DigitalOcean Holdings, Inc. (DOCN) Earnings Call Transcript & Summary

Awesome. My name is Josh Baer, software analyst here at Morgan Stanley, and we're joined today by the CEO of DigitalOcean, Yancey Spruill; and Bill Sorenson, the CFO. Before we get started, some disclosures. For important disclosures, please see the Morgan Stanley research disclosure website, www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representatives.

So we're coming up on a year since IPO. I think DigitalOcean today, looks a lot different than 2 or 3 years ago in many ways. And so I was hoping you could start and talk through some of the biggest changes that have caused really an acceleration and an improvement across most of the key metrics.

Well, we're growing in the upper 30s or mid-30s for all of last year versus mid-20s. And we've done that through a combination of taking net dollar retention across our customer base, which was basically flat the prior year up to the mid-teens, upper teens exiting the year. Our ARPU grew from upper teens to approaching 30%. Our customer growth got in the upper single digit from lower single digit in the prior year. So accelerated customer growth, accelerated revenue per customer and an accelerated net dollar retention, which for me is sort of the gold standard for a subscription business. And we exited the year approaching $500 million in revenue. On the -- we call that grow smarter, which when we came in 3 years ago was really to take a business that was growing mid-20s and accelerated to growing faster than our market, which is an upper 20s business for SMB and developers. The second piece is we transformed the efficiency and scalability of the business by taking EBITDA margins essentially from low 20s to 30-plus percent, and took free cash flow from negative something ugly, and we don't want to talk about it, to 6%. So growth plus profitability, which was 6% in -- on the Rule of 40 was over 40 in a year. So transformed that by [ 3,100 ]. So that's what we call efficient growth, which is what we call grow smarter, build efficiency and scale in everything we do.

Perfect. And Yancey, thinking ahead into in the coming years? What are some of your key priorities?

So we still want to sustain, obviously, really high growth. We've targeted our first $1 billion in 2024, which we're well on track to get there. How we do that is we're focusing on building out some sales capabilities to supplement self-serve, outbound sales and a channel and a partnership capability. We've sort of added sales from nothing a couple of years ago to 3% of total revenue last year, mostly focused on farming our cohort and inside sales, if you will, to the cohort, but we're adding outbound sales and partner. We think that we can get sales into the upper single digits over the next several years, lots of opportunity. It's a $70 billion market with 100 million SMBs around the world. 30 million software developers, massive market. Highly fragmented, which is why the channel is going to be such a critical piece for us. We're also investing in product. We brought in a new Chief Product Officer, kind of reengineered a little bit our org structure to get more velocity on product innovation. We've kind of demonstrated by launching the database as a service, our Kubernetes as a Service, our marketplace that we could take that from 0 to over 10% of revenue in just 2 years. So there's a lot of adoption as these customers start small, launch an idea into a business and start to scale very rapidly that the multiproduct adoption is a big opportunity for us. And so those are the 2 priorities to sustain where we're at or even higher growth.

Great. I want to dig into all of that. And maybe focusing on TAM, massive public cloud TAM. Which segment of the market are you addressing? What does your typical customer look like? And also how has your product portfolio evolved to address more of that TAM?

