Digital Ocean is a very interesting company when you look at what they have done. They call themselves "the developer cloud", but primarily we can look at them as a hosting platform. They have always specialized in offering VPS services which are virtual private servers.
For those that don't understand what this means, they allow you to rent out "shares" or portions of a larger server to host your website, api, run machine learning, compile data, or anything else that you want. Servers are powerful and expensive to run. Many companies don't need the full power of a full server in order run something simple like a website. So if you only need 1/50th of a server, then Digital Ocean will rent that to you at a fractional cost of the whole server. You get root access to the server to do anything you want with it, just like if you owned the server yourself. It is also worth mentioning that these server shares are virtually wardened off and sandboxed so that you don't have any knowledge or access to anyone else's server shares even if you are on the same server.
This is where the VPS name comes in. These are "virtual" servers in that they are only seperated virtually, but phsyically are part of a larger shared server. They are "private" in that you have full access to your own share. And they are "servers" in that you can do whatever you want with them.
The VPS/Hosting Landscape
This concept might sound genius and novel, but it has been around for a very long time. There are much bigger and smaller players in the space already, even long before Digital Ocean came on the scene. Digital Ocean's closest competitor would be Linode. But you could also argue any webhost is doing the same thing in a more limited capacity. For example HostGator, Bluehost, and others are all VPS's behind the scenes. But they are much more "locked down" and don't allow full server access. They replace root-level command line access with cute dashboards and point-and-click menus. So that is why these services tend to attract bloggers with more limited tech knowledge, while Digital Ocean and Linode attract developers with knowledge to manage custom servers themselves.
So if Bluehost and HostGator are offering a simplified version of what Digital Ocean is doing then we can also say that Digital Ocean has more advanced competitors too, such as Amazon's AWS, Microsoft's Azure, and Google's Cloud Platform. In recent years, Digital Ocean has expanded its offerings to compete with some of the other services offered by the more advanced competitors including managed databases, block storage, backups, networking services, and more.
So you read that right, if hosting is a spectrum, Digital Ocean sits in the middle of it between simplified solutions and advanced solutions. Are they the goldilocks in hosting or a bastard child that no one wants to claim? Will being a central player make them profitable and a good investment?
To be honest I don't know what to think about Digital Ocean as an investment. Because I don't think they will reach the level of profitability that the other two ends of the spectrum experience.
The simplified hosting solutions like HostGator and Bluehost both make gobs of money charging around $5 per server. They serve a massive audience because everything they do is point and click. Their bread and butter are people that want to start blogs or desire to start an online business (which is literally everyone, everyone's mom, and everyone's grandma). They serve a MASSIVE market that encapsulates nearly everyone. At $5 /mo it really adds up!
These simplified solutions are also able to overprovision their servers because they are in a business where they don't need to guarantee server resources. They operate their business under the same business model as a gym. They sell more memberships than they could possibly fulfill at a time with their given resources because they know that not all of the customers are going to try to use the service at the same time. In fact many of their customers will keep paying without using it at all, planning to use it again in the future. This means that these simple hosts sell these $5 /mo website plans and ⅔ of their customer's don't even use the service but continue to pay. Their profit per server is HUGE.
Digital Ocean on the other hand offers guaranteed server resources. So they promise a certain amount of RAM, CPU Power, and so forth. This means they can NOT overprovision servers. They can only sell a server up to 100% capacity and then must get another server to bring on more customers. The shared hosts can sell servers up to 900% - 1,200% capacity. This means they make less profit per capital investment. Their overall profits will always be lower than a Shared Host. Getting more customers will always mean getting more servers.
On the other end of the spectrum is Amazon AWS, Microsoft Azure, and Google Cloud Platform (GCP). These are advanced platforms that cater to huge startups, Fortune 500 companies and more. They have hundreds of services (over 200 services offered by AWS) from managed databases, ice storage, semi truck powered data transfer, sattelite ground stations, and everything in between and beyond. They cater to larger customers, but most of their customers are paying 5-7 figures or more per month.
So where is Digital Ocean in all of this? Well they are "The developer cloud". They cater to developers. The bloggers and restaurants of the world who run wordpress sites can't boot up a Digital Ocean instance and maintain a Ubuntu server for their website that gets 5 visitors per day. On the other end of the spectrum, Digital Ocean can't compete with AWS, Azure, or GCP in breadth of services or scale.
AWS, Azure, and to a lesser extent GCP, are the hot names right now. The president of AWS just became CEO of Amazon. And AWS is Amazon's most profitable division right now. Microsoft is bringing in record profits in its Azure division right now too. So the stock market is fairly enamoured with cloud services right now and I think they will like the Digital Ocean play too.
But I do hestitate a little because if I had to bet on a single player right now, Digital Ocean would probably be at the bottom of the list. My first bet would be Microsoft with Azure. AWS is a very good service with a great reputation. They are the industry leader. But Azure is going after the whales in the space. They are targeting every Fortune 100 company, who already had a dedicated Microsoft rep, and is convincing them to switch over to Azure with huge incentives to do so. These companies are whales. When Microsoft gets one of these clients to switch over it is often millions of dollars a month in service fees. Yes, per month! Thats where my money is at.
Digital Ocean Finances
Digital Ocean claims in their S-1 that they have $357M ARR across 570,000 customers. This means they are averaging about $52 per month per customer.
Hmmm, that seems kinda low...
