Why I Bet My Dollars on Cloudflare Stock
I don't do this often. Write about my own money, I mean. There's something uncomfortably revealing about admitting where you put your actual dollars, not just where you think other people should put theirs. But here we are.
I bought Cloudflare. A few tranches across a few months. Not a small position for someone of my means. And I want to put into words why I did it, if only so I can come back here when it's down 30% — as it will be at some point, because everything is down 30% at some point — and remember whether I was thinking clearly or just caught up in a story.
The moment I started paying attention
It wasn't the earnings calls. It wasn't a newsletter tip or a podcast segment. It was a Thursday afternoon when I was setting up a new domain and I noticed, almost as a background fact, that Cloudflare's free tier was doing things that enterprise networking products charged serious money for a few years ago. DNS, DDoS protection, CDN, basic Zero Trust. All free. For everyone.
That's not a marketing stunt. That's infrastructure strategy. They're buying the bottom of the market — the developers, the indie builders, the small operators — and turning them into a pipeline for enterprise deals five years from now. I've seen that playbook work before. I wanted in.
What I actually think the business is
Cloudflare is not, in my reading of it, primarily a cybersecurity company. The way it gets categorized on financial terminals misses the point. It's closer to what AWS was before anyone admitted AWS was the real business — a platform for running compute and traffic at the edge of the internet, at a scale that is genuinely difficult to replicate. The network they've built across 330+ cities is not a product feature. It's a moat. Physical infrastructure, negotiated peering agreements, years of routing optimization. You can't buy that moat on a Tuesday.
The SASE and Zero Trust story is the near-term revenue narrative, and it's a real one. Enterprises are dismantling their legacy perimeter security stacks. Cloudflare is one of maybe three or four credible platforms to absorb that spend. That's a large, structural tailwind with a multi-year runway.
But the longer arc — the one I'm actually betting on — is Workers and the edge compute platform. The idea that you can run code at the network layer, 50 milliseconds from any user on earth, is not just an interesting developer toy. It's a fundamentally different architecture for applications. If that vision captures even a fraction of the workloads currently running on centralized cloud, the revenue math changes dramatically.
The numbers I can live with
I'm not going to pretend I bought this at a cheap valuation. I didn't. Cloudflare has never been cheap on conventional metrics. When I initiated my position the forward EV/Revenue multiple was in the mid-twenties. That's a number that requires you to believe in the growth trajectory, not just the current business.
What I anchored on instead: revenue growth consistently above 25% annually, net revenue retention above 125% for years running, gross margins in the high 70s with a credible path into the low 80s as the mix shifts toward higher-margin software. The operating losses are narrowing in a way that suggests the unit economics are working, not just deferred. They're not burning cash to prop up a broken model. They're investing at a pace that the business can eventually absorb.
I built a rough model. Base case: revenue crosses $3 billion in the next eighteen months, free cash flow turns meaningfully positive in the year after that. If the market continues to assign a growth premium to the name, the stock follows the fundamentals. Bull case: Workers takes off, the AI inference edge story gets priced in — that's a different conversation entirely. Bear case: macro tightens, enterprise deals push right, the multiple compresses before the FCF inflection arrives. I know the bear case. I've sat with it. I still bought.
The AI angle I keep turning over
Here's the thing I don't see discussed enough. Every major AI application has a latency problem at the edges of deployment. You can run inference in a hyperscaler datacenter, but getting that response to a user in Lagos or Manila or Kraków in under 100 milliseconds is an infrastructure problem, not a model problem. Cloudflare's network is, arguably, the best-positioned commercial infrastructure to solve that last-mile AI latency problem.
They've made the right early moves — AI Gateway, Workers AI, the Vectorize database product. None of these are dominant businesses yet. But they're establishing the on-ramp. When enterprises start demanding low-latency AI features embedded in their applications — and they will — Cloudflare will be in the room.
I don't know if they win that bet. But I know I want exposure to it.
What keeps me up at night
Matthew Prince. I mean that with respect, not as an insult. Founder-led companies with a strong technical vision are great until they're not. The moment the founder's conviction outpaces the business's capacity to execute, things get messy. Cloudflare has expanded its product surface aggressively. At some point, breadth without depth becomes a liability. I'm watching whether the platform bets are converting into real revenue or just generating demo traffic and developer enthusiasm.
Competition is the other thing. Zscaler and Palo Alto are not standing still. Microsoft has security ambitions that extend across its entire enterprise base. Akamai is still in the room. The Zero Trust market will not be a Cloudflare monopoly. The question is whether they take enough of it to justify the valuation — and I think the answer is yes, but with a confidence interval I hold loosely.
Why I'm writing this
Partly accountability. If I write down the thesis, I'm more likely to evaluate new information against it rather than defaulting to confirmation bias. Partly because personal finance writing that's actually personal is rare. Most of it is either cheerleading or disaster-porn, and neither teaches you anything about how actual decisions get made.
I could be wrong about this. The stock could go nowhere for three years while the market rotates to something I'm not holding. I might need the money before the thesis plays out. These are known risks, not catastrophic surprises.
But when I look at the internet — how traffic actually moves, how applications are actually built, where the architecture is going — Cloudflare is sitting at a junction I want to own. That's what my dollars are saying.
We'll see if I was right.