Cloudflare: A Long-Term Owner's Perspective

At the industry level, this is not a declining sector or a traditional mature one, but a sector still expanding while intensely crowded. Cloudflare itself sorts its competitors into three groups: on-premises network-hardware vendors, various point-solution cloud vendors, and traditional public-cloud providers. External cross-checks support this: Akamai's security revenue grew 11% year over year in Q1 2026; Fastly's total Q1 revenue grew 20% with security revenue up 47%; Zscaler's revenue grew 26% in its fiscal Q2 2026; and Palo Alto Networks' revenue grew 15% in its fiscal Q2 2026. Demand across cybersecurity, Zero Trust, application delivery, and edge computing is still growing overall, but the profit pool is not effortlessly the property of any single player.

If I distill the question to "is this a business I would be comfortable holding if the stock market closed for five years," my answer is: at a good price, I would gladly hold the business; at today's price, I would not move as if acquiring the whole company. Business understandability score: 4/5. Industry attractiveness score: 4/5. It is a strong company in a good industry, but not the kind of simple business you can see through at a glance, with a profit structure already firmly set.

Moat and Management

Cloudflare's moat is neither the "brand premium" of classic consumer goods nor the physical monopoly of railroads or power grids. It is closer to an architectural moat plus platform breadth plus data-and-distribution scale plus gradually accumulating switching costs. The single global network, the unified platform, and the continuously expanding service layer that the company has long emphasized are its most important strategic assets. For customers, the more DNS, WAF, DDoS, firewall, Zero Trust access, Workers, storage, and AI-related services are layered onto the same platform, the higher the migration cost and the more the purchasing decision tilts toward "the whole platform" rather than "a single tool." Cloudflare is not the cheapest data-center operator, but it is likely one of the best at productizing, engineering, and subscription-izing network capability.

The moat should not be overstated, though. Network effects here are weaker than those of social platforms, exchanges, and payment networks; more precisely, it has a "data flywheel" and a "product flywheel" rather than strong network effects in the traditional sense. Brand advantage exists, but shows up more among developers, site operators, and security teams than among mass consumers. Scale advantage and cost advantage are present, because carrying more services on one global network raises resource-reuse rates; but this is partly offset by heavyweight rivals such as AWS, Azure, Google Cloud, Akamai, Palo Alto, and Zscaler. Regulatory or license-based barriers are weak, and patent barriers are not central. In other words, Cloudflare's moat is closer to a "hard-to-replicate integrated engineering system" than to a "legally codified exclusive right."

Broken down in more detail, my judgment is as follows:

The factual basis for these judgments rests on the company's "unified platform plus global network plus multi-product expansion plus competitive landscape plus financial structure," and the parts about culture and operating ability are inferences rather than precisely quantifiable facts.

In my view, this moat is currently closer to stable-to-slightly-widening than clearly narrowing. The reason: Cloudflare's breadth is still expanding and its enterprise penetration is still deepening; yet at the same time, the large security platforms and public clouds are also accelerating their integration, and platform-level competition has not eased. To replicate Cloudflare, a competitor must not only spend to build a network but also invest for the long term in product integration, the control plane, distribution channels, and customer mindshare. Replication is therefore costly and slow, but not impossible, merely difficult. Moat-strength score: 3/5.

On management, Cloudflare's strengths and weaknesses are both pronounced. The strength: the company remains founder-led; Matthew Prince currently holds about 41.7% of voting power, Michelle Zatlyn about 10.5%, and all officers and directors combined about 53.1%. For long-term strategic continuity, that is a strength; for outside-shareholder governance, it is a clear constraint. The debt instruments the company uses are essentially low-coupon convertibles, and as of Q1 2026 it had not drawn on its $400 million revolving credit line, indicating the capital structure is not fragile.

The weaknesses are not small either. First, stock-based compensation is heavy. Total stock-based compensation expense reached $451.5 million in 2025; for common shareholders, this is not an "inconsequential non-cash expense" but potential dilution. Second, the dual-class structure means outside shareholders lack a real check on key governance matters. Third, the 2024 proxy disclosed that the incremental costs related to Matthew Prince's and Michelle Zatlyn's personal use of the company-leased aircraft were roughly $1.551 million and $0.954 million, respectively; this will not change enterprise value, but it is not the most restrained shareholder-culture signal in my book. On balance, I score management and capital allocation: 3/5.