We don't have exact per-customer numbers from the other players but we do know that Amazon made about $40B in 2020 and was worried that it was slower than expected because companies were cutting cloud expenses due to the pandemic. AWS did say that they have now exceeded 1,000,000 customers. This means that AWS is averaging between $3,000 - $5,000 per customer.
At $52 per customer per month, Digital Ocean would need 57.6M customers in order to compete with AWS, which is simply not possible or practical. There are not enough server-needing customers to justify that. We aren't talking about needing a little growth here, we are talking about it simply not being possible.
Digital Ocean needs whales, huge companies (Fortune 500) that have massive server budgets. Microsoft has been pulling whales from AWS and locking them in. Digital Ocean is not only late to the game but they can't please the whales. It is like a off-strip Super8 motel in vegas trying to pull in whales from Ceasar's Palace on the strip. It is just hard to concieve.
It is also concerning that Digital Ocean has not been profitable (at least as far back as their S-1 goes). Right now they thrive off attracting developers through their $5 a month server plans (although I have noticed they recently upped the price to $6 for "premium" CPUs with NVMe SSDs lately). Is this sustainable?
Currently they are losing around $10M per month. Loooking at their P&L statements there aren't a lot of obvious places to clean the business up to increase profitability. They are currently spending ~$20M per month on research and development. One could hope this would decrease with time. It is possible they are planning a massive new roll-out. But at the same time, this would need to stay constant to compete in the changing landscape of hosting. At best we could say this could be reduced by half, which would bring overall profits back towards net-zero. But again, I don't know how practical a 50% reduction in R&D really is.
I also think that Digital Ocean will need to increase their marketing spend in order to compete with the larger competitors and gain marketshare. Right now they spend about $9M per month on sales and marketing which really is probably not high enough. I would expect that category to increase over time with how competitive this market currently is. This of course reduces profitability, but seems all but inevitable in order to increase long-term viability.
This moves into my other point about Digital Ocean. They pride themselves on being a self-service company that is grown organically by the community that it operates. I get it, this is a really cool aspect of the company. Everyone loves organic growth too. But how sustainable is this organic growth model? Not very many large public companies, if any, can rely on organic growth. Even a company that has mastered organic growth (which I would argue Digital Ocean has already done) will need to get other growth channels added in there.
A private company can sustain itself like Digital Ocean is doing, but to go public they need to get dirty. We need to see enterprise sales teams hired and that marketing budget to double, triple, or quadruple.
Digital Ocean Red Flags
There were a few other red flags I see with the Digital Ocean S-1. First, they only have an NPS score of 65. This isn't terrible for a large business, but Digital Ocean is only just now entering "large business" territory. We would expect a developer focused service to have a higher NPS score. But it isn't a terrible score either so I don't want to diminish the value of a 65 NPS. I just expected more. However I would bet that AWS has an NPS score well below 50. But Digital Ocean's competitive advantage is mostly its developer relationships so the NPS score is an important factor in that. AWS's competitive advantage is price and service scope.
I mentioned this in the marketing analysis, but I think it needs to be said again here. One red flag I see is that Digital Ocean needs to get their hand "dirty". I love (as a consumer) that Digital Ocean prides itself with the values listed above. They are community focused and "Love is at our core". But "Simplicity in all we DO" will not cut it when I look at the business from an investor's perspective.
Why I Am Skeptical
I have been using Digital Ocean for years, since they first launched. I have been a huge fan and according to my affiliate reports I have sent them hundreds of [paying] customers. I have thousands of dollars of Digital Ocean credit right now that I can use and this very blog is hosted on Digital Ocean as we speak.
I really love Digital Ocean... as a consumer... as a developer. I'm a fan.
But I am nervous to invest with them as a investor.
This gets to my ultimate point. I think that Digital Ocean is a great company and I hope them continued success. I have watched them grow and develop into an incredible company. But I think they should have raised another private round of funding before going public, if going public at all.
I am worried that Digital Ocean will not be able to please Wallstreet the way it currently is. This means selling their soul to the devil and losing track of what makes them special as a "developer cloud." I look at what Digital Ocean needs to do to compete in the public space. It means much more growth than they have now. They can not reach the growth they need to get to by leveraging the developer community. This means they need to please shareholders and C-levels. What makes those people happy will require turning their back (at least partially) towards developers. Developers loved them because they were the anti-AWS, the anti-Azure, and the anti-GCP. But if they have to become an AWS in order to please Wallstreet (and they have a fiduciary responsibility to do so) then they will lose what makes them special.
Is it worth it? I'm not sure.
You have to leave your heritage behind to be something grander than you are now. But what if the grandeur that you seek is unattainable? Is it worth abandoning the customers that got you where you are today if reaching the level of AWS is unattainable anyway?
This is my concern. Digital Ocean needs to become AWS and Azure. They need to steal those customers. But how can they compete against them? These are companies who have money to burn and relationships established. These other companies have reputations (Amazon and Microsoft and Google) that please the C-Levels and make shareholders happy. This is the fight that Digital Ocean is up against.
We will see where their valuation is. The saving grace is that tech valuations are insane right now (80-100+ times revenues), and they might be able to get something done. But we need long term results and I am waiting for that plan. For today, I have my money on Microsoft and Amazon. I'll probably buy some early shares of Digital Ocean as a gesture of faith (and a gamble that they can pull it out).
I am rooting for you Digital Ocean, even though I doubt you can win.