Cloudflare: The Network is the Computer ($NET Deep Dive)

One of the most disruptive companies on Earth

I write publicly because genuine long-term thinking is rare. Instead of short-term “book reports,” I make an effort to uncover real long-term insights by breaking things down to first principles. Then I write about it so that anybody could understand me.

I focus on 10-year growth (or decay) stories; businesses with emerging or established moats (disadvantages), trading at asymmetric prices. I sometimes indulge on good companies temporarily losing their way.

Curiosity is my superpower and I believe every thing is a unique puzzle.

All my models are on GitHub (@realVasileios).

Table of Contents

1. Thesis

2. What Cloudflare Really Does

3. The Growth Story

4. The Moat

5. Valuation

6. Risks

Thesis

Cloudflare is consolidating all corporate networks into one, as well as building a disruptive cloud computing platform, Cloudflare Workers. Not only is the revenue runway long, but Cloudflare has also developed strong competitive advantages as the ultimate lowest-cost producer, amongst others. At 26x LTM sales, Cloudflare is a premium stock; ultimately you get what you pay for, and if you are long-term minded, it’s easy to realize that even at these levels, Cloudflare is bargain.

What Cloudflare Really Does

Cloudflare is often mischaracterized as a “cybersecurity stock.” Although cybersecurity is part of what Cloudflare does, it’s a far cry from defining completely the company. Cloudflare is an Internet infrastructure company, and thus, a proper deep dive must begin by explaining how Cloudflare fits within the Internet.

What is the Internet?

The Internet is roughly 70,000 physical networks of cables in the ground, air frequencies, cell towers, and various buildings that are stitched together to transport information between any two points on earth in a fast way. Given how fragmented the Internet is, it’s extremely important to have great cooperation between these networks so that information flows easily across the earth. This cooperation is facilitated by a set of common rules and standards that every network follows.

If this is already too complex, picture the road system, because it’s nearly identical to the Internet and how information moves. The road system is a compilation of many networks of roads that are stitched together in intersections. For example, each town has local and county roads, which connect to larger and longer state roads, which then connect to the Interstate highway system, which is equivalent to what people call the Internet backbone. Also, the road system has addresses that have zip codes as well as laws and rules that all drivers follow to reach their destinations. The Internet is similar.

Today, the Internet serves only as “almost dumb pipes”, meaning its purpose is only to transport information rather than process information. Processing of information and software happens only at corporate and cloud data centers which you can think of as being located at the edges of the Internet and not within the Internet.

I say “almost dumb pipes” because there is a little bit of processing that happens on the Internet today. When the Internet started being used more broadly in the late 1990’s, there was a big bottleneck that emerged. Imagine if I lived in Germany and I tried to access a website in San Francisco. Given the physical distance and Internet congestion, at best, the website was slow to load, and at worst the website never arrived in Germany because of a connection failure.

To solve this networking bottleneck, Akamai came up with an ingenious idea. Akamai placed servers all around the world, near Internet users, so that people would access their websites from nearby Akamai servers rather than servers where websites originated. Akamai placed these servers in buildings within the Internet. These buildings are either the buildings of a network owner/operator (i.e. Verizon) or Internet Exchanges (i.e where networks connect to exchange information). This way, if I lived in Berlin, I would access the SF website from the Berlin Internet Exchange which is only milliseconds away from me and not SF. In other words, Akamai gave the Internet memory by saving websites and other Internet properties near consumers across the globe. In technical terms, Akamai is a Content Delivery Network or CDN.

What is Cloudflare?

Similar to Akamai, Cloudflare started as a CDN company as well, though the big difference is that Cloudflare provided services for small companies and websites (i.e. low end of market), whereas Akamai serviced big companies. Because Cloudflare made its CDN available for free, 20% of websites signed up for the service. With this large scale Cloudflare leveraged its CDN to build a global network of servers within Internet buildings, and then develop additional services such as cybersecurity.

You can think of Cloudflare similarly to Amazon. Amazon started selling books to poor college students which is not a very lucrative market. Nevertheless, the books gave a launchpad for Amazon to build warehouses and a logistics network. Once you have that, you can sell any product you want out of your warehouses. Amazon doesn’t own the roads but pays tolls/taxes to use them to transport goods to consumers.