Financial Quality and Owner Earnings

Start with the most important trend. From 2020 to 2025, Cloudflare's revenue grew from $431.1 million to $2.1679 billion, a five-year compound annual growth rate of about 38%; this is genuine high growth. Over the same period, the GAAP operating margin improved from -25% in 2020 to -9% in 2024, but was still -10% in 2025; in other words, operating leverage is improving, but the company has not yet crossed the threshold into "high-quality profitability." Gross margin has long stayed in the 75%-78% range, indicating the business model has a solid gross-margin base, but the drop to 71% in Q1 2026 reminds us that infrastructure expansion and new-business investment are not free.

The table below is compiled from Cloudflare's 2020-2025 10-Ks, the Q1 2026 quarterly report, and the company's official full-year earnings press releases; unless otherwise noted, figures are on a GAAP basis. Year-over-year changes, CAGR, and some ratios are calculated by the author from the financial statements; the precise 2025 free cash flow was not independently re-verified in the materials extracted this time, so the cell is left blank rather than fabricated.

If your central concern is "are these profits real cash profits," the conclusion is: partly real cash, partly still accounting- and SBC-driven surface improvement. On one hand, the improvement in operating cash flow is real, reaching $603.1 million in 2025; on the other hand, 2025 operating cash flow included a $451.5 million add-back of stock-based compensation, and working capital that year also drained $97.3 million. Looking further at the 2025 income statement, interest income was $131.2 million while interest expense was only $8.77 million; that is, part of the narrowing net loss came from interest income on the large cash balance rather than from a highly mature core business. The same holds in Q1 2026: the operating loss was $61.99 million, but interest income of $40.17 million pulled the pre-tax loss down to $21.40 million. This shows Cloudflare has not yet reached the stage where "operations alone can easily roll out profit."

On the balance sheet, Cloudflare is not fragile. As of March 31, 2026, the company had $932.2 million in cash and $3.2317 billion in available-for-sale securities, for total liquidity of about $4.164 billion; meanwhile, net convertible debt on the books was about $3.268 billion, corresponding to net cash of about $896 million. The current ratio is close to 2x, and the revolving credit line is undrawn. The real near-term pressure is not "will it survive" but how the convertibles maturing in 2026 are handled, plus potential dilution while the stock price is high. The company has disclosed that, based on the March 31, 2026 closing price of $206.34, the if-converted value of the 2026 convertibles exceeded principal by about $101.4 million, so that portion was classified as a current liability. For shareholders, ample cash is good, but potential dilution cannot be ignored.

Changes in the share count are not to be dismissed either. Total weighted-average shares rose from 299.8 million in 2020 to 348.4 million in 2025, a five-year increase of about 16%, roughly 3% annualized. As long as stock-based compensation stays this heavy, even if the company generates cash, the per-share intrinsic-value growth that reaches common shareholders will be diluted. For Buffett-style analysis, this must be treated as a real cost.

Owner Earnings Estimate

Buffett-style "owner earnings" is not a simple copy of the company's reported free cash flow; it asks: after subtracting the reinvestment required to maintain competitiveness, and counting the items that carry a real cost to shareholders, how much distributable cash is left?

Using 2025 as the anchor, I offer a conservative version:

• Net loss: about -$102.3 million.

• Operating cash flow: about $603.1 million, which is the result after adding back non-cash items such as depreciation, amortization, and stock-based compensation, and already reflects working-capital changes.

• Non-cash expenses requiring special treatment: The most important is stock-based compensation of $451.5 million. In accounting terms it is added back to operating cash flow, but from a common shareholder's standpoint it represents future equity dilution and cannot be treated as "free cash given away."

• Maintenance capital expenditure: unknown, and the financial statements cannot split it out precisely. I adopt an assumption: using the property-and-equipment-related depreciation and amortization of $167.5 million disclosed in the 2025 statements as a reference, maintenance capex should be at least in the $130 million-$170 million range. This assumption is conservative but not exaggerated, because the network itself needs continual server and software upgrades.