Well, there's public estimates of 30 million software developers going to 50 million by the end of this decade, 50 million, 100 million small, medium-sized businesses around the world and 40 million net new are created every year. Collectively, that group spends $70 billion on cloud infrastructure today, and that's projected to go well into the $100-plus billion over the next several years. So it's a massive -- but how you get there is really small numbers. And so our principal route to market is we bring millions of people to the website a month, tens of thousands sign up. Frequently, they start at $10 to $15 a month. And like we cited in the case study in the earnings call 2 weeks ago, we had a customer 8-person start-up in Tel Aviv 8 years ago that came to the site, started $20 a month. Last month, they're $170,000. So a $2 million run rate over 8 years. They have a 400-person business that they've built on DigitalOcean platform. So that's a long tail cycle, how we nurture, germinate these customers, incubate them, if you will, and then they grow. And then those customers now are driving 85%, 84% of our total revenue. So 100,000 of our 610,000 customers are -- they take that journey. They start small. They're testing an idea. They're figuring things out. We have a low acquisition cost model to get them here. We keep them through our documentation, our support, our simple, easy-to-use service. It's low cost. We offer support to every customer, so that's highly differentiated from the other cloud providers. And we're all open source, so we're not trying to lock them into some kind of proprietary tech stack. Make it simple and easy for them to stay and to test and to grow and to scale when they're ready to scale. So that's our customer. We're now supplementing that with -- as we build out a more complete cloud set of capabilities with the PaaS services that I mentioned, we're able to bring those customers in day 1, and so that's what we're supplementing with the sales capability. So it's budding entrepreneurs. We have customers in every country that we can legally sell in the world. 70% our revenues outside the U.S. If you were to look at a heat map of our customers, our revenue will look like global GDP. And we have a low x cost way to access those customers. And obviously, we're getting very efficient at delivering for them as proxy by free cash flow generating today.

You mentioned your international exposure. With the war in Ukraine top of mind, just wanted to ask how investors should think about the exposure to Russia and Ukraine and any impact for you.

Yes. Well, it's an awful situation, so we should think about it like that. We have about 1.5% of our revenues in Ukraine. It's growing much faster than the total company. And we have a little bit more than that in Russia, growing slower than the total company. As of now, we are working with the folks in Ukraine, frankly, and in Russia, we haven't shut anybody off. Our position is people use our platform for force for good. And the individuals, the entrepreneurs in Russia aren't the government. That's our posture today. That could change, obviously. And in Ukraine, we're doing what we can to support the customers through a challenging time.

And the other angle to the exposure question is around the investor base. And I just want to kind of throw that out if there's anything to address as far as some of your...

Well, our largest investor is Access Industries, which is a family office for Len Blavatnik. And Len is from Kyiv. He came to the United States from the USSR in 1977. He's not Russian, he's Ukrainian. We should all know the distinction on that right now. He is educated in the United States, made his money in the United States. He's an American citizen who left an ugly place called the USSR and came here, and has obviously realized an incredible dream. He's a citizen here. He's a citizen in England. And the Queen of England has knighted him. And all the money in Access Industries is his own. It's a family office. And he's been incredibly supportive of DigitalOcean since the Series B.

Excellent.

So he's American. Made his money in 90s, long, long time ago before all of these folks that are in power now came to power.

Great. You had touched on some of the differentiation between some of the hyperscalers. I was hoping to double click on differentiation versus AWS and some others, and also to throw in pricing into the mix, and if you could provide some insight into how you're able to offer such a lower price compared to some of the hyperscalers.

Well, we differentiate on 4 key areas with the hyperscalers. It's simplicity -- simple, easy, intuitive. In the time we've been on the stage here, you could be up and running coding on the platform. Actually, for about 15 minutes or so, you could have been coding on the platform. So it's really easy, which is critically important. I know a lot of people that we talk to on the buy side think simplicity is a buzzword, but it really is. And if you talk with our or any start-up, the fact that whatever product they engage with is easy to use is like a really, really important feature to them. They don't have an IT department. They don't have a DevOps department that they can roll back on. The product experience has to stand on its own. We also provide documentation. Tens of thousands of digital documents, tutorials that drives many millions of people to the website a month. Even customers at the large hyperscalers who don't even -- and customers of ours come to our website to learn how to do things in open-source software. So that's a huge differentiator. Again, no DevOps, no IT, no expertise around it. So documentation is really important to them. We offer support to every customer. And so again, we have many customers that come to us from the hyperscalers, $2 million, $5 million business growing really rapidly, not spending enough to matter. And as their business evolves, they want the help to support the documentation. So that's a big differentiator. And then we're all open source. So we're not locking people into a tech solution. It's just simple and easy to use. And then we're low price. The average customer that comes over here from AWS, Azure, et cetera, saving 50% and not losing any performance. And you could see Stack Overflow and -- one of the major technology reviewers, sort of puts us pretty high on the performance scale relative to the hyperscaler for the use case that our folks need for us.