Cloudflare is nearly identical, only swap “books” for “CDN”, “warehouse” for “server”, and “logistics network” with “network.” Cloudflare started out as a CDN company for SMB, which is not a very lucrative market. Nevertheless, CDN allowed Cloudflare to build a fleet of servers around the globe and a network of cables that connects all of them. Cloudflare doesn’t own cables between its servers but rents them from others. Once you have servers everywhere and all that is established, the products you can sell are limited only by the imagination of your R&D department, since a server is just a generic piece of hardware that can run anything you want. CDN is nothing but one product your servers can run.

At heart, Cloudflare is a cloud, but much different than AWS etc. in that its servers are much more distributed across the globe.

The Growth Story

So what types of products does Cloudflare sell out of its network?

There are three types of products Cloudflare sells: Web Application Services, Corporate Networking, and Cloud Computing. The former is how Cloudflare started, and the latter two is where it’s going. Let’s break them down.

Web Application Services

Web Application services is how Cloudflare started and it includes CDN. In general, this is the segment that services Internet assets such as websites or software applications. These “applications” have a need to be fast, reliable, and secure when a customer accesses them. Imagine the American Express website or mobile application. Amex wants two things. First it wants a fast website and mobile app. To achieve that, Cloudflare uses its CDN to make it fast by saving the website and parts of the app in its servers, which are very close to users. Second, Amex wants to protect its website and app from hackers and make sure they are always up. To achieve that, Cloudflare directs all the traffic to its servers first, where it gets to inspect it and make sure it is legitimate before it hits the Amex servers. In this way, Cloudflare acts as a shield.

One common attack is a Distributed Denial of Service (DDoS) attack where millions of devices send a request to the same server at the same time to overwhelm it. Since Cloudflare sits in front of the server and inspects all traffic, it can discard the traffic from the DDoS attack.

There are many more products in this segment, though CDN and security of websites and applications is the crux of it. This is by far the least interesting and more mature market for Cloudflare. Again, web application services is how Cloudflare started and it is what gave it the means to build its global network of servers.

Corporate Networking

In general, networking services are about getting secure access to the Internet. For consumers, that’s home and mobile phone Internet access, whereas for corporations, it means connecting offices, data centers, and WFH employees. Although consumer networking services are straightforward, setting up access to the Internet is far from trivial for corporations today. Cloudflare plays in the corporate networking arena.

To understand how revolutionary Cloudflare is here, we must understand how things are today.

Just like websites and applications, when a company accesses the Internet from its various locations, it wants everything to be fast, reliable, and secure. Particularly secure. To achieve that a company has to stitch together parts and pieces from 20+ vendors as well as hire a big networking team. A corporate network consists of two things: physical connectivity to the Internet and specialized hardware servers.

The way connectivity works for corporations is not simple. First there is a cable that goes into an office or data center that’s owned or leased by somebody like AT&T. If a company has global offices in 100 countries, it’s likely that AT&T doesn’t own or have relationships with cable owners in those countries. As a result, the company, without any buying power, has to strike agreements with a mesh of Internet providers all over the world which is uneconomical and cumbersome.

Once you have connectivity, you must have security and routing. The way companies today ensure security is that all the Internet traffic to/from all employees must first go to a company data center for inspection and then to its final destination. To achieve this inspection, a company buys expensive specialized hardware servers such as firewalls, routers, switches etc. from companies such as Palo Alto, Fortinet, and Cisco, that serve only one purpose, and stitches them together. For example, if I am accessing “http://google.com” from my corporate computer, my request goes first to my company’s data center and then to Google. This type of round trip causes latency and a bad experience. Again, the networking team goes to many vendors to buy these specialized hardware servers and install them in not only its data centers but also every single office. It’s not unlikely that a fortune 500 company would have 30,000+ of these.

Cloudflare is a completely new paradigm for both connectivity and specialized hardware.

On the connectivity front, there is currently no global Internet service provider since Internet providers have historically been a regional business. As a result, a company must go to multiple providers. Cloudflare is ending that by creating one global network where it leases cables from every single network provider across every single country, and thus, will be able to give access everywhere economically, since Cloudflare would have significantly more buying power.

On the hardware front, because Cloudflare has generic hardware servers it can write all the functions that the specialized hardware execute into software and run them out of its servers. As a result, when a company employee uses the Internet, a “http://google.com” request will not have to go to a company data center to be inspected. Rather, it can just go to Cloudflare’s servers, which are milliseconds away, get inspected and Cloudflare will take it to its final destination fast and securely.