On this basis, conservative owner earnings ≈ operating cash flow $603 million - stock-based compensation $452 million - maintenance capex $130 million-$170 million ≈ -$18 million to +$22 million, with a central value near zero. That is, if you treat stock-based compensation as a genuine shareholder cost, Cloudflare today is closer to "roughly breakeven" than to "already producing high-quality, large profits." This is one of the most important conclusions in this report. It does not negate the quality of the business, but it heavily shapes the valuation conclusion.

My judgment on owner earnings is therefore:

• Conservative version: near zero, even slightly negative.

• Neutral version: if only part of SBC is treated as an economic cost, it can be modestly positive, but the evidence is not solid enough.

• Company/market optimistic version: closer to free cash flow, but that would overstate what common shareholders can truly distribute.

At a market capitalization of about $69.66 billion, Cloudflare's valuation multiple on "conservative owner earnings" is not well described by traditional value methods; even using the audited 2024 free cash flow of $166.9 million, the P/FCF exceeds 400x. This is not a "cheap stock" but a highly long-duration growth asset.

Valuation and Margin of Safety

Discounted Owner Earnings

Because Cloudflare's conservative owner earnings are currently near zero, a DCF is extremely sensitive to the long-term growth rate, the terminal margin, the path of SBC decline, and the dilution rate. Here I avoid the "false precision" of decimal points and give a range instead.

My core assumptions are as follows:

Under these premises, my rough per-share intrinsic-value estimate is:

• Conservative intrinsic-value range: $80-$110

• Fair intrinsic-value range: $120-$160

• Optimistic intrinsic-value range: $180-$230

This means the current price of $197.56 is only roughly defensible under the optimistic scenario; in the neutral and conservative scenarios, it is by no means cheap. Put differently, what the market gives Cloudflare today is not a "price after reasonable skepticism" but a "delivery price under high confidence."

Relative Valuation

The comparables table below is based on recent official results and mid-May 2026 market quotes. Because the companies have different fiscal years and capital structures, and because this report did not perform original-filing re-verification of every peer's latest EV/EBITDA, P/B, and ROIC, I would rather mark those items not uniformly verified / not applicable than fabricate a table that merely looks tidy. What is more meaningful is the latest growth, GAAP/cash profit quality, and a simple sales multiple.

From this angle, Cloudflare's problem is not "a poor company" but "a valuation that prepays too much good news." Even measured against richly valued software-security companies, NET clearly sits on the more aggressive side on a sales multiple.

Asset Value and Liquidation Value

For Cloudflare, the asset approach offers almost no support for a high valuation. As of March 31, 2026, cash and available-for-sale securities totaled about $4.164 billion, net convertible debt was about $3.268 billion, equivalent to net cash of about $896 million. Against roughly 353.4 million shares, net cash works out to about $2.5 per share. The current stock price is close to $198. This means almost everything the market is buying is future earnings power and platform position, not existing realizable assets. For value investors, such a name naturally demands more valuation discipline.

Margin-of-Safety Judgment

My conclusion is therefore clear:

• Is the current price cheap enough? No.

• The most fragile assumption in the valuation: long-term growth around 20%, a continuous decline in SBC as a share of revenue, and an eventual owner-earnings margin reaching the mid-teens.

• If growth falls short: even if Cloudflare remains a good company, returns could be mediocre, because the current price is too sensitive to growth.

• If margins decline or stay in the low-to-mid range long term: the investment shifts from "high-return growth" to "high valuation, low cash return."

• If the valuation multiple contracts: multi-year or even permanent capital loss is entirely possible.

• Is this a case of "good company, bad price"? Yes.

• Is it worth waiting for a better price? Yes.

The operating ranges I offer are:

• Ideal buy range: $95-$125

• Acceptable holding range: $125-$170

• Clearly overvalued range: above $190

These ranges are not technical-analysis levels; they are derived backward from "conservative value, fair value, and the need to keep a margin of safety."

Risks, the Bear Case, and Opportunity Comparison

Start with the most important risk. Competitive risk ranks first. Cloudflare operates not in a monopoly market but in a platform war crowded with Akamai, Zscaler, Palo Alto, Fastly, and the large public clouds and traditional cybersecurity vendors. Cloudflare itself acknowledges in its annual report that competitors include on-premises hardware vendors as well as various point-solution providers and traditional public clouds. For such an industry, the scariest outcome is not demand vanishing but the profit pool being continually siphoned off.