And so on the pricing differential, are you able to offer that because of your efficient self-service model, on your co-location strategy?

Yes. I think we have a more targeted set of capabilities. So we don't -- our product list isn't extensive, it never will be extensive. Part of simplicity is not overwhelming people with product that they don't need. It's relevant product. We have a set of data center and infrastructure that we manage increasingly well that's not as broad. We're not -- we have an infrastructure and product set and a go-to-market that's targeted towards a -- the use cases of small businesses and developers is much tighter than the broad enterprise customers that the hyperscalers sort of focus on. And so as a result, we could be a lot more efficient in where we're investing. And so that's why our prices are where they are. And it's not clear. I think a lot of times people ask that question, and there's some assumption that AWS is a cost-plus pricer. We don't have costs as low as AWS, so it's not that. It's a strategic decision for them to price the way they price. We do know that when people come over from those folks, they're saving 50% on the bill.

Great. I wanted to ask one on customer acquisition. You talked about near-term returning toward high single-digit growth as far as customer counts and eventually targeting 10%. And so I guess I'm wondering what's needed to get to that level? What gets you confidence -- what gives you confidence to get there?

We were upper single digits last year. We kind of ran into some headroom -- headwinds as we -- correlate it a lot with cryptocurrency getting into the 60-plus thousand, a lot of fraud, a lot of bad actors coming to the platform. And so we normally have a run rate of people. We are the Internet in some ways to think about our company. People come here to start test ideas and start companies, and there's bad side of the Internet. So we routinely are kicking a couple of thousand people off a month that was elevated in the middle of the year in Q4. And so we have changed our protocols, et cetera. And we said we would outrun that. It's not unusual for things to surge and sway on the dark side of the web. And so we have a good process. We have a significant investment in security. That's a core value proposition that differentiates us on the lower end. We haven't talked about Vultr and OVH, but we have a significant investment in security and fraud protection for our customers, which is really important. So that brought customer acquisition down, but we'll outrun that. And pretty confident in that. We'll do that by our core strategy around organic content generation websites and continue to use better filters, new features like we launched Google Pay in Q4, which is accelerating customers. So there's a number of things that we can do to sort of outrun that and comfortable that we'll get back to a higher level of growth. But again, aggregate customer growth is in the story because of the customer journey that I talked about. 100,000 of 610,000 drive 84%, 85% of the economics in terms of top line revenue of the company because we incubate so many small companies, and so that's growing 24%. And I think increasingly -- and we look forward to the Investor Day we have more time to talk with you all about that dynamic. That's unusual for a company to have customers take such a significant journey that drives the economics of the business. But we're still comfortable we can get that upper single digits back to double digits at some point.

Great. And thinking about your largely SMB customer base and this economic environment, how do you think about potential impact to that base in the event of recession? Or -- you know, yes.