To contextualize what’s happening take the following analogy. In the 1990’s and 2000’s, every person had a ton of specialized hardware: walkman, CD player, a map, books, home phone, computer, DVD player, iPod etc. In other words, people had to go to many places to satisfy their basic need of entertainment / information / communication etc. Today, through software all this specific hardware can be consolidated and bundled into the iPhone. Cloudflare is the exact same. Corporate networks are riddled with specialized hardware, connectivity vendors, and inefficiencies. Cloudflare, through software, can turn all the specialized hardware functions into software in its servers. All that to say that today companies go to 20+ vendors and hire huge networking teams, whereas with Cloudflare you can just hit a button and have the best network in the world.

It’s usually said that in technology there are two way to make money: “bundling and unbundling.” You can think of Cloudflare as bundling corporate networking.

Cloud Computing

Cloudflare is building a cloud computing platform, called Cloudflare Workers. Workers is a significantly better way to do cloud computing than the current paradigm of AWS, Azure, and Google Cloud and will be taking an increasing amount of share as the platform matures. There are many reasons why Workers is a better cloud computing platform.

Ease of Use

Cloud computing companies such as AWS have made building an IT backend significantly easier than corporate data centers. This makes sense. Companies want to build applications. It’s obviously much easier and efficient to build that application by renting a couple servers from AWS vs. having to hire a team and build an entire data center.

Nevertheless, what most don’t realize is that AWS is not a cloud. Rather, it is an outsourced data center. AWS builds the data center, buys the servers, and connects them to the Internet, but beyond that, it’s the job of the customer to figure out how to use these servers. When a customer goes to AWS, they choose a specific data center and specific servers within that data center. To do that, a customer has to build a whole team to figure out the most efficient use and all the steps one must take to use AWS. Again, it’s important to understand that AWS is just an outsourced data center where AWS abstracts away some of the physical elements of using servers. The customer still has tons of work to do to run and maintain their IT backend with AWS, and unfortunately, customers aren’t good at that, resulting in a big waste of server time.

Contrary to that, Cloudflare is extremely easy to use, and that’s because Cloudflare is a serverless cloud. This means that Cloudflare abstracts away not only the physical aspects of cloud computing but all aspects. All the client has to do when using Cloudflare is to give the code they want to run, and then Cloudflare figures out where and how to run that code.

What this means is simple. With corporate data centers you needed millions of dollars and a huge team to run an application. With AWS you need a huge team to run an application. With Workers you just need a couple of software engineers to write the applications because all the mundane stuff has been taken care of by Cloudflare. With Workers, companies can focus exclusively on writing business logic rather than maintaining unnecessary IT backend infra.

Better Pricing Model

Since AWS is an outsourced data center, a customer rents physical servers to use. As a result, the way AWS charges is per server per hour, in spite of how much a customer uses these servers. This dynamic forces customers to overbuy servers just in case their applications see large spikes. As a result, the utilization levels of the servers that are rented is low for AWS customers.

In contrast, Cloudflare charges only for usage. As a result, customers pay only for what they use. This is a significantly fairer model and customers don’t have to overbook servers in order to gain scale when they need it; Cloudflare scales up and down applications when needed.

Vendor (un)Lock-in

The basic business model of cloud providers is to lock you into their data centers, and it goes like this. A company starts using the data center of a cloud provider, and moves all data inside that data center. At this point, the customer has completely locked themselves into the cloud provider’s data center because if it decided to migrate the data outside the cloud provider’s data center it would have to pay massive fees (called “egress fees”) to move this data out.

In contrast, Cloudflare has 0 egress fees for using Workers, which makes it significantly more customer friendly, as it doesn’t attempt to lock you in. One corollary is that Cloudflare is significantly more incentivized to use innovation as a way to keep customers, which is not the case for AWS, since their mechanism to keep you is egress fees.

Price

The final and perhaps most important advantage of Workers and Cloudflare at large is that it is extremely cheap. Depending on the different cloud services you use, Cloudflare can be 80-90% cheaper than conventional clouds. The reason Cloudflare can price cloud services so low is outlined below (Moat section).

The Moat

Although Cloudflare’s growth runway is impressive, the competitive advantage is even more outstanding. There are a few competitive advantages, though the most important one is that Cloudflare is the lowest cost producer of all its services.

Let’s take it from the beginning.

First, it’s important to understand the economics of Cloudflare. A business like this has three types of costs: bandwidth (70%), data center (20%), servers (10%). Data center is the rental and electricity charges Cloudflare pays, whereas ”servers” is easy to understand.