The second risk is technology and business-model pressure. In Q1 2026 gross margin fell to 71%, and the company explicitly pointed to third-party technology services, network and bandwidth costs, and rising server depreciation. This shows that the company's expansion into AI, the developer platform, and more complex enterprise services will not necessarily lift margins as naturally as a traditional SaaS would. If infrastructure and inference-related costs keep rising over the next few years while customers' willingness to pay does not keep pace, the market's expectation that "the platform converges into a high-margin engine" will be forced to reset lower.

The third risk is shareholder dilution and convertibles. Stock-based compensation expense of $451.5 million in 2025 is already very high; meanwhile, the convertibles maturing in 2026 were classified as a current liability in the Q1 2026 report and, at the period's stock price, are already in a position that could dilute common shares. For a company that has not yet built thick GAAP profit, these two factors will keep eroding per-share value.

The fourth risk is regulation and content liability. The company's annual report explicitly warns that laws on content, copyright, platform liability, data privacy, and cybersecurity-incident disclosure could raise litigation and costs; it also disclosed that, in October 2025, a Japanese court found it liable for damages in a case involving infringing content on a website that used its services, and that the company has appealed. Such risks usually will not destroy a company, but they raise operating complexity and compliance costs.

The fifth risk is the high valuation itself. This is not a short-term volatility risk but the core source of "permanent capital loss" risk. If you buy at a very high price today, and five to seven years out the company proves capable only of "good but not exceptional" margins, then even if you did not pick the wrong company, you may have picked the wrong return.

The strongest counterargument, I would put this way:

Cloudflare may be a very good company, but the current price requires it to become a nearly great one. What bears truly watch is not "the product is poor" but: first, real owner earnings net of SBC are still near zero; second, gross margin has not risen linearly with scale and instead came under clear pressure in Q1 2026; third, the market is awarding it the valuation of a top-tier platform-style security/infrastructure asset, while its profit maturity is nowhere near that yet. If any one of these links fails to deliver, the valuation could de-rate first.

What facts would make me admit I was wrong, or force me to rewrite the thesis? At least the following: if large-customer growth slows markedly over the coming quarters and net retention falls back below 110% and stays there; if gross margin stays stuck in the low 70s long term and cannot recover; if stock-based compensation persists at a high single-digit to low double-digit share of revenue or above; if handling the 2026 convertibles brings worse-than-expected dilution; if management starts substituting aggressive M&A for organic improvement. All of these would weaken the core premise that "the larger the platform, the better the unit economics." Conversely, if over the next two to three years the company can hold growth around 20% while bringing down both SBC as a share of revenue and the GAAP operating loss, I would meaningfully raise my valuation.

Compared with other opportunities, if you are a conservative value investor, I do not think NET is clearly superior to the index, much less clearly superior to more mature peers that have already delivered cash flow. Akamai grows far more slowly but its real profitability is clearly more mature; Palo Alto's platform and profitability are further along the delivery curve than Cloudflare's; Zscaler strikes a more mature balance between growth and cash flow. By contrast, Cloudflare's appeal comes mainly from "it could be very large in the future," not "it is already cheap today." If your portfolio could hold only 5 assets, I do not think Cloudflare today earns a place on risk-adjusted return.

Checklist and Final Verdict

Investment Checklist

The conclusions above combine official company disclosures, peers' recent results, and the current market valuation.

Final Investment Verdict

【Final Rating】 Watch

【One-Sentence Investment Thesis】 Cloudflare is a long-term platform business worth respecting, but buying at today's price means you are paying mostly for "future perfect delivery" rather than "a present margin of safety."

【Core Bull Case】

• A unified global network carrying a multi-product platform, with strong business scalability and continually expanding service breadth.

• Revenue led by subscriptions, a diversified customer base, and steadily rising enterprise penetration give solid revenue quality.

• Revenue growth is still high: up 30% for full-year 2025 and 34% in Q1 2026.