Yes. You know what's interesting? When we were last here, 2 years ago this week as we were shutting down and everybody is freaking out, we came to the conference out here and we were getting to know folks as a private company and they said, "Oh, you guys are going to 0. This pandemic is going to be awful for SMBs. They're all -- they're not going to pay you. They're all going to go to business and you're done." And that churn was 17%, 18%, now it's 10%. Net dollar retention was 99%, now it's 116%, 117%. Growth was 24%, and now it's 35%, 37%. What we found has been very amazing about the resilience of small businesses and how misunderstood or not understood they are sort of in the broader market. It's half of GDP. And so we didn't really skip a beat. In fact, our platform -- if you talk to our customers, most of them are running remote businesses. So you talk to the founder in Pakistan and his co-founders in Singapore, the engineers are somewhere in Europe, and so they didn't have a lot of the disruption. They're running digital businesses, tech-enabled businesses, a lot of media streaming, ad serving, crypto, blockchain, quantitatively intensive business models which have all done well. And we have a virtual infrastructure. And most of our customers, if not all of them, even through sales, it's usually virtual, the conversation. So our business operates in the Internet. And our company was 60% remote going into all this, and we kind of transitioned to remote first for everybody. And so we didn't have the disruption that was perceived and the acceleration in our business and the resiliency during a pandemic. We're not going to see, I hope, a worse economic environment than we saw 2 years ago in terms of mass unemployment, uncertainty, all the rest of it. And I think our customer base came through. And that's not too unusual. I think if you look at GoDaddy, Wix, some of these other small businesses that were formed in the middle 2000s, they outran the financial crisis. I think cloud actually unleashes small business, it doesn't inhibit them.

Great. I want to loop Bill into the conversation. Can you talk about capital intensity, where CapEx as a percent of revenue is today, where it can go, how that influences the free cash flow margin, and ultimately, where there is room for leverage in the model and efficiency?

So when we started here, the picture was pretty bleak. And we were spending about 50% of our revenue in terms of funding the CapEx, and we were being told numbers that suggested we were going to have to raise hundreds of millions of dollars to build out server farms. And as we learn more and more about the business, we were able to basically take a longer-term view in terms of our planning, which then allowed us to work with our major suppliers, server manufacturers, the chip manufacturers to basically give them some idea of our growth, which led to improved pricing for us across the board. Before, we were [ paying ] retail plus because we were buying just in time, and now we were taking a longer-term view. The other thing that we've been doing technologically is looking at the capacity of the servers themselves and how they're utilized, and we're finding more ways that we can effectively increase the overall output, revenue output of those individual servers. And third, we're taking advantage of technological advancements that are improving the overall servers themselves as we invest in newer and newer models of the servers and the guts within them, we're finding additional benefits. So all of those things combined has basically cut the payback on these servers to -- by 50%. Most of our servers are paid back within 6 to 8 months from when we purchased them, and then we depreciate them over 5 years, and we run them longer. And now we're in the process of replacing our overall fleet, a maintenance program which we began 2 years ago, which will be replacing everything over the next 3 to 4 years, embedded in our numbers, which will lead to incremental yield relative to that. What does this all mean? CapEx as a percentage of revenues this year came down around 25%, where -- implicit in our guidance is a low 20% type of number. We see it going below 20% as we go out beyond next year. But the most important thing and one we continue to emphasize is while we're making material capital investments, we are cash flow positive today. Last year, 6%. This year, we provided guidance of 8% to 10%. As we repeatedly say, we like to beat and raise. And we're doing everything we can to continue to increase that. And part of our long-term goal that Yancey referred to earlier of 2024 of $1 billion of revenues, match that up with $200 million of free cash flow. And we think those are things that make the performance of our business, while there may be a capital-intensive issue to it, it's still a big cash flow generator.

Great. I'll ask a couple more and then see if there's any questions from the audience. So with that, CapEx framework in mind, how should we think about your data center expansion strategy?

Well, we've identified one region we'll be expanding into this year. And we have 2 other regions on the calendar for some time frame after that. We're looking at basically where we feel a closer proximity to the customer is going to make a difference. And so we see real opportunities to grow there. Plus there as well, we're talking to the major partners that we have in terms of our growth plans and their growth plans. So we think there's real expansion opportunities for us overseas as we continue to build out and complement the current customer base. I mean I think it's one of the fascinating things, at least to me when we joined the company, we're in 185 countries. Every country where we legally can do business -- so we can't be in North Korea, we can't be in Syria, places like that, but we have customers everywhere. So no matter where growth comes in entrepreneurship or the small end of the market or the developer that you just heard, Peter McKay, talk about, we'll be there to capture that revenue. So we'll be getting closer and closer to them as we move forward.