“Bandwidth” is the amount Cloudflare pays to rent other people’s cables to move data around the Internet. In general, there are two types of physical networks on the Internet: “customer” and “transit” networks. Customer networks have consumers (i.e Spectrum) whereas transit networks are there to just move traffic between networks and charge a fee for that. In our road system example, transit networks would be the Interstate and state highways. There aren’t any homes on them and only serve to bring you from town to town, aka the customer networks.

Advantage #1: Lowest Cost Producer: locating in customer networks

When Cloudflare started out, it decided to give its services on a freemium basis. That was genius. At that point (i.e. 2009) there was a huge need by small to medium businesses/websites for the Application Services, such as CDN and cybersecurity. As a result, millions of people signed up for Cloudflare’s free services, which gave Cloudflare huge scale very fast. Today, Cloudflare provides Application Services for 20% of the entire Internet. This scale is the backbone for Cloudflare’s low cost advantage.

Having such large scale entices customer networks to put Cloudflare in their network building in order to bypass transit networks. The way this works is that customer networks have consumers who use the Internet. When a consumer requests a website, the network operator must go out to the Internet and pay a bunch of transit networks in order to retrieve that content and serve it back to its customer. This adds to the network operator’s cost and since the network has to go far to retrieve the content, the speed and experience to the customer is low.

Since Cloudflare has such large scale and delivers so much content, the network operator has an incentive to put Cloudflare servers in its buildings and serve the websites from there rather than having to pay transit networks. Also, since Cloudflare servers are within the network and only one hop from the consumers, the experience becomes faster and better. Finally, the network operator pays for Cloudflare’s electricity and data center costs as well.

The final result for Cloudflare is that it pays no bandwidth and no electricity in many of its data centers. This gives an incentive to Cloudflare to continue expanding and spreading more servers around and cooperating with customer networks because it crushes down its cost structure. In other words, Cloudflare has many of its servers in places where it pays barely any costs. As a result, all the services that Cloudflare offers (application, networking, Workers) have extremely high gross margins, allowing Cloudflare plenty of room to be competitive.

Advantage #2: Lowest Cost Producer: bandwidth agreements

Even though Cloudflare is in many networks and that number will be increasing, it will take some time before it stops paying transit networks. The scale of Cloudflare gives it an ability to strike wholesale agreements for bandwidth. What this means is that it pays only for one-way bandwidth. Imagine that traffic goes either to consumers or from consumers. The way Cloudflare structures its deals, it pays only for the largest of the two paths. Since Cloudflare is the largest CDN in the world, it pays for outbound traffic, aka going to consumers. As a result, any traffic that’s inbound, Cloudflare doesn’t pay any bandwidth. This is important because all the networking and cybersecurity, and increasingly Workers, services all are inbound traffic, and thus they are near 100% gross margin, allowing Cloudflare plenty of room to win on price.

Advantage #3: Lowest Cost Producer & Network Effects: peering

The 70,000 networks of the Internet are split into 3 tiers depending on their size, with Tier 1 being the 20 largest networks in the world. Amongst all these networks, there are two types of relationships: paid and unpaid. Paid relationships are simple. For example, a small Tier 3 network in St. Louis wants to access content in NYC. In order to reach that content it must use the cables of a Tier 2 transit network, which then must use the cables of a Tier 1 transit network. In this sequence, the Tier 3 network pays the Tier 2 and the Tier 2 pays the Tier 1 network.

On the other hand, unpaid relationships are different. In the above example, if the Tier 2 network had to use just another Tier 2 network, then it wouldn’t have to pay for the usage. We call this type of relationship network “peering,” and it’s an unpaid relationship when two networks of equal size exchange traffic. Peering is simply an unpaid exchange of traffic between two networks.

What’s important to keep from all this is that the larger the traffic you move around the Internet, the more leverage you have with other networks. That’s because all networks want to transfer traffic fast and efficiently. If your network is the best way to route traffic, then other networks want the routes your network has to offer. As a result, the larger you are, the more likely other networks are to peer with you, the more versatile your network is.

This is obvious by the 20 largest physical networks in the world, such as Lumen. These Tier 1 networks peer with each other, and thus pay no bandwidth to move traffic around. It’s the same with Google. Google carries so much traffic around the Internet that every network wants to peer with Google to make their own networks better. In other words, there is a network effect amongst networks. The larger you are, the more networks want to be associated with you, which makes your network cheaper and more performant, which means that even more traffic comes to you, which means even more networks want to be associated with you.

