awesome. With that, we're at the at the end of time. Thank you so much Ron. Thank you so much, Kip, for joining us. to talk to us about the product strategy, and thank you for giving us the update on Fastly. Really appreciate it.

Thank you.

Thank you.