• A sound balance sheet, positive net cash, low financing costs, and low near-term survival risk.

• Strong founder control and high strategic continuity.

【Core Bear Case】

• Conservative owner earnings net of stock-based compensation are near zero, still a distance from a Buffett-style "high-quality cash cow."

• Q1 2026 gross margin fell to 71%, showing expansion is not naturally high-margin.

• The valuation is extremely high, and the current price holds up only in the optimistic scenario.

• The dual-class structure and continual dilution work against common shareholders.

• The competitive environment is very strong, and the industry profit pool is not clearly locked up in Cloudflare's hands.

【Key Assumptions】

• Enterprise customers keep expanding, and net retention stays in a healthy range.

• Gross margin returns to around the mid-to-high 70s over the next two to three years.

• SBC as a share of revenue declines steadily rather than staying elevated.

• Handling the convertibles does not cause worse-than-expected dilution.

• The company can convert platform breadth into an owner-earnings margin improvement over a ten-plus-year horizon.

【Fair Buy Price】 $95-$125. This range is derived backward from "fair value of $120-$160, then keeping a 20%-25% margin of safety"; if you are more aggressive, you could raise the cap on an initial tracking position to about $130, but that no longer counts as a typical value margin of safety.

【Target Holding Period】 10 years or more. For a business like Cloudflare, making real money depends not on next year's EPS but on whether the next decade can string together platform expansion, share gains, and profit delivery. Over a shorter horizon, the value framework loses much of its meaning.

【Expected Annualized Return】 This is opinion plus assumption, not fact:

• Conservative scenario: 0%-3%

• Neutral scenario: 5%-8%

• Optimistic scenario: 10%-13%

I do not offer more because the starting valuation is too high, and a sizable share of future return will be eaten by valuation contraction.

【Maximum Loss Risk】 If over the next three to five years Cloudflare's growth falls to the low-to-mid double digits while both gross margin and SBC improvement disappoint, the market may re-rate it from a "top-tier long-term platform" to a "good but ordinary infrastructure-software company." In that scenario, from the current price, a 50%-70% downside is not unimaginable. This is not because the company would go bankrupt, but because the price you pay today is too high.

【Tracking Metrics】 Going forward, I will keep watching these:

• The number of customers with annualized revenue above $100,000.

• The dollar-based net retention rate.

• Whether GAAP gross margin recovers.

• SBC as a share of revenue.

• The relationship between operating cash flow and receivables and deferred contract-acquisition costs.

• The settlement method and dilution impact of the 2026 convertibles.

• Whether channel-partner revenue share brings higher efficiency.

• Whether the GAAP operating loss keeps narrowing, rather than the net loss being dressed up by interest income alone.

【Signals That Trigger Reassessment】

• Net retention falls persistently below 110%.

• Large-customer growth stalls markedly.

• Gross margin stays below 73%-74% long term.

• SBC stays elevated long term, and per-share dilution keeps approaching or exceeding the pace of the past five years.

• Major M&A clearly worsens capital allocation.

• Management starts substituting "storytelling" for "delivering cash-flow improvement."

【Final Recommendation】 If you are putting new money to work today, I would not advise ignoring the price just because "the company is excellent." Cloudflare deserves a spot on a high-priority watch list, deserves long-term tracking, and deserves a fresh look after a meaningful pullback or further fundamental delivery; but from the standpoint of a long-term business owner rather than a short-term trader, today is more about patient waiting than rushing to act. If you already hold it at a low cost, I lean toward continuing to hold and tracking delivery quality closely rather than mechanically selling because the short-term valuation is high; but if you are not on board yet, do not mistake a "good company" for a "good price."

Open Questions and Limitations

This report has three limitations to state plainly: first, the precise 2025 free cash flow was not separately re-verified to final in the official text extracted this time, so for owner earnings I used the more conservative, plainer "CFO - SBC - maintenance capex" framework; second, peers' EV/EBITDA, P/B, and ROIC were not subjected to a full, uniform original-filing re-verification, so I did not force-fill the table for the sake of completeness; third, the personal settings in your original template for "goals, risk tolerance, and buy/hold/sell" were not filled in, so the rating above leans toward a long-term value investor in the general sense rather than fully personalized advice.

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.