Cloudflare carries not only 20% of the Internet, but its networking and Workers businesses will be bringing on significant amounts of new traffic. This way it is highly likely that Cloudflare will be the largest network in the world, and will completely eliminate most bandwidth fees in the future, similar to Google. The one variable that will determine that is whether Cloudflare can increase its traffic even more.

Advantage #4: Lowest Cost Producer: high asset utilization through serverless model

Cloudflare has significantly better utilization on its assets than any of its competitors. This is because Cloudflare operates as a serverless platform. What this means in the context of utilization is that Cloudflare can run its services from any one of its 100s of locations. For example, Cloudflare can run, if it deems appropriate, a security service out of Pakistan even though the customer is in the US, only because the Pakistan servers are being underutilized since it’s late at night there. In other words, serverless means that Cloudflare decides where the code runs.

This is in contrast to AWS. For AWS, it’s the customers that choose what data center to use. AWS rents out servers to customers and its customers are then responsible for figuring out how to utilize those servers. As a result of this fragmentation, there are tons of AWS servers that are idly running and customers are paying for, since customers are not in the business of efficiently utilizing cloud servers.

Advantage #5: Lowest Cost Producer: high asset utilization through isolates

To understand isolates you must understand the historical context. In general, since servers were invented we have been trying to maximize their utilization through technology.

We first started in the 1990’s with what was called bare metal servers. What this means is that one server could run only one application at a time. As a result the server was idle for much of the time.

In the late 1990’s VMware invented virtualization, which basically allowed us to run multiple applications in the same server, which made everything much more efficient. The issue with corporate data centers though was that outside of work hours, the data center would sit idle.

Virtualization gave birth to cloud computing, since cloud providers could take advantage of having many customers that operate 24/7. Beyond virtualization, Google invented containers which was a technology that allowed us to fit even more applications in the same server, almost like virtualization 2.0.

Finally, Cloudflare had the issue of operating out of small data centers with very few centers and had to invent a new and more granular way to do computing. As a result, they invented isolates, which is basically a technology that allows it to fit 100x more applications than containers. As a result, isolates make Cloudflare significantly more capital and utilization efficient.

Imagine it this way. Cloudflare’s servers do 100x what an AWS server does, even though it’s the same server. As a result, AWS has to buy significantly more servers to achieve the same result.

Advantage #6: Lowest Cost Producer: best-in-class capital efficiency

Cloudflare is significantly more capital efficient than AWS. All else equal, the same amount of people require the same amount of servers. For example, let’s assume that 1m people need 10,000 servers to service their cloud computing needs. Whether you place these servers in one big data center or 1000 small ones, it’s the same as far as servicing the people. Obviously a company like AWS is the former and Cloudflare is the latter. The Cloudflare structure is significantly more capital efficient.

AWS builds big data centers which takes a lot of planning, forecasting, and overbuilding so you don’t miss out on customers. The capital commitments are enormous and basically AWS invests ahead of demand and tries to predict it for an entire continent. On the contrary, Cloudflare invests behind demand. For example, because it has a serverless model and many data centers, Cloudflare places a few servers in the south side of San Francisco in somebody else’s building. If demand picks up, it can send down a technician and easily plug in a few more servers. This way Cloudflare can tailor its servers purchases precisely to the demand for its services, which also helps with utilization.

Advantage #7: Counter-positioning

Cloud providers are highly unlikely to create a structure such as Cloudflare’s because it is highly counter-positioned to their own business model. Here is how this works, through an analogy. When Amazon.com was coming up as an online bookstore, everybody thought that Barnes & Noble would eat their lunch. B&N had much better access to capital, a better know-how on books, and a much more experienced team. What nobody realized was that B&N was counter-positioned to building the online bookstore. That’s because its economic incentives were trapped inside each individual store. In other words, B&N had a huge incentive to bring customers to the store and lock them in there because it would sell them high margin products like chocolates, and also it had to justify huge investments and personnel.

In a very similar way, Cloudflare is counter-positioned to existing cloud providers. For a cloud provider, their economic incentives are trapped within each big data center because their incentive is to bring customers and workloads back into a data center and keep them there through egress fees. In contrast, Cloudflare runs a serverless and no egress fee model. This means that it has an ability to run any application for any client from any one of its thousands of data centers, depending on how it sees fit. Similar to how Amazon.com can sell any book to any customer from anywhere.