Excellent. Are there any questions?

I was curious if edge compute was an opportunity that DigitalOcean considered branching into, given the data center footprint that you already have and in light of the recently announced Akamai acquisition of Linode.

I'm sorry, I didn't -- I missed the first part of that.

Yes, just whether edge compute was an opportunity you consider branching into.

Yes. Absolutely, yes. So we are -- we have a partnership with Cloudflare on our serverless. We have a partnership with Stack Overflow. So we're in the edge. We offer that opportunity for customers for improved performance for static sites, et cetera. And that's definitely one of our priority areas. We don't have a long list of what we need on the product side. But that is one area we need a more comprehensive -- that we're looking at alternatives currently to enhance and have a broader sort of comprehensive capability to sort of merge the core infrastructure as a service globally, highly performant, compute-intensive with edge capabilities to serve, with a lot of our business models, the media streaming, education, ad serving want both. And so we're looking to evolve -- so that is a priority area for us.

Question upfront.

Now I know you talked about churn going from 17% and 18% to 10%. I think you've called out before that the first 90 to 100 days, customers are most likely to churn. So I think you've also talked about kind of engaging with them more in that time period. Just wondering what are the specific things that you've done in that window of time? And like why do you think that's working?

Yes. We talked -- we throw these numbers around 17% to 10%, it's all very interesting. We're an SMB-focused company. I mean 75% to 80% is sort of middle of the fairway. So when you're talking about 90% gross retention in SMB, pretty good. So we're proud of that. We've done a number of things and really operationalized the business. One of the beauties of a self-serve business is it costs you very little to get lots of customers. The problem is you don't engage with them unless they come to you for support or other activity. And we kind of flipped that to a much more proactive model. We use digital signals and monitoring and data science to look at a customer's journey from when they hit the website. What tutorials do they read? What do they contemplate when they sign up? What don't they sign up for? What's that first 30-, 60-day use case looking like? Are they ramping? Are they installing? Are they using everything? And then we do a lot of digital engagement with them to figure out where they may be stalling. We send them tons of -- not tons, but we'll send them e-mails if appropriate. Have huge open rates, 20% open rates for our customers. We know e-mail pretty good from our history, that's off the chart. So these developers, these early-stage folks, are really hungry to optimize on the platform and they need that help. So we will always provide support, but by being more proactive, it's kept people longer. And we've really driven up retention in that first year, and then that carries into the second year. Customers don't leave our platform typically after they stay a year, they really don't leave and so that's been a big contributor. So it's been a wholesale change, a transformation, we like to say, in terms of how we engage with our customer, change the relationship, be much more proactive in helping them get onboarded onto the platform, be successful in that first year, which really raises the stakes and the probabilities that they're going to be with us a really long time. Yes?

You said you've gotten [indiscernible].

The question is do we have a lot of exposure to crypto. Crypto is a small percentage of our revenue, sort of single digit, low single-digit percent. The reason they're attracted to our platform for blockchain and, let's say, the positive uses for crypto is really because we have a really high-performing compute capability. So really computational intensive business models are run on our platform, whether it's streaming ad serving models where you're calculating a lot of complex calculations, media businesses like streaming video, et cetera, computationally intensive. So that's why they're attracted. We're obviously managing the security, which is a part of our core value proposition, and we'll continue to -- that's not an episodic thing. That's a constant thing for those of us that are on the Internet, involved in that business. But it's a real opportunity for us given a lot of the business models that are emerging. And we're really well positioned relative to our price point, our value, our low-latency global network to capitalize on that opportunity as it has evolved.

Great. We're over time. Thank you, Yancey and Bill, for a great conversation.

Thanks, Josh. Thank you.

Thanks, Josh.

Appreciate it.