Fresh Letter from Matthew Prince and Michelle Zatlyn from CloudFlare
To our potential shareholders:
Cloudflare launched on September 27, 2010. Many great startups pivot over time. We have not. We had a plan and have been purposeful in executing it since our earliest days. While we are still in its early innings, that plan remains clear: we are helping to build a better Internet. Understanding the path we’ve taken to date will help you understand how we plan to operate going forward, and to determine whether Cloudflare is the right investment for you.
Cloudflare was formed to take advantage of a paradigm shift: the world was moving from on-premise hardware and software that you buy to services in the cloud that you rent. Paradigm shifts in technology always create significant opportunities, and we built Cloudflare to take advantage of the opportunities that arose as the world shifted to the cloud.
As we watched packaged software turn into SaaS applications, and physical servers migrate to instances in the public cloud, it was clear that it was only a matter of time before the same happened to network appliances. Firewalls, network optimizers, load balancers, and the myriad of other hardware appliances that previously provided security, performance, and reliability would inevitably turn into cloud services.
Network Control as a Service
We built Cloudflare to provide the suite of cloud services we anticipated customers would demand as they looked to replace their on-premise, hardware-based network appliances. That was an audacious goal and it shaped both our business model and our technical architecture in ways that we believe differentiate us and provide us with a significant competitive advantage.
For example, since we were competing with hardware manufacturers, usage-based billing never made sense for our core products. In the on-premise hardware world, when you suffered more cyber attacks you didn’t pay your firewall vendor more, and when you suffered fewer you didn’t pay them less. If we were going to build a firewall-as-a-service — or any other network appliance replacement — we needed predictable, subscription-based pricing that reflected how companies wished they could pay for their hardware.
We also knew that more data gave us an advantage no hardware appliance could match. Like an Internet-wide immune system, we could learn from all the bits of traffic that flowed through our network. We could learn not only about bad actors and how to stop their attacks, but also about good actors and how to optimize their online experiences. Since more data helped us build better products for all our customers, we never wanted to do anything to discourage any potential customer from routing any amount of traffic, large or small, through our network.
Efficiency is in Our DNA
This core tenet of serving the entire Internet forced us to obsess over costs. Efficiency is in the DNA of Cloudflare because it had to be. Being entrusted with investors’ capital is a privilege and we make investments in our business always with a mind toward being good stewards of that capital. Moreover, while it was tempting to just pass along costs like bandwidth to our customers, we knew if we were going to provide a compelling value proposition against hardware we needed to be ruthlessly efficient.
To achieve the level of efficiency needed to compete with hardware appliances required us to invent a new type of platform. That platform needed to be built on commodity hardware. It needed to be architected so any server in any city that made up Cloudflare’s network could run every one of our services. It also needed the flexibility to move traffic around to serve our highest paying customers from the most performant locations while serving customers who paid us less, or even nothing at all, from wherever there was excess capacity.
We built Cloudflare’s platform from the ground up with a full understanding of our audacious plan: to literally help build a better Internet. We didn’t run separate networks to provide our different products. We didn’t use expensive, proprietary hardware. We didn’t start with one product and then attempt to Frankenstein on others over time. Our platform was purpose-built to efficiently deliver security, performance, and reliability to customers of every size from day one. And our platform has allowed us a level of efficiency to achieve the gross margins of leading hardware appliance vendors — 77% in the first half of this year — but with the greater predictability of a SaaS business model.
Our Platform Approach
For some it may be challenging to categorize our business because our platform includes an incredibly diverse set of capabilities. We provide security products like firewall and access management, performance products like intelligent routing, and reliability products like vendor-neutral load balancing — all as a service, without customers needing to install hardware or change their code.
We also have functions that play supporting roles to the products we sell. For example, we built one of the fastest, most reliable content delivery networks not because we were targeting the CDN market, but because we knew caching was a necessary function in order to efficiently deliver our core products. We built the world’s fastest authoritative domain name services, not to sell DNS, but to deliver service levels we knew our customers needed.
We provide features like CDN and DNS for free to all of our customers. We will continue to implement this strategy; onboarding more customers onto our platform and capturing value from our highly differentiated products that, once using any part of Cloudflare’s platform, are only a click away.
Potential investors who are new to Cloudflare sometimes ask questions like: “What will you do if CDN bandwidth prices continue to fall?” We remind them we’ve given CDN away for free since Cloudflare launched in 2010, not because we were trying to disrupt the CDN space, but because the much more valuable products we provide our customers need a highly optimized global caching network to perform up to our standards.
We Create More Value Than We Capture
But there is another reason for taking the approach that we do. Cloudflare has always put our customers first and prioritized creating much more value than we capture. We work to get customers onto our platform because, once on board, we know we will be able to solve so many of their problems over time. We aim to make the combined value of the products on our platform significantly more than customers can get from any combination of point solutions.
In the past, to deliver Internet security, performance, and reliability not only required an organization to buy rooms full of expensive network appliances but also to hire IT teams to manage them. While there were some companies that could afford this, the cost was prohibitive for many. Instead of serving only those that could have paid the most, we intentionally made the decision to start by focusing on organizations and individual developers that had previously been underserved. We made our products not only affordable, but easy to use.
And we didn’t stop there. We have continued to improve with every bit of traffic we have seen. In doing so, we have moved up market to the point that, today, approximately 10 percent of the Fortune 1,000 are paying Cloudflare customers. We think one of the best ways to measure the value we deliver is our Net Promoter Score of 68 among paying customers, rivaling some of the best consumer brands in the world. Not only are we obsessed with our customers, but our customers are obsessed with us.
We Are Focused on Consistent Growth Over the Long Term
One of the characteristics of the world’s greatest SaaS companies is that they typically enter a market in some small way and then use that toehold to expand their relationship and move up market. We learned from the great SaaS companies that came before us. This strategy has resulted in consistent, long-term — rather than explosive — growth. Contrast this with companies that only build a better mousetrap. They initially experience heady growth shifting defined spend from one product to another, but the challenge they then face is existential: what’s their second, third, and fourth act? Cloudflare doesn’t have this problem.
We will continue to invest in R&D so long as it demonstrates a significant return. Our investment philosophy is oriented around making many small, inexpensive bets — quickly killing the ones that don’t work, and increasing investment in the ones that do. While we will consider M&A when opportunities present themselves, our bias is toward internal development tightly integrated into our efficient platform. We aim to build a massive business — slowly and consistently.
Finally, there are two of us signing this letter today, but three people started Cloudflare. Lee Holloway is our third co-founder and the genius who architected our platform and recruited and led our early technical team. Tragically, Lee stepped down from Cloudflare in 2015, suffering the debilitating effects of Frontotemporal Dementia, a rare neurological disease.
As we began the confidential process to go public, one of the early decisions was to pick the code name for our IPO. We chose “Project Holloway” to honor Lee’s contribution. More importantly, on a daily basis, the technical decisions Lee made, and the engineering team he built, are fundamental to the business we have become.
It has indeed been an incredible journey to have built Cloudflare into what it is today. We are grateful to our customers for their business and trust, to our team members for their dedication to our mission, and to our shareholders, and potential shareholders, for their support and encouragement.
And we’re just getting started.
Matthew Prince Michelle Zatlyn
Co-founder & CEO Co-founder & COO