The reason all this is important is because if an existing cloud provider decided to open up the gates and allowed customers to run things from anywhere they wished without egress fees, then big swaths of a cloud provider’s data center could go completely unutilized, or it would be a complete nightmare for an AWS to predict utilization. It would be if customers in California could access the fifth avenue B&N easily; no other store would work.

Advantage #8: Bundling

As we discussed earlier, networking and the Internet used to be a fragmented industry with thousands of service providers. Companies would have to stitch together solutions from tens of vendors and hire big networking teams to figure all that out. The argument we are making here is that Cloudflare is moving to consolidate the industry by vertically integrating through the power of software and its network.

In other words, we are going from an unbundled to a bundled world for networking. And even though the Internet is the most decentralized and distributed structure, it will soon become consolidated, with Cloudflare being the chief consolidator.

It certainly made sense that the Internet was fragmented when it started and that we would use multiple hardware companies. Nevertheless, “software is eating the world” is about to hit the Internet, with Cloudflare replacing all the hardware and connectivity providers with software.

Advantage #9: Innovation Machine

Given how novel Cloudflare’s mission is, it’s important to have an organization with a strong innovation muscle, because it’s hard to predict exactly where things may head. Cloudflare has an incredible and unique innovation muscle that is enhanced by the millions of SMB customers.

Cloudflare innovates in the same way as Facebook and Google. Because they have so many customers, they create a half-finished product, test it against a small segment of customers and ferociously make edits and developments before it’s ready to be released to others.

In addition, Cloudflare completely dogfoods its entire portfolio of products since everything is written up on Cloudflare Workers.

Advantage #10: Data Moat

Given its size and reach, Cloudflare has high visibility on Internet traffic and patterns that nobody else has.

Advantage #11: Switching Costs

Cloudflare’s products are mission critical for a business and thus not easily replaceable.

Advantage #12: Excellent and incentivized management team

Management have nearly the entirety of their net-worth in Cloudflare and have done a tremendous job bringing the company to where it is. They have complimentary skills and a long-term mindset.

Valuation

Given its high revenue growth (34% in 4Q25) and breakeven profitability, Cloudflare is valued on an EV/Sales multiple. Given the strength of the moat and long growth runway, the company is valued at a premium valuation of 26x 2025 sales. This level of valuation assumes flawless execution and continued growth, two aspects that Cloudflare has proven time and time again.

As with all high quality companies you get what you pay for. It’s unlikely Cloudflare will get much cheaper from today, and thus, for long-term minded shareholders, today’s prices can be characterized as fair.

Risks

There are two primary risks to the Cloudflare story. First, that the TAM for its cloud computing platform, Cloudflare Workers is much smaller than what it is for traditional clouds, and second, its corporate networking business has presently a lot of competition.

Workers TAM Exaggerated

Cloudflare Workers is often compared to large-scale cloud platforms like Amazon Web Services, Microsoft Azure, and Google Cloud Platform, which can create expectations that may be somewhat optimistic relative to its design. A more measured view suggests that while Workers is a compelling edge-compute platform, its total addressable market may be smaller and take longer to develop than commonly assumed. Even under generous scenarios where a large portion of existing cloud workloads migrates, the effective revenue opportunity compresses due to materially lower pricing and structural limitations, including constraints around long-running processes, language support, and data-intensive workloads.

In addition, applications built specifically for Workers are not easily portable, and migrating existing systems requires meaningful re-architecture, which can slow adoption. The platform also operates within the broader constraints of Cloudflare’s network, which may limit flexibility compared to general-purpose cloud environments, and it is not currently well suited for GPU-intensive AI workloads that are driving a growing share of cloud demand. Taken together, while Workers represents an important innovation, its long-term scale may be more specialized and modest than comparisons to hyperscale cloud providers might suggest.

Corporate Networking Competition

Cloudflare’s opportunity in corporate networking is often framed as significant, though a more tempered perspective suggests that growth may be gradual given both internal and external factors. The competitive landscape is both crowded and well-established, with many players benefiting from deep enterprise relationships and distribution advantages. At the same time, corporate networking adoption may progress incrementally, as it involves organizational change, vendor consolidation, and is typically viewed as a cost center rather than a direct driver of revenue. Cloudflare’s integrated platform approach is differentiated, but purchasing decisions for networking and security are often made across different teams, which can limit cross-sell momentum. Additionally, while Cloudflare has a strong cost advantage, the high-margin nature of the market gives competitors room to compete on pricing, potentially reducing the impact of that advantage over time.